Okay, I've read the GeoFX "Technology of the Tubescreamer" article as well as the "AMZ tone control mods" article and I'm a little confused about the interaction of the various components.
Here is a link to the article for reference: http://www.geofex.com/Article_Folders/TStech/tsxtech.htm
The calculation for the rolloff of the 1K resistor and .22uf cap to ground immediately after is gain stage is stated as being 723Hz. That makes sense. Now, the boost stage is where it gets hairy. The GeoFX article claims that the R-C rolloff calculation for the cap in the feedback network is 3.2KHz when the filter is all the way open. He gets this number by calculating the combination of 220ohm resistor to ground and the .22 cap attached to the wiper, and I've also seen this same calculation repeated before here on these forums with the same numbers. That's what I don't get. Don't you have to include the 1K resistor that is also in the path of the feedback network to get the rolloff value? So wouldn't the rolloff actually be 593Hz? Could someone please explain to me why the 1K resistor is not included in the calculation for the R-C filter? I thought I was starting to understand this stuff, but now I am really confused.
It seems to me that it would make sense to if the rolloff was actaully (1K+220ohms)/.22uf which equals 593Hz, not 3.2KHz like the article explains, because that would much better explain the midrange hump in the Tubescreamer that everyone is always complaining about. If the first cap is rolling off everything above 723Hz, and then the feedback network is boosting everything above 593Hz, then that means you would get a big midrange hump between 593 to 723Hz. That makes a lot more sense to me than a boost above 3.2KHz, which would leave a midrange dip between 723 and 3.2k when the tone control is maxed, which I don't hear at all in the tubescreamer.
Going on this assumption, then to get an even frequency response you just get rid of the 220ohm resistor to ground on the wiper. Now you have a 1K/.22uf=720Hz rolloff, followed by a 1k/.22=720hz boost, and they should theoretically cancel each other out!
So, I lost the 220 resistor and put the whole tone section on a bypass switch so I could flip between them. Guess what... with the tone control maxed and the 220 ohm resistor gone it sounded almost identical to the sound of the whole tone section being completely bypassed. Hmmmm... Not satisfied yet, I used a synth to run some fast frequency sweeps through the pedal and hooked it up to a spectrum analyzer. Yup, with the 220 resistor gone the response is nearly flat as a rail.
So, the for final answer to the question "How do you get rid of the mid-range hump in the Tubescreamer": jumper the 220ohm resistor in the feedback network, and of course, also do the typical fat mod to the gain stage. It seems that the whole tone stack on the tubescreamer is designed specifically to be midrange booster. The initial high frequency rolloff when you start to turn the tone knob down is just a consequence of boosting even lower midrange which sounds like a high cut when the gain of the whole stage is 1. That's why there is never a real treble boost or scooped sound with the tubescreamer. It just goes from more to less midrange honk. I guess that's part of it's charm.
So is the GeoFX article wrong or am I just crazy? Ugghhh this is frustrating! Why do I have such a hard time understanding all of this stuff? Please help me understand this better.
...
Ohh, and one more question: How do you calculate the rolloff frequency for the small 51pf cap in the feedback loop of the clipping stage. I just don't understand how that works at all. Don't you need a resistor to ground somewhere after the cap in order to get a high pass filter?
QuoteDon't you have to include the 1K resistor that is also in the path of the feedback network to get the rolloff value? So wouldn't the rolloff actually be 593Hz? Could someone please explain to me why the 1K resistor is not included in the calculation for the R-C filter?
The 1K resistor is between the output and the negative input of the op-amp, and that's why I'm inclined to think it only sets the gain of the stage and does not effect the frequency response. The tone control is in between the negative and positive inputs. I'm sure you noticed it, but I think the current flows towards the output of the opamp and that's why the 1K resistor would be "after" the tone control in the circuit, and thus wouldn't affect the roll-off frequency of the filter.
QuoteOhh, and one more question: How do you calculate the rolloff frequency for the small 51pf cap in the feedback loop of the clipping stage. I just don't understand how that works at all. Don't you need a resistor to ground somewhere after the cap in order to get a high pass filter?
The 51p resistor lets more highs through. That means the negative feedback for highs is bigger, and thus highs are attenuated. It is explained in the article. You can in your mind think of the Drive pot, the 51k resistor, the diodes and the 51pf cap as a single resistor with a fixed impedance at any given frequency, and the 0.047uF cap and the 4K7 resistor as another. The impedance of the first resistor is the parallel combination of the cap, the diodes and (drive pot setting + 51k in series). The impedance of the second imaginary resistor would be the series combination of the 0.047uF cap and the 4K7 resistor. These two imaginary resistors form a voltage divider between the output and ground. Obviously, the impedance changes with frequency.
I'm hoping someone else can phrase this in a more understandable way ;)
Quote from: MohiZ on April 08, 2009, 05:37:44 PM
QuoteDon't you have to include the 1K resistor that is also in the path of the feedback network to get the rolloff value? So wouldn't the rolloff actually be 593Hz? Could someone please explain to me why the 1K resistor is not included in the calculation for the R-C filter?
The 1K resistor is between the output and the negative input of the op-amp, and that's why I'm inclined to think it only sets the gain of the stage and does not effect the frequency response. The tone control is in between the negative and positive inputs. I'm sure you noticed it, but I think the current flows towards the output of the opamp and that's why the 1K resistor would be "after" the tone control in the circuit, and thus wouldn't affect the roll-off frequency of the filter.
I was under the impression that in a feedback network the current flows from the output of the opamp into the negative input (isn't that why they call it feed-BACK), which means it would have to pass through the 1K resistor before the cap to ground which I would think would have to be included in the calculation. The current definitely isn't coming from the lug connected to the positive input of the opamp because I can disconnect that lug entirely and the tone control functions relatively the same for the first 1/4 turn of the knob. Current wouldn't be flowing OUT of the negative input would it? That's not how I understand opamps work. Also none of this explains why the tone control magically becomes "ruler flat" when I remove the 220ohm resistor, which only makes sense if both networks are rolling off at the same point, 723Hz, one being a cut and the other a boost.
With AC, does current really "flow" in a particular direction anyway? Doesn't it just oscillate back and forth, the output of the opamp being what "forces" that oscillation at the output based on what it "sees" at the positive and negative inputs? The only sense of "direction of flow" would be if you placed component "A" in front of that output, everything after that particular component will see the result of that component's effect on the current that is being oscillated by the output of the opamp. In this case, the 1K resistor would be between the "force" of the output and the cap to ground, so it would have to be included in the calculation, right?
This is my first pedal build and I feel very strange having to argue with people that are obviously much more knowledgeable than I am, but something is telling me that this is not correct.
Quote from: MohiZ on April 08, 2009, 05:37:44 PM
QuoteOhh, and one more question: How do you calculate the rolloff frequency for the small 51pf cap in the feedback loop of the clipping stage. I just don't understand how that works at all. Don't you need a resistor to ground somewhere after the cap in order to get a high pass filter?
The 51p resistor lets more highs through. That means the negative feedback for highs is bigger, and thus highs are attenuated. It is explained in the article. You can in your mind think of the Drive pot, the 51k resistor, the diodes and the 51pf cap as a single resistor with a fixed impedance at any given frequency, and the 0.047uF cap and the 4K7 resistor as another. The impedance of the first resistor is the parallel combination of the cap, the diodes and (drive pot setting + 51k in series). The impedance of the second imaginary resistor would be the series combination of the 0.047uF cap and the 4K7 resistor. These two imaginary resistors form a voltage divider between the output and ground. Obviously, the impedance changes with frequency.
I'm hoping someone else can phrase this in a more understandable way ;)
That actually helped a lot. It was very understandable. Thank you Mohiz.
I guess my questionis then: How do you calculate the rolloff frequency off a RC network with a resistor and capacitor in parallel? I tried doing a google search but i didn't find much that was helpful. I see these parallel resistor capacitor filter networks in feedback loops all the time, but I've never seen any information on how they are calculated. I know this particular example becomes exceedingly complex, but let's just say we look at it in one particular state. With the diodes at full clip you can eliminate them, and let's say you have the pot at full gain, so that would be 551K in parallel with 51p in combination with 4.7K to ground. Now if I could figure out how to calculate that particular state I could start shifting numbers around to estimate different states, which would help me greatly to wrap my head around it. Anyone care to point me in the right direction? I appreciate the help. :icon_biggrin:
You calculate de Frec cutoff in all the filters the same ways. The only thing that changes is the type of filter being them: Parallel, series, PIs, etc. Then, in a subcategory you have: Hig pass, low pass, band pass and stop band ;)
Quote from: SISKO on April 08, 2009, 08:40:25 PM
You calculate de Frec cutoff in all the filters the same ways. The only thing that changes is the type of filter being them: Parallel, series, PIs, etc. Then, in a subcategory you have: Hig pass, low pass, band pass and stop band ;)
Okay, but how do I apply that formula to find the cutoff frequency of a filter that consists of a resistor and capacitor in parallel? I already did a google search and didn't come up with anything useful. If I already knew how to apply the R-C formula to this type of filter I wouldn't be asking. At least a link to some specific information would be useful.
The formula I am familiar with is Fc = 1/ (2 . pi . R . C)
Do I just plug the parallel resistor into R with the parallel cap in C?
SO essentially this: in-----l C l------------out
|
z
R
z
|
V
is calculated the same way as this?
in-------------------l C l-------------
| |
| |
--------W R W--------------------out
Quote from: Projectile on April 08, 2009, 07:48:55 PM
I was under the impression that in a feedback network the current flows from the output of the opamp into the negative input (isn't that why they call it feed-BACK), which means it would have to pass through the 1K resistor before the cap to ground which I would think would have to be included in the calculation.
i learned early on to stop thinking about which way the electrons run... because in all truth, the little buggers run the other way!
just thinking outloud here, this isn't meant to be a full answer or a lesson,
the opamp works by changing the output so the inputs match, so the 1k resistor is just showing the minus input a signal that's "so much" less than the output... its the upper leg of a voltage divider, except that the bottom leg is frequency dependent (depends on the tone pot setting essentially.) So i think the best way to thinking about that circuit is that it provides frequency dependent gain (or attenuation!).
Set to one side, it filters out the highs going into the minus/feedback input, this translates to a boost of those frequencies at the output, and because of the filter cutoff, it effectively undoes the low pass filter that comes between the opamps. At the opposite setting, it lets the minus see the full frequency response, so the output already has the highs cut from the low pass filter between the opamps.
Jack Orman (AMZ) does a great discussion of this tone control, as well as RG.
Quote from: aziltz on April 08, 2009, 10:29:46 PM
Quote from: Projectile on April 08, 2009, 07:48:55 PM
I was under the impression that in a feedback network the current flows from the output of the opamp into the negative input (isn't that why they call it feed-BACK), which means it would have to pass through the 1K resistor before the cap to ground which I would think would have to be included in the calculation.
i learned early on to stop thinking about which way the electrons run... because in all truth, the little buggers run the other way!
just thinking outloud here, this isn't meant to be a full answer or a lesson,
the opamp works by changing the output so the inputs match, so the 1k resistor is just showing the minus input a signal that's "so much" less than the output... its the upper leg of a voltage divider, except that the bottom leg is frequency dependent (depends on the tone pot setting essentially.) So i think the best way to thinking about that circuit is that it provides frequency dependent gain (or attenuation!).
Set to one side, it filters out the highs going into the minus/feedback input, this translates to a boost of those frequencies at the output, and because of the filter cutoff, it effectively undoes the low pass filter that comes between the opamps. At the opposite setting, it lets the minus see the full frequency response, so the output already has the highs cut from the low pass filter between the opamps.
Jack Orman (AMZ) does a great discussion of this tone control, as well as RG.
Yes, this is all true, but it completely misses the point. What is in question here is not a general idea of how the filter works, but where the rolloff points specifically fall on the frequency spectrum when the filter is all the way open. GeoFX claims that the feedback network boosts frequencies above 3.2Khz, bringing back up most of the treble that the first 720Hz filter rolled off. I'm saying that this appears to be wrong and that the feedback network actually boosts above 593Hz, which falls BELOW the frequency of the first filter. This means that you get a broad midrange hump around 650 that rolls off gradually in either direction and then steeply in the highs. This is the characteristic midrangey sound of the tubescreamer. it never goes away because you never really boost the highs, just remove the hump by first boosting even lower frequencies and then rolling off more treble as you roll the knob back. The most important part of the midrange honk of the tubescreamer tone stack IS that 220 ohm resistor. If you jumper that resistor the tonestack appears to go flat. you get -720Hz rolloff and then +720Hz boost and the midrangy honk is gone.
The AMZ discussion is great if you are looking for alternatives to the original design but he also seems to entirely miss what is going on with that resistor. The article doesn't state anything wrong like the GeoFX article, but it kind of ignores that resistor's function, which is that it turns the whole tonestack into a midrange honk machine. The tonestack isn't really useful for anything else, so if you don't like midrange, then you are probably better off using one of his alternatives.
I've read a lot of articles an posts discussing the inner workings of the tubescreamer and the issue of it's midrangeyness comes up a lot, but I kind of shocked that no one has ever pointed out that you can make the tubescreamer flatten out almost entirely by just jumpering that resistor. I've done frequency sweeps through the pedal into a spectrum analyser and this appears to be exactly what is going on.
At least it appears that way...
But Really I'm a total newb at all this and it bothers me that my understanding of this circuit would contradict the published information that so many people are familiar with. What I really need is just someone to answer either, "Yes, that is correct the GeoFX article is wrong," or "No that's not how it works, here is where your confusion is" and give me a clear and specific correction.
I am very interested in audio electronics, but I don't have anyone else to help me with this stuff, so I have to rely on forums like this one when there is a contradiction in my understanding. It is very important that I clear it up, so I can move on. I guess I really don't understand why that is so difficult. This isn't rocket science. They're just simple little audio circuits, right? But for some reason whenever I ask a specific, precise question on this forum just to help clarify my rudimetary understanding, all I ever get are really vague and dismissive answers and it's is very frustrating. I just don't get it. I've never had this experience on any other internet forum. I come on here and I see all this really knowledgeable people designing and building really amazing things, so I know my questions must be simple and elementary, but it's so difficult to get any help or direct answers here. I've honestly had better results posting my electronics related questions on other, non-electronics related music forums than I have here, which I find very odd. The whole thing leaves me rather frustrated and makes me feel like I'm the butt of some kind of joke here. WTF? Am I just a complete idiot or something?
this could easily be put to rest if someone would/could simulate that thing in spice. I unfortunately don't have the skills yet.
sorry i missed your point the first time. In the back of my mind i was thinking that the mid hump is set by the fixed low pass at 720Hz and the Cap to ground off the feedback of the CLIPPING amp, only because that's where the BYOC Overdrive II has its Mid Hump/Flat/Full Mod.
although, if the fixed roll off is >720Hz, and the Tone Control Boosts from 593Hz and up, that would "raise" the Mid Hump with the Tone Control wouldn't it?
I'm just as intrigued/confused as you are. I grade for a undergraduate electronics class currently, and in 2 weeks we will be learning Spice in-depth. I plan on adding this to my basic dissection of the Tube Screamer and other Standard Circuits.
Whether this is right or wrong or just one big grey area, thanks for bringing it up and pointing it out.
Quote from: aziltz on April 08, 2009, 11:53:10 PM
this could easily be put to rest if someone would/could simulate that thing in spice. I unfortunately don't have the skills yet.
sorry i missed your point the first time. In the back of my mind i was thinking that the mid hump is set by the fixed low pass at 720Hz and the Cap to ground off the feedback of the CLIPPING amp, only because that's where the BYOC Overdrive II has its Mid Hump/Flat/Full Mod.
although, if the fixed roll off is >720Hz, and the Tone Control Boosts from 593Hz and up, that would "raise" the Mid Hump with the Tone Control wouldn't it?
I'm just as intrigued/confused as you are. I grade for a undergraduate electronics class currently, and in 2 weeks we will be learning Spice in-depth. I plan on adding this to my basic dissection of the Tube Screamer and other Standard Circuits.
Whether this is right or wrong or just one big grey area, thanks for bringing it up and pointing it out.
Sorry aziltz, that little rant wasn't directed at you. I appreciate your help. This forum just confuses me. I think I just need to go back to school. I'm probably too busy looking for things on the internet that would probably be better learned with the help of a teacher in a classroom.
Just to boil this down to two questions for anyone following this thread:
-Is the 1K resistor in the feedback network of the tubescreamer tone control calculated in combination with the 220 ohm resistor to ground, which sets the rolloff frequency of the .22 cap on the wiper below 720kHz when the tone control is turned all the way up? (This would appear to make the well circulated GeoFX article incorrect)
-If not, then why isn't the 1K resistor included in RC filter calculation?
...Note that according to my rudimentary testing, simply jumpering the 220ohm resistor to ground with the tone knob turned all the way up appears to flatten out the signal entirely, making it sound like the tone stack has been bypassed. This would seem to confirm that the 1K resistor is in fact part of the RC filter network and the 220 ohm resistor just offsets the rolloff frequency down compared to the previous 720hz rolloff, where without the 220 ohm resistor the opposite 1K/.22ohm cut and 1K/.22 boost would just cancel each other out.
Thanks.
QuoteOkay, but how do I apply that formula to find the cutoff frequency of a filter that consists of a resistor and capacitor in parallel? I already did a google search and didn't come up with anything useful. If I already knew how to apply the R-C formula to this type of filter I wouldn't be asking. At least a link to some specific information would be useful.
The formula I am familiar with is Fc = 1/ (2 . pi . R . C)
It becomes very complicated to calculate it precisely, and that's why it's most often approximated, like so many other things in electronics ;) The low roll-off freq is approximately Fc = 1/ (2 . pi . R1 . C1) = 720 Hz and the high roll-off is Fc = 1/ (2 . pi . R2 . C2) = 5660 Hz, where in this circuit R1=4K7 C1=0.047uF R2=551K C2=51pF. In truth, the roll-off frequency is where the impedance of the imaginary resistors I was talking about in my earlier post is equal. That is, the parallel combination of R2 and C2 is equal to the series value of R1 + C1. This becomes an equation which I couldn't calculate :icon_lol: But in the same way, the roll-off of a single pole RC filter is the point where the impedance of the cap and resistor are the same, R = 1/(2 . pi . f . C), which can be re-arranged (or whatever, English is not my native tongue) to the familiar f = 1/(2 . pi . R . C)
Please read this, this is elaborated more at the end of the article: http://gaussmarkov.net/wordpress/parts/op-amps/op-amps-4-divided-negative-feedback/
QuoteWith AC, does current really "flow" in a particular direction anyway?
There's really no current flowing in the input of the op-amp. It just reacts to voltage changes, as you said. Actually, I was talking about the DC current, in audio circuits you can think of the DC conditions and AC conditions as two separate things. But maybe you are right about that.
Nevertheless, if you take into account the 1K resistor, then on the same basis you should also take into account everything that comes after the op-amp as well. BUT, the whole point of an op-amp is to have infinite input impedance and zero output impedance. What comes after the op-amp shouldn't affect what is before it. Maybe I just trust R.G. Keen's knowledge too much but I think there is no mistake in his article.
QuoteBut for some reason whenever I ask a specific, precise question on this forum just to help clarify my rudimetary understanding, all I ever get are really vague and dismissive answers and it's is very frustrating. I just don't get it. I've never had this experience on any other internet forum. I come on here and I see all this really knowledgeable people designing and building really amazing things, so I know my questions must be simple and elementary, but it's so difficult to get any help or direct answers here. I've honestly had better results posting my electronics related questions on other, non-electronics related music forums than I have here, which I find very odd.
I think most people here are not really that much electronics gurus than they are effects gurus. Many people just learn this stuff by trial and error and by substituting different parts and getting insight at what everything does. It's all so imprecise anyway so your ears are your best guides. But as far as understanding the electronics go, nothing beats proper education and I've learned lots more from this basic electronics course that I'm attending than I've learned reading various pages over the Internet.
Anyway, I think this is a very interesting point you've made and I'm sure someone comes up with a definitive answer. I totally believe you when you say the freq response becomes flat by jumpering the 220 ohm resistor, but I'm just not sure I agree as to why it does so. Thanks for this thread! :)
after sleeping on this, i might have something worthwhile to suggest.
I'm not convinced that the .22uF and 220Ohm act as a low pass entirely, and RG's article argues this as well. The gain of the non-inverting op amp is 1 + R2/R1 where R1 = 1K, and R2 = .22uF + 220Ohm, which means the gain is frequency dependent (we knew this). What I'm trying to suggest though, is that this cannot be explained as a low pass filter exactly. YES, the cut off frequency of R2 (Z2 really) is calculated the same way as any other filter, but its slightly more complicated than a simple roll off. You could calculate the frequency response using complex impedance i suppose. Let me rephrase, I plan on doing that calculation eventually when I'm trying to document some stuff for this circuit, but not today, sorry.
The other thought that just occurred to me, i misstated earlier that this action "undoes" the .22uF/1k roll-off, and that is incorrect. It acts to level it off above 3.2kHz, yes, but frequencies between 720Hz and 3.2kHz are still only affected by the .22uF/1k. THAT is our mid-hump, I believe. It never goes away, since the active tone control only affects the content above 3.2kHz.
I do however, agree with RG's article when it describes the other end of the tone control. The 20K Pot serves to isolate the .22uF and 220o from the feedback, and it provides additional roll off, pre-op-amp, above 3.2kHz. He does not go into detail about the gain of the op-amp in this situation, which I think would be helpful to the reader, but i have a feeling it gets reduced to 1. Again, I'm not 100% on this.
Fun discussion. Again, I'm not trying to answer you entirely, I'm just learning this stuff currently myself.
One thing to keep in mine i think is, the the Clipping stage adds gain above 720Hz, and a lot of upper frequency content, so while a tone control or stack might seem like its limiting things a lot, its actually just reigning in the harmonics so that things aren't harsh. I struggled with that for a while.
Quote from: aziltz on April 09, 2009, 09:10:11 AM
after sleeping on this, i might have something worthwhile to suggest.
I'm not convinced that the .22uF and 220Ohm act as a low pass entirely, and RG's article argues this as well. The gain of the non-inverting op amp is 1 + R2/R1 where R1 = 1K, and R2 = .22uF + 220Ohm, which means the gain is frequency dependent (we knew this). What I'm trying to suggest though, is that this cannot be explained as a low pass filter exactly. YES, the cut off frequency of R2 (Z2 really) is calculated the same way as any other filter, but its slightly more complicated than a simple roll off. You could calculate the frequency response using complex impedance i suppose. Let me rephrase, I plan on doing that calculation eventually when I'm trying to document some stuff for this circuit, but not today, sorry.
The other thought that just occurred to me, i misstated earlier that this action "undoes" the .22uF/1k roll-off, and that is incorrect. It acts to level it off above 3.2kHz, yes, but frequencies between 720Hz and 3.2kHz are still only affected by the .22uF/1k. THAT is our mid-hump, I believe. It never goes away, since the active tone control only affects the content above 3.2kHz.
I do however, agree with RG's article when it describes the other end of the tone control. The 20K Pot serves to isolate the .22uF and 220o from the feedback, and it provides additional roll off, pre-op-amp, above 3.2kHz. He does not go into detail about the gain of the op-amp in this situation, which I think would be helpful to the reader, but i have a feeling it gets reduced to 1. Again, I'm not 100% on this.
Fun discussion. Again, I'm not trying to answer you entirely, I'm just learning this stuff currently myself.
One thing to keep in mine i think is, the the Clipping stage adds gain above 720Hz, and a lot of upper frequency content, so while a tone control or stack might seem like its limiting things a lot, its actually just reigning in the harmonics so that things aren't harsh. I struggled with that for a while.
Interesting...
I think you are right that the interaction of the feedback network is probably more complicated than I have explained it, but it still doesn't explain what happens when you remove the 220 ohm resistor. I would suggest anyone to just try it. It takes five minutes to breadboard the tone control circuit. You don't even need the pot. Just use a resistor in it's place since we are only discussing the tone knob at full position. Then, put the whole thing on a dpdt bypass switch. If you jumper the 220 ohm resistor, and flip the switch back and forth, the two signals are remarkably similar. I can't say that they are exact, because there is always going to be some interactions when you are filtering and boosting signals like that, but it sounds nearly the same.
The only way I can figure out how to describe this is to say that the 1K resistor and .22uf cap act as a 720Hz rolloff in the feedback network, which essentially boosts the same frequencies that were just cut earlier by the same amount, giving a flat response to the whole filter network. When I run sine sweeps through the circuit and look at it on a freqency spectrum analyzer I see the same thing: it looks ruler flat. When I put the 220 ohm resistor back in the circuit it is a little bit harder to see exactly what frequency is the center of the hump because the slope is so gradual, but I am definitely seeing a broad hump and it's peak is definitely much lower than 3.2Hz. What I am clearly NOT seeing is what is described in the GeoFX article. If frequencies were being rolled off below 720Hz and brought back up above 3.2Hz, then I would expect to see a dip in the frequency content between 720Hz and 3.2Hz. I am not seeing any kind of dip like that on the plot, and I don't hear anything like that either. The content between 720Hz and 3.2Hz would be the bulk of what most people consider "mid-range" content (I usually consider mids to be 300Hz-5KHz, but I couldn't find a standard definition), so it should clearly sound like a mid-range scoop if the filter works as described in the GeoFX article. I've never heard anyone describe the Tubescreamer as having a mid range scoop, period. It's always been known for it's mid-range presence.
The only thing I can figure is that the GeoFX article is wrong, which baffles me because that article has been so widely circulated that I would think that by now someone would have caught the error. I really wish I could just get someone to confirm this so I can move on. Where are all the "expert" engineers here that can calculate R-C filters in their sleep? I get the distinct feeling that I am being ignored by the folks who could answer this easily. I can't blame them. They are probably sick of reading threads about tubescreamers, but thank you to those who have tried to help. It has been much appreciated.
Quote from: Projectile on April 10, 2009, 02:16:17 AM
What I am clearly NOT seeing is what is described in the GeoFX article. If frequencies were being rolled off below 720Hz and brought back up above 3.2Hz, then I would expect to see a dip in the frequency content between 720Hz and 3.2Hz. I am not seeing any kind of dip like that on the plot, and I don't hear anything like that either. The content between 720Hz and 3.2Hz would be the bulk of what most people consider "mid-range" content (I usually consider mids to be 300Hz-5KHz, but I couldn't find a standard definition), so it should clearly sound like a mid-range scoop if the filter works as described in the GeoFX article. I've never heard anyone describe the Tubescreamer as having a mid range scoop, period. It's always been known for it's mid-range presence.
The 720Hz cut off is a low pass, meaning it cuts frequencies
above 720Hz. Our confusing tone control only acts to control things above 3.2kHz. Considering the gain of the clipping stage is 1 below 720Hz and higher above, everything running into the tone section has a lot more harmonic content above 720Hz. The Tone section attemps to reign that back in to mostly flat, and in the process, adds a mid hump somewhere bertween 702 and 3.2 like you said. I don't think the Geo Article is incorrect here.
For reference, here's the section on Tone:
"Tone and volume control stages
Cutting out harsh high frequency harmonics seems to be one of the underlying principles of the TS series. Following the clipping stage there is a 1K resistor leading to a 0.22uF capacitor to ground. This acts like a simple RC lowpass filter, with the rolloff point being 723Hz. This means that the output of this stage is down 20db (10:1) at 7230 Hz, and another 6db (20:1) at 14KHz, close to the top of the audio range. From this simple lowpass filter, the signal goes to the active tone control stage. The control is a 20K potentiometer strung from the (-) to the (+) input of the second opamp section. The wiper of the control is tied to a series RC combination to ground. This RC is a 220 ohm resistor and a 0.22uF capacitor. As a series network, at frequencies above the point at where the capacitive impedance is less than 220 ohms (which happens at about 3.2KHz), the network just looks like the 220 ohm resistor. At frequencies below that point, the capacitor impedance gets larger as the frequency goes down until at some point the capacitor impedance is large even compared with the full resistace of the tone control (20K); this happens at about 36Hz, below guitar frequencies.
The tone control operations are easiest to see if you assume that the tone control is at one end or the other of its range. When fully toward the (+) end, the capacitor shunts the frequencies above 3.2KHz to ground; when fully toward the (-) end, the capacitor shunts feedback frequencies above 3.2KHz to ground. This means that at the (+) side, the signal gets another -6db/octave high frequency rolloff, while when it's at the (-) side the signal finally gets some treble boost, +6db/octave above 3.2KHz. Note that the "boost" actually just levels off the -6db/octave induced by the 1K/0.22uF network ahead of the active control stage, so the treble is just not being cut any more above the turnover frequency for the tone control stage when fully at "treble".
The opamp is set up as a noninverting buffer, which just means that there is no net signal loss through the tone control stage, a gain of 1 - if you can find a frequency where there isn't otherwise a boost or cut from something else. "
Quote from: aziltz on April 10, 2009, 11:48:51 AM
Quote from: Projectile on April 10, 2009, 02:16:17 AM
What I am clearly NOT seeing is what is described in the GeoFX article. If frequencies were being rolled off below 720Hz and brought back up above 3.2Hz, then I would expect to see a dip in the frequency content between 720Hz and 3.2Hz. I am not seeing any kind of dip like that on the plot, and I don't hear anything like that either. The content between 720Hz and 3.2Hz would be the bulk of what most people consider "mid-range" content (I usually consider mids to be 300Hz-5KHz, but I couldn't find a standard definition), so it should clearly sound like a mid-range scoop if the filter works as described in the GeoFX article. I've never heard anyone describe the Tubescreamer as having a mid range scoop, period. It's always been known for it's mid-range presence.
The 720Hz cut off is a low pass, meaning it cuts frequencies above 720Hz. Our confusing tone control only acts to control things above 3.2kHz. Considering the gain of the clipping stage is 1 below 720Hz and higher above, everything running into the tone section has a lot more harmonic content above 720Hz. The Tone section attemps to reign that back in to mostly flat, and in the process, adds a mid hump somewhere bertween 702 and 3.2 like you said. I don't think the Geo Article is incorrect here.
Sorry, but I can't make a bit of sense out of that argument. If the first part of the filter is rolling off at 720Hz and the second part of the tone control is only controlling frequencies above 3.2KHz, then there is no way it could be adding a hump between 720 and 3.2K. It would be an audible cut in that range, regardless of how much hamonic content the clipping stage is adding there. If you think the gain stage is adding THAT much midrange content then just take it out of the equation all together. Let's just talk about the tone control by itself. I'll bypass the clipping stage and do my tests again.
This is getting ridiculous. Why is it so difficult for someone to confirm whether or not the 1K resistor in the feedback loop is supposed to be taken in account when calculating the rolloff frequency of the .22 cap and 220 ohm resistor to ground? It is a very simple question, and that is all I'm basically asking. If you are trying to find a way to fit the theory in the GeoFX article to what I am finding, then start with the fact that the frequency response of the whole filter section goes basically flat when I jumper the 220 ohm resistor. That is something I can measure with near certainty.
In the meantime, does anyone knows of any good measurement software that can actually send out a sine sweep and generate a frequency plot? I would be very grateful. At the moment I am just using a digital synth with an LFO routed to a self-oscillating filter as a signal generator, and a audio visualizer tool as a specrum analyser. It's a pretty rough way of doing it, but it works for getting a general picture. I would prefer something that could give me a bit more accurate reading though. Thanks.
duplicate post deleted.
Forgive me for trying to help, perhaps I miss-communicated. I was just pointing out that its a Fixed Low Pass at 720Hz, not High Pass as you said in the post previous to mine.
I think what I said makes a lot of sense, and the Geo article is in line with that as well, I'm just stating that I agree with RG's explanation thus far.
The clipping stage increases everything over 720Hz.
Between Op-Amps, everything over 720 is reduced (6dB/Octave).
The tone control has a gain of 1, and can either cut above 3.2kHz at the input, or cut in the Feedback loop (adding more highs to the output).
I never claimed that the tone control added the mid-hump on it's own. In fact, I believe its the combination of the Clipping Amp's Filtering and the Fixed Low-Pass between stages, but I haven't had time to calculate that. That's just my hunch. I don't believe you'll find a mid-hump without the Clipping amp.
On to your actual question:
The easiest way to see what the 1k Resistor does its to put a pot in there. My bet all along has been that its just setting the gain to 1, but I cannot be sure, because I don't recognize the 20K Pot between (+) and (-) inputs as any typical op-amp circuit. You seem capable of doing this. I can suggest Spice or 5Spice or LTSpice, a program where simulating would be very simple. I'd do it myself, but I haven't learned the ins and outs yet. It's coming in a week or two for the lab I'm teaching.
EDIT: Thinking a little more, I'd like to suggest that the 220 and .22uF is not a typical low pass filter in that, its not a simple voltage divider with upper leg R, lower leg C. I see it more as, a side path, or a dead end side street. (Getting figurative here i suppose) Vin to (+), has a side route to ground via the 220/.22uF. The 20k controls the amount dumped to ground, but doesn't affect the cut-off point. Likewise, Vout runs through the 1K, to the (-) and has a side route to ground through the 220/.22uF, amount controlled by the 20k. I believe this is because the main signal is not passing straight through the 20K resistor as it is in the 1k/.22uF low-pass prior to the tone control. To test this I would disconnect the 20K from the (+) input. If this is the case, I believe the Tone Pot would only INCREASE treble, because it can only cut the treble on the feedback leg, but not the input. Just a theory.
I can, however, offer an explanation for why the response goes flat when you short the 220ohm resistor. Recalculating what the cut-off would be, assuming a resistance of 1 for the shorted resistor, the cut-off frequency is now 723kHz!!!, well above the human range of hearing. The 220oHm resistor controls the cut-off, put a pot in there and you've got a way to control the cut-off frequency as well.
For the record, I'm not some newb throwing out suggestions out of my rear. I'm an experimental research physicist. I work with electronics everyday and on during any other week I would love to breadboard this, document it, and give a definitive answer. I just may do that for myself at some point, but I haven't as much as a free moment right now. My apologies that all I can offer is discussion.
Quote from: aziltz on April 10, 2009, 10:01:55 PM
Forgive me for trying to help, perhaps I miss-communicated. I was just pointing out that its a Fixed Low Pass at 720Hz, not High Pass as you said in the post previous to mine.
I think what I said makes a lot of sense, and the Geo article is in line with that as well, I'm just stating that I agree with RG's explanation thus far.
The clipping stage increases everything over 720Hz.
Between Op-Amps, everything over 720 is reduced (6dB/Octave).
The tone control has a gain of 1, and can either cut above 3.2kHz at the input, or cut in the Feedback loop (adding more highs to the output).
I never claimed that the tone control added the mid-hump on it's own. In fact, I believe its the combination of the Clipping Amp's Filtering and the Fixed Low-Pass between stages, but I haven't had time to calculate that. That's just my hunch. I don't believe you'll find a mid-hump without the Clipping amp.
On to your actual question:
The easiest way to see what the 1k Resistor does its to put a pot in there. My bet all along has been that its just setting the gain to 1, but I cannot be sure, because I don't recognize the 20K Pot between (+) and (-) inputs as any typical op-amp circuit. You seem capable of doing this. I can suggest Spice or 5Spice or LTSpice, a program where simulating would be very simple. I'd do it myself, but I haven't learned the ins and outs yet. It's coming in a week or two for the lab I'm teaching.
EDIT: Thinking a little more, I'd like to suggest that the 220 and .22uF is not a typical low pass filter in that, its not a simple voltage divider with upper leg R, lower leg C. I see it more as, a side path, or a dead end side street. (Getting figurative here i suppose) Vin to (+), has a side route to ground via the 220/.22uF. The 20k controls the amount dumped to ground, but doesn't affect the cut-off point. Likewise, Vout runs through the 1K, to the (-) and has a side route to ground through the 220/.22uF, amount controlled by the 20k. I believe this is because the main signal is not passing straight through the 20K resistor as it is in the 1k/.22uF low-pass prior to the tone control. To test this I would disconnect the 20K from the (+) input. If this is the case, I believe the Tone Pot would only INCREASE treble, because it can only cut the treble on the feedback leg, but not the input. Just a theory.
I can, however, offer an explanation for why the response goes flat when you short the 220ohm resistor. Recalculating what the cut-off would be, assuming a resistance of 1 for the shorted resistor, the cut-off frequency is now 723kHz!!!, well above the human range of hearing. The 220oHm resistor controls the cut-off, put a pot in there and you've got a way to control the cut-off frequency as well.
For the record, I'm not some newb throwing out suggestions out of my rear. I'm an experimental research physicist. I work with electronics everyday and on during any other week I would love to breadboard this, document it, and give a definitive answer. I just may do that for myself at some point, but I haven't as much as a free moment right now. My apologies that all I can offer is discussion.
Sorry, but much of what you have said is about the way this tone control circuit functions is completely wrong. I have done a lot of experimenting and I am 100% sure at this point.
First of all if you read what I posted earlier, I already HAVE disconnected the tone control pot from the positive input of the opamp, and like I said earlier, "It doesn't make a discernible difference in how the tone control functions when the pot is turned all the way up." I can disconnect the wire and barely hear a difference. So, we can ignore that.
Your explanation of how the signal can goes flat when I remove the 220 ohm resistor doesn't make any sense. I am really starting to think you are pulling my chain. If the first part of the tone control network rolls off 6db/octave from 720Hz up (before the feedback network), then how on earth could the signal be flat if the feedback network has a roll off point beyond the range of human hearing!??? Huh!?
I apologize if I am coming off as abrasive, but I am very frustrated. You see, I am not a theoretical research physicist. I am merely an unemployed college drop-out who decided to build a tube screamer for fun. I don't know calculus or physics, but when I tinker with things I like to have a good working model of how they function in my head, so I started reading about electronics. After a while I felt I had a pretty good understanding of how this stuff works, but some of the models in my head were directly contradicting what I was reading in the GeoFX article, among other places. So, I thought I would ask here to see if anyone could clear up the gaps in my understanding. Instead, what has happened is that it became very clear to me that the explanations I have been given are obviously wrong. In order to confirm my understanding I had to do some experimentation that further proved that the information circulating is bad. I would have thought that upon argument and examination all of the "experts" here would either clear up the issue for me, or simply realize their folly and fix the error. I am a novice at this, but it is painfully obvious to me that my understanding is at least somewhat correct, and that the "experts" here are completely wrong. The data doesn't lie. You can't imagine how frustrating it must be for me as a novice to see very knowledgeable and highly educated people persist in making glaring errors. It's really quite shocking! What in the hell is a newbie like me supposed to do in that kind of situation? Over 200 people have apparently read this thread, but so far all I have gotten in response is more errors. It really bothers me that I could possibly be seeing gaping holes in a theoretical physicist's basic understanding of how electrical currents are controlled by an RC filter, when I'm just a freaking newb! But I know I'm not completely nuts, so what in the hell is going on!!!!???? ARRGGGGGHHHH!!!!!
Before your mind blows up, I'm going to throw in my 2 cents again.
QuoteI can, however, offer an explanation for why the response goes flat when you short the 220ohm resistor. Recalculating what the cut-off would be, assuming a resistance of 1 for the shorted resistor, the cut-off frequency is now 723kHz!!!, well above the human range of hearing. The 220oHm resistor controls the cut-off, put a pot in there and you've got a way to control the cut-off frequency as well.
First of all, this doesn't seem right to me, aziltz. As you said, it's not a typical voltage divider filter, so you can't calculate the roll-off frequency like this? The roll-off should be Z1=Z2, where Z1=220R+1/(2*pi*0.22uF*f) and Z2=the resistance of the tone control pot. This equation should be solved for f (frequency) and it's made more complicated by the fact that it affects both the non-inverting and inverting sides of the op-amp. If the 220ohm resistor is shorted, then the tone control looks like a normal low pass filter with the tone control as a variable resistor. The roll-off seen from the (+) side should simply be: f=1/(2*pi*20k*0.22uF) ~ 36 Hz when the tone control is fully on. But this only concerns the frequencies from the (+) side of the op-amp to the (-) side. So it's actually more complicated than you might first think.
I tried to simulate this circuit in LTSpice, but I already deleted the results, so forgive me for not posting the graphs. The tone control only added a very slight hump at about 1 kHz when fully on, but I agree with aziltz that the hump is mainly caused by the combined efforts of the clipping stage and the tone stage. Shorting the 220ohm resistor produced a quite flat frequency response for the tone stage when the tone control was fully on, but with the tone down the treble was rolled down.
You guys are killing me.
I have some pretty accurate data for the isolated tone control section now, and the mid hump IS mostly from the clipping section after all, but I'm still pretty sure my general idea about how the tone control section works and how the roll-off value is calculated are for the most part correct... at least more in the right general direction than the GeoFX article, which is completely wrong. The data seems to back that up although the actual interactions are a little bit more complicated than I first understood. You MUST account for the 1K resistor in the calculation, and the 220 resistor also. The 220ohm resistor's interaction is a lot more complicated because it doesn't just set the roll off. I don't know where you guys learned how to apply all of these formula, but your basic concepts about how the R-C filter works in this circuit leave me scratching my head. I'm not going to claim to fully understand this stuff, but I can definitely say that the way you guys analyze this stuff seems to contradict very basic foundational concepts about how the different components of a circuit interact with electrons. That's what's really bothering me.
I'll post some stuff when I get all my data collected, but I'm still working on the clipping section to figure out exactly how the caps in that filter work. It seems the AMZ article is also wrong. WTF?!!! Once again, I honestly don't know where you guys learned how to apply these formula, but I think everyone needs to go back and re-learn some of the basic concepts of how these things function, because there is obviously a lot of errors and misunderstanding floating around. I've done a ton of reading and research at this point and I'm 99% sure my general thinking is correct and the internet is freaking crazy. Weird.
The thing is, this tone circuit is not your basic RC-filter, so your basic formulas won't apply unless you formulate your own. I've never seen an op-amp with the inverting and non-inverting inputs connected together like that, either. I'd really like to hear you view on how these components interact with electrons. Maybe it would clear some things up. There have been many times that I've been sure I understand something perfectly only to be proven completely wrong with an even more logical explanation. I don't claim that I understand this circuit perfectly and I'm not claiming that you're wrong, I just hope someone like R.G. himself could swoop in here and explain his article.
Quote from: MohiZ on April 12, 2009, 06:47:49 AM
The thing is, this tone circuit is not your basic RC-filter, so your basic formulas won't apply unless you formulate your own. I've never seen an op-amp with the inverting and non-inverting inputs connected together like that, either. I'd really like to hear you view on how these components interact with electrons. Maybe it would clear some things up.
Well, I'm not going to get into the gory details at the moment, but as far as the negative and positive terminals of the opamp being connected together is concerned it's pretty straightforward. Ignoring the 220ohm resistor at the moment, which just complicates things, when the wiper is all the way toward the negative terminal the 20K impedance of the pot + the 1K resistor after the output of the first opamp equals 21K in an RC combination with the .22uf cap on the wiper. This passes nearly the entire frequency range for the positive input side of the tone control to ground, so you can essentially ignore it. In fact, if you dissconect the lug from the positive terminal while the pot is turned all the way up you can barely detect an audible difference. You can't really see a difference on the frequency plot either. The same thing happens when you turn the pot all the way in the other direction: you get 20K +1K in the feedback network and .22uf on the wiper, essentially taking it out of the equation.
Furthermore, when the pot is rotated to it's midpoint you basically have 11K with .22uf RC filter from one lug, and 11K with with .22uf RC on the other lug. Now, I know the interactions of these filters is far more complicated than i am describing because the cap is seeing current from the output of both opamp stages, but for all practical purposes we can say that the rolloff is still really low, likely below 100hz, so we can basically take the entire tone control pot out of the picture at it's midpoint. If you remove both lugs while the pot is at it's midpoint there is only barely a discernible difference in the frequency response by ear, and you can also barely see any change on the frequency plot. What we have is essentially a tone control that only does anything at it's extremes. This is probably why it uses that fancy "W taper" or "eq taper" pot, which is gradual at either end and very steep in the center. Though even with such a steep change at the center, the biggest eq changes only happen at the top quarter and bottom quarter rotation of the pot. It's pretty dead in the middle.
Quote from: MohiZ on April 12, 2009, 06:47:49 AM
There have been many times that I've been sure I understand something perfectly only to be proven completely wrong with an even more logical explanation. I don't claim that I understand this circuit perfectly and I'm not claiming that you're wrong, I just hope someone like R.G. himself could swoop in here and explain his article.
That's why I posted here in the first place. I fully expected to get a more logical explanation.
I simulated the opamp with the tone control with Aplac, a circuit analysis software and here are the results. I ignored the 1k/.22u low-pass filter right before the op-amp for the sake of simplicity. When looking at these results you have to remember it cuts off some treble, and the clipping stage freq response also affects the total freq response at this point.
Here is the response at the output of the opamp, with 5 different tone control settings. We can clearly see that all it does is boost treble. The bass frequencies have an amplification of 0 dB, that is gain of 1. The point of 3 decibel boost seems to be a little less than 600 Hz.
(http://www.aronnelson.com/gallery/main.php?g2_view=core.DownloadItem&g2_itemId=39496&g2_serialNumber=1)
Jumpering the 220 ohm resistor only seems to steepen the treble boost curves:
(http://www.aronnelson.com/gallery/main.php?g2_view=core.DownloadItem&g2_itemId=39499&g2_serialNumber=1)
The 1k resistor affects gain. Of course, the 3dB roll-off point changes, but it seems it's only because the amount of total boost changes - the curves stay the same shape. Note the different vertical scale:
(http://www.aronnelson.com/gallery/main.php?g2_view=core.DownloadItem&g2_itemId=39502&g2_serialNumber=1)
These might be slightly more informative. Here's the response of the whole tone stage, with the RC lowpass filter at the beginning. I've arranged the scales so that you can directly see the 3dB cutoff frequency at the bottom of the graph.
http://www.aronnelson.com/gallery/main.php/v/diyuser/ts3.JPG.html?g2_imageViewsIndex=1
It's interesting to note that the op-amp does not affect the frequency response AT ALL when the tone is rolled full down. It's also interesting to see that you were totally right, the frequency response is flat with the 220ohm resistor shorted. But, only when the tone control is on full. There's still a mid hump when the tone control is somewhere in the middle.
The highs-boosting op-amp seems to create the mid hump in conjunction with the low-pass filter right before it. The peak is, however, only about one decibel, so I'm not sure if it's even audible. It does seem that the Geofex article is wrong.
(http://www.aronnelson.com/gallery/main.php/v/diyuser/ts3.JPG.html?g2_imageViewsIndex=1)
Quote from: Projectile on April 12, 2009, 07:55:36 AM
That's why I posted here in the first place. I fully expected to get a more logical explanation.
While I think you'd get a lot opinions that you can trust, a lot of people on here are simply enthusiasts, but I don't think people overstep when giving advice. Let's face it, you don't need an EE degree to build stomp boxes, not to say there aren't some fabulously brilliant people on here. Of course there are.
I've been responding because I was interested in this topic, and I wasn't seeing many members contribute. I tried to offer a few educated guesses, but I made sure to indicate what was a hunch. There are no "gaping hole's in my understanding of how RC filter's affect electrical currents", and I take offense to that remark. No, I don't know everything about this circuit. I haven't yet studied it, but as I said before, this is not a basic application of op-amps and RC Filters. I said I was an experimentalist, not a theorist, and there's definitely a difference. My occupation should make no difference here, and your sarcastic and snide remarks are uncalled for. I was never preaching to you, or anyone, about this. I just haven' had the time to breadboard this, and so for the sake of discussion, I offered a few suggestions, based on the basics of RC Filters and Op-Amps.
I feel like I've been set up to be shut down. You solicited explanations, then ran your own tests. No need to call bullshit on us here. Your posts have included more and more examples of things you've discovered on the breadboard, and I applaud you for testing things out. I would be really excited to read all your findings once you complete what you are testing.
Discussion and suggestion really are the cornerstone of experimentation and ultimately, research and education. If I've offended your or wasted your time by offering up what were my first instincts as to how the circuit works, then I apologize. Perhaps it would have been better for my occupation to remain unmentioned. I think I am done sharing on this topic for now. I don't wish to create enemies in this community, especially over a misunderstanding such as this. Good Luck in your endeavors, and I hope you find the answer and share it for others to hear.
MohiZ,
Thank you for your simulations. I think these kinds of calculations are a great tool here dealing with these kinds of questions. I hope I can learn to use these programs as effectively in the next few weeks so I can begin my own explorations.
Sorry Aziltz, but I came here as a novice looking for advice from people who I assumed were clearly more knowledgeable than I am. The issue seemed fairly straight forward to me and I just got frustrated to no end when after hundreds of people had read my post I was still getting back responses from obviously educated people that appeared very clearly to be wrong, and no one was willing to back me up or give me a correction in any way that makes sense. I apologize for being abrasive.
Thank you MohiZ for your analysis. Your findings are pretty close to what I am seeing since I isolated the tone control and refined my simple measurement system. At this point I still don't quite understand exactly how the 220 resistor effects the filter, but it is very clear that it works in conjunction with the 1K resistor to set the roll off point (in this case the boost point) to a value slightly below 720Hz, but it seems to only effect the roll off point very slightly. Most of its effects seems to be more in softening the curve of the filter. It doesn't contribute to the mid hump as much as I had thought, but the fact remains that the 220ohm resistor and the .22 cap alone do not set the roll off point of the filter. I don't know where so many people got the notion that you can calculate RC filters that way, because the AMZ article makes the same mistake in its analysis of the clipping section of the tube screamer. It seems very obvious to me with only my elementary knowledge of how these components work that you cannot calculate a filter that way. I'm just baffled that so many people have read this post and those articles are so well circulated, yet nobody has caught these errors when it seems so apparent to just a novice like me. I find it very odd, and the whole thing has left me rather frustrated and confused. Thank you for posting some data to help to clear up the confusion. Now at least I know I'm not completely crazy.
QuoteMost of its effects seems to be more in softening the curve of the filter. It doesn't contribute to the mid hump as much as I had thought, but the fact remains that the 220ohm resistor and the .22 cap alone do not set the roll off point of the filter. I don't know where so many people got the notion that you can calculate RC filters that way, because the AMZ article makes the same mistake in its analysis of the clipping section of the tube screamer.
I agree with you there. The 220ohm resistor and .22 cap don't even LOOK like an RC filter. I'm pretty sure the 1K resistor only sets the gain of the op-amp. But of course that affects the roll-off point of the stage, because the op-amp only boosts highs. Could you post a link to the AMZ article?
Sorry for not being able to answer your question, but you see, I'm relatively new to this stuff, too. I've done tons of research and studying, though. It seems strange that there's not many members answering to this thread. I'm sure there are people on the forum who could answer this. Maybe they got bored reading the long posts :icon_biggrin:
Here's the AMZ article:
http://www.muzique.com/lab/fatt.htm
I'm in the middle of doing frequency plots of this filter right now, but I'm having trouble because when I remove the diodes it's very easy to send the opamp into clipping when I mess with resistor values. Regardless, it is pretty clear at this point that the AMZ article is also wrong. The 4.7K resistor once again does not by itself set the rolloff of the filter. The 51K resistor is what actually interacts most with the .047 cap to set the rolloff point. As you turn up the gain pot and add more resistance to the feedback loop, the rolloff point drops and more bass comes through. What makes it more complicated though is when you add the diodes in. When the voltage gets high enough the diodes open and the cap sees more current, so the filter starts cutting more highs into the negative input. If the diodes were to stay completely open then the cap would block all frequencies, but since the diodes are usually in some state part way between fully open and fully closed, they just seem to put a kink in the whole interaction and soften the effect of the gain pot on the roll off point. What is clear is that the gain pot still effects the roll off point. It is definitely NOT set alone by the 4.7K resistor, like it says in the AMZ article, and it is not 720Hz. It depends greatly on the amount of current in the feedback loop, which is controlled by the pot and the 51K resistor.
This is what I would expect from my elementary understanding, and it is in line with what I am seeing on the frequency plots. I just can't figure out how these "experts" keep getting these things wrong when it seems so obvious to me as a novice that they wouldn't work like that. I also don't understand why there is a complete lack of interest in the fact that there are all of these errors and misunderstandings circulating about. I don't get it. :icon_confused:
I'll try and add something to the discussion but bear in mind that I'm not an expert, and what I'm saying is based on what little knowledge I have and from playing with LTSpice for half an hour.
First read this http://en.wikipedia.org/wiki/Cutoff_frequency (http://en.wikipedia.org/wiki/Cutoff_frequency) especially the bit about the -3dB point. With a standard RC filter with no active parts involved the cutoff frequency, which is the point where the output is 3dB less than the input is set by the equation F = (2*Pi*R*C). Depending on whether it's a high pass or a low pass filter the volume then drops off on a slope above or below these frequencies.
I'm saying this so that hopefully we're on the same page for the next bit, or at least in the same book :)
To look at the first stage what I did was remove the diodes and run it at 30volts so there's no clipping messing things up.
With the gain set at minimum so you've just got the 51k resistor and 51p cap in the feedback loop, you get a maximum boost of about 21 dB, if you follow the line down to the point 3dBs lower, ie:18 dB you should see that it's at about 720Hz. Which is where the RC filter equation for the 4k7 resistor and 47n cap says it should be.
If you then make the total resistance in the feedback loop 551k (maximum gain) you get a maximum gain of about 40dB and the -3dB point appears to be at about 550Hz. I think this is because like R.G. says in the TS article the lowpass cutoff frequency of the 51p and the resistance in the feedback loop lowers as the resistance increases. This then starts to interact with everything else so you don't get the answer you were expecting. To verify this if you remove the 51p the maximum gain increases slightly and the -3dB point moves back to 720Hz.
So basically I don't think there's anything incorrect in R.G.s or Jack's articles. The RC filter cutoff for the 4k7 and 47n resistor is 720Hz, it's just that the other things in the circuit change the overall frequency response. They aren't changing what those 2 components do.
They're maybe not giving you the full picture or giving a simplified explanation, but they're not wrong in what they're saying.
I'm guessing its a similar story for the tone control but I haven't had chance to look at it in much detail. I think the portion of the tone control between the 200n and the negative input added to the 220R resistor set the rolloff frequency.
Don't know if any of that helps, it might make someone who knows what they're talking about jump in and tear what I've said to pieces if nothing else ;D
Quote from: slacker on April 13, 2009, 07:56:20 AM
To look at the first stage what I did was remove the diodes and run it at 30volts so there's no clipping messing things up.
With the gain set at minimum so you've just got the 51k resistor and 51p cap in the feedback loop, you get a maximum boost of about 21 dB, if you follow the line down to the point 3dBs lower, ie:18 dB you should see that it's at about 720Hz. Which is where the RC filter equation for the 4k7 resistor and 47n cap says it should be.
If you then make the total resistance in the feedback loop 551k (maximum gain) you get a maximum gain of about 40dB and the -3dB point appears to be at about 550Hz. I think this is because like R.G. says in the TS article the lowpass cutoff frequency of the 51p and the resistance in the feedback loop lowers as the resistance increases. This then starts to interact with everything else so you don't get the answer you were expecting. To verify this if you remove the 51p the maximum gain increases slightly and the -3dB point moves back to 720Hz.
THAT, makes a lot of sense to me.
One thing we all need to keep in mind here is, the easiest way to describe the way something works its to break it down and simplify it. Take RG's explanation of the Tone Curve of the TS Clipping Stage as an example. Yes, the full frequency response depends on everything attached to the op-amp, but the easiest way to describe it (in words) is to separate the 4k7/4n7 720Hz Low Pass from the 51pF/1M Gain control. While I can't speak for them directly, I think this is what the Author's intended to do with their articles.
Thank god for these simulation tools though!
Quote from: slacker on April 13, 2009, 07:56:20 AM
I'll try and add something to the discussion but bear in mind that I'm not an expert, and what I'm saying is based on what little knowledge I have and from playing with LTSpice for half an hour.
First read this http://en.wikipedia.org/wiki/Cutoff_frequency (http://en.wikipedia.org/wiki/Cutoff_frequency) especially the bit about the -3dB point. With a standard RC filter with no active parts involved the cutoff frequency, which is the point where the output is 3dB less than the input is set by the equation F = (2*Pi*R*C). Depending on whether it's a high pass or a low pass filter the volume then drops off on a slope above or below these frequencies.
I'm saying this so that hopefully we're on the same page for the next bit, or at least in the same book :)
To look at the first stage what I did was remove the diodes and run it at 30volts so there's no clipping messing things up.
With the gain set at minimum so you've just got the 51k resistor and 51p cap in the feedback loop, you get a maximum boost of about 21 dB, if you follow the line down to the point 3dBs lower, ie:18 dB you should see that it's at about 720Hz. Which is where the RC filter equation for the 4k7 resistor and 47n cap says it should be.
If you then make the total resistance in the feedback loop 551k (maximum gain) you get a maximum gain of about 40dB and the -3dB point appears to be at about 550Hz. I think this is because like R.G. says in the TS article the lowpass cutoff frequency of the 51p and the resistance in the feedback loop lowers as the resistance increases. This then starts to interact with everything else so you don't get the answer you were expecting. To verify this if you remove the 51p the maximum gain increases slightly and the -3dB point moves back to 720Hz.
So basically I don't think there's anything incorrect in R.G.s or Jack's articles. The RC filter cutoff for the 4k7 and 47n resistor is 720Hz, it's just that the other things in the circuit change the overall frequency response. They aren't changing what those 2 components do.
They're maybe not giving you the full picture or giving a simplified explanation, but they're not wrong in what they're saying.
I'm guessing its a similar story for the tone control but I haven't had chance to look at it in much detail. I think the portion of the tone control between the 200n and the negative input added to the 220R resistor set the rolloff frequency.
Don't know if any of that helps, it might make someone who knows what they're talking about jump in and tear what I've said to pieces if nothing else ;D
Once again, that doesn't make any sense. You people are driving me freaking crazy!
If you roll off highs in a negative feedback loop, you BOOST highs in the output, which looks like a bass cut, but what you are actually doing is boosting highs. The point at which the highs are down 3db because of the filter isn't going to be where the bass rolls off 3db at the output, it's going to be where the BOOST STARTS, which is AT THE BOTTOM OF THE CURVE.
I'm trying to help, but I seriously can't put up with this nonsense any longer unless someone is going to back me up. I feel like I'm wasting my time and just going around in circles. The GeoFX article is clearly wrong. We have already seen that with Mohiz's Aplac analysis. I feel like we are moving backwards instead of making progress and it is very frustrating. I am fairly confident that the AMZ article is also wrong. I'm kind of shocked that nobody else can see these clear errors. Everyone just seems intent on kludging the data to fit the GeoFX and AMZ articles rather than stopping for a moment and actually thinking for themselves.
Look, you can't build an RC filter like this:
(http://img13.imageshack.us/img13/1083/rc1l.jpg)
I'm a newb at this stuff, and even that is very clear to me. I don't see why you people can't understand that. If you still can't see why that can't be an RC filter by itself, then may god help you because you obviously don't understand the very basics of how electrical circuits work.
In order have a filter of any kind, you have to create some sort of voltage divider. Like this:
(http://img518.imageshack.us/img518/334/rc2a.jpg)
R1 is not necessary for creating a filter, but R2 is essential! You cannot calculate the rolloff of this filter without taking in account R2, period. You also have to account for R1 in the calculation, but just using R1 and C1 to calculate the filter is ridiculous. It doesn't make one bit of sense. Why is that so hard to understand?
Now look at the tubescreamer circuit. The output of the opamp is the source of the voltage potential. The filter in the gain stage is on the other side of the 51K resistor and the tone pot. The 51K resistor + the tone pot, in combination the .047 cap is what calculates the rolloff. The 4k7 resistor is involved in that calculation too, but I don't understand exactly how it works into the calculation, just that is softens the curve and appears to also effect the rolloff by some small amount. This is, of course, with the diodes TAKEN OUT of the feedback loop. If you JUMPER the diodes, then the negative input of the opamp sees the output without any interference by the filter. The filter is essentially out of the equation. Get it?
Same thing with the filter section. We can disconnect the lug of the pot that goes to the positive input of the opamp, because it doesn't really effect the signal when the tone control wiper is all the way on the negative input side of the pot. That has been established. Okay, now the output of the opamp is the source of the voltage potential that the negaive input "sees". The 1K resistor is your R2 from the graphic above, the .22cap is C1, and the 220 ohm resistor is R1. See how that works? You NEED that 1k resistor for the calculation. It is not something you can ignore out of convenience. It is essential to the calculation of the RC rolloff. To just use the 220 resistor and .22cap in the RC calculation is FLAT OUT WRONG. The actual filter is rolling off nowhere near 3.2Khz, period.
I'm getting really frustrated over this and I'm thinking I'm going to abandon this forum altogether. No offense, but I feel like I'm talking to a bunch of dunces here that don't understand simple concepts. I'm a newb for christ's sake! I know you people are not retards. You all seem very intelligent and educated. If anything I am the dunce in the room, which is why I am so incredibly confused? Am I the butt of some elaborate joke? Hello??!!!!!
Quote from: aziltz on April 13, 2009, 03:20:13 PM
Quote from: slacker on April 13, 2009, 07:56:20 AM
To look at the first stage what I did was remove the diodes and run it at 30volts so there's no clipping messing things up.
With the gain set at minimum so you've just got the 51k resistor and 51p cap in the feedback loop, you get a maximum boost of about 21 dB, if you follow the line down to the point 3dBs lower, ie:18 dB you should see that it's at about 720Hz. Which is where the RC filter equation for the 4k7 resistor and 47n cap says it should be.
If you then make the total resistance in the feedback loop 551k (maximum gain) you get a maximum gain of about 40dB and the -3dB point appears to be at about 550Hz. I think this is because like R.G. says in the TS article the lowpass cutoff frequency of the 51p and the resistance in the feedback loop lowers as the resistance increases. This then starts to interact with everything else so you don't get the answer you were expecting. To verify this if you remove the 51p the maximum gain increases slightly and the -3dB point moves back to 720Hz.
THAT, makes a lot of sense to me.
One thing we all need to keep in mind here is, the easiest way to describe the way something works its to break it down and simplify it. Take RG's explanation of the Tone Curve of the TS Clipping Stage as an example. Yes, the full frequency response depends on everything attached to the op-amp, but the easiest way to describe it (in words) is to separate the 4k7/4n7 720Hz Low Pass from the 51pF/1M Gain control. While I can't speak for them directly, I think this is what the Author's intended to do with their articles.
Thank god for these simulation tools though!
No, you cannot separate the 4k7/4n7 720Hz Low Pass from the 51pF/1M Gain control. That doesn't make any sense. It's not merely a convenience. It drastically changes the calculations. This is not an explanation. It is a complete fallacy. You simply cannot do that. There is a world of difference between a cap seeing the low impedance output of an opamp, versus the same cap seeing the high impedance of a 51K pot and a 500K gain control. That impedance is ESSENTIAL in setting the rolloff of the filter. If you take those parts out of the equation, then the filter doesn't work anymore, at all.
AAAAARRRRRGGGGGGHHHHHH!!!!!!!!!
LOOK... you really should be more polite if you want help, even if you disagree. referring to members as dunces and insulting their understanding of electronics will get you no where. you seem like a smart guy. get an reference book and figure it out if you don't like the answers provided, no reason to rip everyone a new one.
good luck.
btw, the 4k7/4n7 isn't a simple RC filter. It sets the gain bandwidth of the clipping stage because, below the RC cut-off, its a essentially an open circuit, above it, its essentially a 4k7 resistor. I say essentially because I'm simplifying things. If you don't want to simplify, play with the differential equations.
(http://i107.photobucket.com/albums/m297/biggy-boy/lawn_chair.gif)
(http://i107.photobucket.com/albums/m297/biggy-boy/popcorn.jpg)
Now I agree with Projectile. If you think about it, considering the tone knob resistance as "large", we can replace it with an open circuit when it's either full on or full off. With this simplification, the tone stage looks like this:
(http://img4.imageshack.us/img4/2470/tubescreamer.jpg)
Now it's starting to look easier to calculate. If we short the 220 ohm resistor, it's even easier. When the tone is full off, after the input there's just a low pass filter with two caps to ground in parallel. And since parallel caps add up, we get its roll-off: f = 1/(2*pi*0.44u*1k) ~ 360 Hz. The op-amp after that is just a unity gain buffer. And take a look back at my analysis here, the middle graph, the first blue line (tone control fully off position): http://www.aronnelson.com/gallery/main.php/v/diyuser/ts3.JPG.html?g2_imageViewsIndex=1
Its 3dB roll-off pretty much exactly 360 Hz in the graph so the approximation was correct!
When the tone is fully on, and the 220 ohm resistor shorted, it's like a LPF with a highs boosting op-amp configuration after it. First of all, the Low Pass Filter roll-off before the op-amp is f = 1/(2*pi*0.22u*1k) ~ 720 Hz. Calculating the op-amp's bandwidth roll-off frequency is easy, too. The 1k resistor from the output and the .22u cap from inverting to ground, form a LPF, so f = 1/(2*pi*0.22u*1k) ~ 720 Hz. Note that everything below this frequency is fed to the inverting input of the op-amp, thus REDUCING those frequencies. Thus, the op-amp boosts frequencies above 720 Hz.
With the LPF and the highs-boosting op-amp the sum of their frequency response should be flat, since both their roll-off is 720 Hz.... and it is! The tone control sort of smoothly morphs between those response curves.
All the 220ohm resistor does in my mind is "prevent" the frequency response from becoming completely flat. As you can see in the graphs, with the 220ohm resistor shorted the graphs look the same at first, but it's as if the range of the tone control is extended in the high frequency range.
So, in light of this realization and my graphic analysis I'd say Projectile is correct and the Geofex article is wrong. Good job pointing that out!
EDIT: Some mild spell checking after proofreading.
Quote from: MohiZ on April 14, 2009, 03:00:15 AM
Now I agree with Projectile. If you think about it, considering the tone knob resistance as "large", we can replace it with an open circuit when it's either full on or full off. With this simplification, the tone stage looks like this:
(http://img4.imageshack.us/img4/2470/tubescreamer.jpg)
Now it's starting to look easier to calculate. If we short the 220 ohm resistor, it's even easier. When the tone is full off, after the input there's just a low pass filter with two caps to ground in parallel. And since parallel caps add up, we get its roll-off: f = 1/(2*pi*0.44u*1k) ~ 360 Hz. The op-amp after that is just a unity gain buffer. And take a look back at my analysis here, the middle graph, the first blue line (tone control fully off position): http://www.aronnelson.com/gallery/main.php/v/diyuser/ts3.JPG.html?g2_imageViewsIndex=1
Its 3dB roll-off pretty much exactly 360 Hz in the graph so the approximation was correct!
When the tone is fully on, and the 220 ohm resistor shorted, it's like a LPF with a highs boosting op-amp configuration after it. First of all, the Low Pass Filter roll-off before the op-amp is f = 1/(2*pi*0.22u*1k) ~ 720 Hz. Calculating the op-amp's bandwidth roll-off frequency is easy, too. The 1k resistor from the output and the .22u cap from inverting to ground, form a LPF, so f = 1/(2*pi*0.22u*1k) ~ 720 Hz. Note that everything below this frequency is fed to the inverting input of the op-amp, thus REDUCING those frequencies. Thus, the op-amp boosts frequencies above 720 Hz.
With the LPF and the highs-boosting op-amp the sum of their frequency response should be flat, since both their roll-off is 720 Hz.... and it is! The tone control sort of smoothly morphs between those response curves.
All the 220ohm resistor does in my mind is "prevent" the frequency response from becoming completely flat. As you can see in the graphs, with the 220ohm resistor shorted the graphs look the same at first, but it's as if the range of the tone control is extended in the high frequency range.
So, in light of this realization and my graphic analysis I'd say Projectile is correct and the Geofex article is wrong. Good job pointing that out!
EDIT: Some mild spell checking after proofreading.
Quote from: MohiZ on April 14, 2009, 03:00:15 AM
Now I agree with Projectile. If you think about it, considering the tone knob resistance as "large", we can replace it with an open circuit when it's either full on or full off. With this simplification, the tone stage looks like this:
(http://img4.imageshack.us/img4/2470/tubescreamer.jpg)
Now it's starting to look easier to calculate. If we short the 220 ohm resistor, it's even easier. When the tone is full off, after the input there's just a low pass filter with two caps to ground in parallel. And since parallel caps add up, we get its roll-off: f = 1/(2*pi*0.44u*1k) ~ 360 Hz. The op-amp after that is just a unity gain buffer. And take a look back at my analysis here, the middle graph, the first blue line (tone control fully off position): http://www.aronnelson.com/gallery/main.php/v/diyuser/ts3.JPG.html?g2_imageViewsIndex=1
Its 3dB roll-off pretty much exactly 360 Hz in the graph so the approximation was correct!
When the tone is fully on, and the 220 ohm resistor shorted, it's like a LPF with a highs boosting op-amp configuration after it. First of all, the Low Pass Filter roll-off before the op-amp is f = 1/(2*pi*0.22u*1k) ~ 720 Hz. Calculating the op-amp's bandwidth roll-off frequency is easy, too. The 1k resistor from the output and the .22u cap from inverting to ground, form a LPF, so f = 1/(2*pi*0.22u*1k) ~ 720 Hz. Note that everything below this frequency is fed to the inverting input of the op-amp, thus REDUCING those frequencies. Thus, the op-amp boosts frequencies above 720 Hz.
With the LPF and the highs-boosting op-amp the sum of their frequency response should be flat, since both their roll-off is 720 Hz.... and it is! The tone control sort of smoothly morphs between those response curves.
All the 220ohm resistor does in my mind is "prevent" the frequency response from becoming completely flat. As you can see in the graphs, with the 220ohm resistor shorted the graphs look the same at first, but it's as if the range of the tone control is extended in the high frequency range.
So, in light of this realization and my graphic analysis I'd say Projectile is correct and the Geofex article is wrong. Good job pointing that out!
EDIT: Some mild spell checking after proofreading.
interesting.
separating full and off was the same method used by RG and AMZ to explain this. Rereading RG's article though, I don't believe it to be wrong in anyway, just over-simplified for our man Projectile's purpose. I can see how his descriptions fit into your plots of the tone curve, (great job btw).
I'm going to do some calculations today.
I believe I can add some more justification to your 220 resistor argument. I have a feeling that the 220 resistor prevents the 1k + .22uF filter from having any effect until you are over the 3.2kHz cut-off (of 220/.22uF). In a nutshell, you get the roll-off curve/slope of the 1k/.22uF, but only in frequencies above 3.2kHz. This is because below 3.2kHz, that leg looks like an open circuit. I will flesh this out as much as possible so please don't bash me before I can determine if I'm wrong or not. This would explain a lot about how series Caps/Resistors to ground act differently from regular RC filters, and hopefully clear up a lot of confusion in this thread.
Yes, it may be that RG's article just looks at this from a different viewpoint. The two filters interact a lot. I'm interested to see what you'll come up, with, aziltz!
Quote from: MohiZ on April 14, 2009, 09:49:25 AM
Yes, it may be that RG's article just looks at this from a different viewpoint. The two filters interact a lot. I'm interested to see what you'll come up, with, aziltz!
instead of paying attention in stat-mech, i did a quick calculation of the gain as a function of frequency of a Low Pass Filter that includes a 2nd resistor in series with the Capacitor to ground. The results are interesting. If ya'll can give me til Thursday after my E&M test I'd be happy to post my math and hopefully a plot of the gain curve.
EDIT: I don't want to jump the gun until I'm sure I did it right and get some plots going.
ya'll don't mind hand-written math in a pdf?
Quote from: Projectile on April 13, 2009, 07:35:26 PM
Once again, that doesn't make any sense. You people are driving me freaking crazy!
Fair enough I'm fully prepared to accept that I'm wrong.
It could just be coincidence, or a quirk of the simulator, that where the signal is 3dB down compared to the max gain corresponds to 720Hz.
The above is a genuine statement by the way, not some sort of snide remark. just in case anyone reads it as one :)
In all seriousness if you've got issues with R.G.'s article why not PM him and put your points to him?
Quote from: Projectile on April 13, 2009, 07:35:26 PM
In order have a filter of any kind, you have to create some sort of voltage divider. Like this:
(http://img518.imageshack.us/img518/334/rc2a.jpg)
ok thats just wrong.
read up on RC filters and you'll find all kinds of variations, parallel arrangements, etc. The thing is, R1/C1 has a certain property at its cut-off frequency, no it doesn't act like a filter on its own, but place it in the feedback loop of an op-amp and it controls the bandwidth of the gain.
Quote from: aziltz on April 14, 2009, 06:46:38 PM
Quote from: Projectile on April 13, 2009, 07:35:26 PM
In order have a filter of any kind, you have to create some sort of voltage divider. Like this:
(http://img518.imageshack.us/img518/334/rc2a.jpg)
ok thats just wrong.
read up on RC filters and you'll find all kinds of variations, parallel arrangements, etc. The thing is, R1/C1 has a certain property at its cut-off frequency, no it doesn't act like a filter on its own, but place it in the feedback loop of an op-amp and it controls the bandwidth of the gain.
Uh... No.
I'd be entirely willing to back down and eat everything I've said, but at this point I haven't actually seen any information that would contradict my understanding of how these filters work. I have done a lot of reading and breadboarding since the beginning of this post, and so far everything that I have learned has done nothing but strengthen my position, and it has shown to me very clearly that you don't really know what you are talking about. Sorry if you take that personally, but I have a hard time seeing it any other way, and believe me, I've tried! You seem to be very good at math and juggling variables. That's great! I can't do that stuff nearly as well as you can, so my props to you. But you seem to have a problem with understanding how this stuff works in context and it is causing you to make obvious errors in the way you are applying your formulas. You don't seem to actually understand why they work.
I'm sure you could get very creative with resistors and capacitors in a feedback loop to create all kinds of interesting filter arrangements. The fact remains that you need some type of voltage divider or you aren't going to get a different frequency response at the output. Sure, without R2 the arrangement may still be filtering certain frequencies and have a rolloff that you could calculate, but it's like a tree falling in the woods with nobody around to hear it. So, it would be a pretty pointless exercise.
My point was not that the 2nd example is the ONLY way to make a filter. My point was that the first example isn't a filter at all. I don't care whether you put it in the feedback network of an opamp or not, it's not going to DO anything. The GeoFX and AMZ articles seem to be claiming that C1 and R1 in the feedback loop of an opamp make a filter. They don't. Can you see why the first example above is not a filter when simply placed in a feedback loop? Opamps have a very low output impedance, so in order to make a filter that works like the one in the example you MUST have R2, and R2 is what get plugged into the formula to calculate the rolloff. It is essential.
I created this thread to clear up my understanding of how RC filters work. Because my own understanding was directly conflicting with the GeoFX and AMZ articles, I initially thought that there must be something I was missing in my general analysis of these filters. The process of discussing and analyzing these filters on in this thread have now cleared up any contradictions I was having with the GeoFX and AMZ articles. I am now quite confident that those articles are in fact incorrect. It doesn't mean that I don't still have questions, but there is no misunderstanding about the GeoFX and AMZ articles any longer. They're just wrong.
Anyway, thank you to everyone who has participated in this thread. Although it was quite frustrating, this thread has been an enormous learning experience for me. What confuses me now is the total lack of interest in the fact that there are obvious errors in well circulated documents, and why nobody has caught them before. I know I'm not the only person who understands this stuff, so why is there such dead silence except for the small handful of people who have particpated in this thread? Hello? Anyone out there? ...weird.
Quote from: Projectile on April 14, 2009, 10:35:15 PM
My point was not that the 2nd example is the ONLY way to make a filter.
BUT that's what you said, friend... I was only reacting to that statement.
Quote from: Projectile on April 14, 2009, 10:35:15 PM
In order have a filter of any kind, you have to create some sort of voltage divider. Like this:
(http://img518.imageshack.us/img518/334/rc2a.jpg)
I don't think anyone is saying that R1/C1 make a filter on their own. HOWEVER! could it be that R1/C1 have an effect on the R1/R2/C1 filter, and that effect is INDEPENDENT of R2? Possibly? Would you sleep at night? I know I would! LETS LEARN!
Here's some data, I finally figured out how to get 5Spice to work.
Low Pass, R=1k, C=.22uF, like our Mid-Happy Friend the Tube Screamer. The cursor is centered about -3dB roll-off.
(http://i132.photobucket.com/albums/q18/aziltz/AC-3.jpg)
Here's a simulation of Projectile's expertly drawn low-pass/hi-shelf filter (foreshadowing?), using the values, R2=1k, C1 = .22uF and R1 = 220. It seems to level off at the value of 3.2kHz, as calculated by R1 and C1. Note: My definition of R2/R1 is reversed, and I forgot to center the cursor on -3dB.
(http://i132.photobucket.com/albums/q18/aziltz/AC-1.jpg)
Putting them together, again, cursor is arbitrary, and my schematic definition of R2/R1 is reversed.
(http://i132.photobucket.com/albums/q18/aziltz/AC-2.jpg)
note: i've used R2=1 ohm as zero, in order to general two plots together. There's no difference in the plots between r2=1 ohm and no resistor within this range.
As you can see, the addition of R1 after C levels off the filter, creating a "shelf" in the frequency response. The overall roll-off of the filter changes slightly, but not enough to warrant shouting BLASPHEMY at those who would explain this using 720Hz and 3.2kHz as "special" frequencies to talk about, more on that in a bit.
Oh, and just for gits and shiggles, here's what happens when we change the upper resistor. Two curves, R=1K and R=10K.
Note: Cursor is Arbitrary, and my schematic labeling of R1/R2 is reverse from the schematic at the top of the page.
(http://i132.photobucket.com/albums/q18/aziltz/AC-4.jpg)
The thing to note here is, the upper leg resistance changes the roll-off frequency, but the point at which things flatten out remains the same, (set by R=220). To me this says they are independent enough that we can talk about the 220/.22uF on its own. To clarify, in my professional DIY opinion (joke), it is not incorrect to talk about the 220/.22uF RC constant separately from the 1k/.22uF as they are more or less independent in this configuration. HIGH FIVE!
Next I intend to simulate how this affects a feedback loop in a gain stage, as well as the complete clip stage and tone stage of the TS, but tonight I've got to study for an Electricity and Magnetism Mid-Term!
Oh, and I'll probably start my own thread, or even post on my website.
Cheers.
If anyone is interested in 5Spice, it seems a bit easier to use than LTSpice, at least for first time users, I can provide some links. And its FREE. I can also point to a tutorial, with example for simulations. It's from the class I teach, associated with my research advisor so i'd rather not post publicly but I will respond to any PM's with the information.
Now THAT makes sense!!!!!!!!!!
Finally!!!!!!!!!
Thank you! Thank you! Thank you Aziltz!
That is where the 3.2kHz value comes from! It turns out the resistor to ground combination creates another apex of the filter curve where the rolloff "bottoms out". When you put this arrangement in a feedback network, that lower apex becomes the point where the boost becomes level and bass starts rolling off. You can in fact look at it independently. So, You were right about that Aziltz! It's just that that slope of the curve depends upon the relationship of r1 and r2, so if we are going to calculate the technical rolloff point, where the signal is -3db, you would have to look at the relationship of r1 and r2. If you get rid of R2 altogether, then the slope becomes zero and the filter disappears, which is why I was saying that you can't have a filter without R2. Interesting... I'm just thinking aloud here.
So, I will eat my words, the AMZ article is basically correct! ...I say "basically" because I still think it's misleading and I don't think the low E string is actually down -20db, since that would depend on the gain pot and the 51K resistor, which set the slope of the curve by defining the other apex... But I'm not going to nitpick. It's close enough in a general sense, so I'll stand corrected.
The GeofX article however is still appears to be wrong, since it claims very clearly that the signal gets a +6db/octave boost above 3.2KHz. This is not true, since 3.2KHz is actually the point where the boost levels off. Everything above 3.2KHz actually gets the same amount of boost, and since the first filter is rolling off everything above 720Hz at a ratio of -6db per octave, the treble is never fully recovered above 3.2Khz
That matches what I am seeing exactly. It all makes perfect sense now. Awesome! Thank you Aziltz! You are the man!
Quote from: Projectile on April 15, 2009, 12:38:35 AM
Now THAT makes sense!!!!!!!!!!
Finally!!!!!!!!!
Thank you! Thank you! Thank you Aziltz!
That is where the 3.2kHz value comes from! It turns out the resistor to ground combination creates another apex of the filter curve where the rolloff "bottoms out". When you put this arrangement in a feedback network, that lower apex becomes the point where the boost becomes level and bass starts rolling off. You can in fact look at it independently. So, You were right about that Aziltz! It's just that that slope of the curve depends upon the relationship of r1 and r2, so if we are going to calculate the technical rolloff point, where the signal is -3db, you would have to look at the relationship of r1 and r2. If you get rid of R2 altogether, then the slope becomes zero and the filter disappears, which is why I was saying that you can't have a filter without R2. Interesting... I'm just thinking aloud here.
So, I will eat my words, the AMZ article is basically correct! ...I say "basically" because I still think it's misleading and I don't think the low E string is actually down -20db, since that would depend on the gain pot and the 51K resistor, which set the slope of the curve by defining the other apex... But I'm not going to nitpick. It's close enough in a general sense, so I'll stand corrected.
The GeofX article however is still appears to be wrong, since it claims very clearly that the signal gets a +6db/octave boost above 3.2KHz. This is not true, since 3.2KHz is actually the point where the boost levels off. Everything above 3.2KHz actually gets the same amount of boost, and since the first filter is rolling off everything above 720Hz at a ratio of -6db per octave, the treble is never fully recovered above 3.2Khz
That matches what I am seeing exactly. It all makes perfect sense now. Awesome! Thank you Aziltz! You are the man!
honestly i feel like we've had the same thing in our heads, and it just doesn't translate into words very well. I wish I had this simulation tool when i first tried to answer your question, we might have avoided some frustration. This all took about 20 mins to construct and run. 5Spice really is a nice, simple simulation too.
Quote from: aziltz on April 15, 2009, 12:46:14 AM
honestly i feel like we've had the same thing in our heads, and it just doesn't translate into words very well. I wish I had this simulation tool when i first tried to answer your question, we might have avoided some frustration. This all took about 20 mins to construct and run. 5Spice really is a nice, simple simulation too.
Same here.
I wonder how one would actually calculate the "technically correct" -3db rolloff point of this curve, since in this arrangement it's not your typical -6db per octave filter anymore. That's where the AMZ article gets into trouble, although I guess it's close enough for DIY discussion.
I tried messing with LTSpice yesterday, but I wasn't confident enough that I knew what I was doing to post the results. It must have taken me 2 hours just to figure out how to get it to output the plot of a simple RC filter, and then I had to attend to other business. The graphs from 5spice look much nicer though!
Now my next question would be, why does the R2 resistor (using aziltz model) control the point where the filter levels off? And why would that "level off" point be calculated using the same general formula that calculates the roll off point of the R1/C1 combination? Hmmmmm...
Duplicate post. Deleted.
Wow! This explains everything. Very good results, Aziltz! In a way, we were just all focusing on different points of the response curve.
QuoteThe GeofX article however is still appears to be wrong, since it claims very clearly that the signal gets a +6db/octave boost above 3.2KHz. This is not true, since 3.2KHz is actually the point where the boost levels off. Everything above 3.2KHz actually gets the same amount of boost, and since the first filter is rolling off everything above 720Hz at a ratio of -6db per octave, the treble is never fully recovered above 3.2Khz
Actually, I checked the GEOFEX article, and it says.. quote..
QuoteNote that the "boost" actually just levels off the -6db/octave induced by the 1K/0.22uF network ahead of the active control stage, so the treble is just not being cut any more above the turnover frequency for the tone control stage when fully at "treble".
I think what RG is trying to say here is the same what we have established as of now. However, I think this still can be misleading.
I think the explanation to why the 220 ohm resistor sets the "level-off" frequency is found from the GEOFEX article: "As a series network, at frequencies above the point at where the capacitive impedance is less than 220 ohms (which happens at about 3.2KHz), the network just looks like the 220 ohm resistor." So the high frequency response stops dropping because the combined impedance of the cap and the resistor can't be less than 220 ohms. That means the minimum amplification factor can't be less than 1k/(1k+220R) ~ 0.18 no matter how high the frequency. An amplification factor of 0.18 equals about -15 dB, which seems to be the value that the response levels off to, looking at Aziltz's graphs!
I agree that it's pretty interesting that the way this "level-off" frequency can be calculated with the same formula as a normal single-pole filter. I wonder what's the math behind it. Aziltz, I think it's a good idea to start a new topic about it with a different title, this one is getting a little long.
I'm gonna have to answer my own post on this. I'm referring to this
QuoteNow my next question would be, why does the R2 resistor (using aziltz model) control the point where the filter levels off? And why would that "level off" point be calculated using the same general formula that calculates the roll off point of the R1/C1 combination? Hmmmmm...
I figured this out. Let R1 be the upper resistor 1k, R2 be the lower resistor 220R and C be the cap 0.22uF.
First think about the filter simplified as a normal RC filter: its amplification is calculated as the voltage of a voltage divider, the cap's impedance divided by the sum of the impedances of the cap and resistor. At low frequencies the cap's impedance is large compared to R2, so we can short R2. At high frequencies the cap's impedance is small so we can short the cap.
(http://img4.imageshack.us/img4/5827/74286824.jpg)
Now we calculate the amplification factors of both of these simplifications.
(http://img185.imageshack.us/img185/3483/48996079.jpg)
Next we calculate the frequency where the amplification of these simplified models is equal, that is, when the frequency response levels off.
(http://img2.imageshack.us/img2/6019/88893487.jpg)
And the answer is, oddly enough, the same as the roll off point of the R2/C combination would be. Only one question: why this isn't properly explained in these articles? I hope you can make out the hand-written equations :P
In conclusion here are the formulas for this type of filters, with the example values of our beloved TS circuit ;):
1) Initial -3dB roll-off: f = 1/(2*pi*R1*C) ~ 720 Hz
2) Level-off frequency: f = 1/(2*pi*R2*C) ~ 3.3 kHz
3) Level-off amplification: R1/(R1+R2). In decibels this is 20*log [the answer] ~ -15 dB.
So at first there's no attenuation, then at 720 Hz the volume starts to drop, and at 3.3 kHz the volume has dropped to -15 dB and ceases to drop, staying at that level even at higher frequencies. All of this is backed by the graphs of aziltz, so the approximation works extremely well!
Quote from: Projectile on April 13, 2009, 07:35:26 PM
Look, you can't build an RC filter like this:
(http://img13.imageshack.us/img13/1083/rc1l.jpg)
I know I'm splitting hairs here, but that's incorrect. Real-life op-amps have low, but non-zero output resistance. So while you're correct that you can't build an RC filter without an "R," you will never find an op-amp without output resistance, so in practice, you
can build a filter like that.
Carry on...
MohiZ,
That's Exactly the kind of stuff I got when doing the math yesterday, i just had trouble making those approximations so I went straight for the simulations, but I got the same curves graphing the complex impedance-calculated gain on my TI89 as i did with the simulation.
So at high frequencies, the lower resistor causes the filter to act like a voltage divider, genius!
QuoteNote that the "boost" actually just levels off the -6db/octave induced by the 1K/0.22uF network ahead of the active control stage, so the treble is just not being cut any more above the turnover frequency for the tone control stage when fully at "treble".
Still seems wrong to me.
The treble IS being cut ABOVE the 3.2KHz. It is compensated for, or rather "brought back up to level", BELOW the 3.2KHZ turnover. The GeoFX article has it backwards.
He did get the first part right though, so he must understand what is going on. It seems like he just made a simple mistake. Anyway, I sent a PM to R.G. about this, so hopefully he can clear this up himself.
Quote from: MohiZ on April 15, 2009, 07:18:41 AM
I agree that it's pretty interesting that the way this "level-off" frequency can be calculated with the same formula as a normal single-pole filter. I wonder what's the math behind it. Aziltz, I think it's a good idea to start a new topic about it with a different title, this one is getting a little long.
That's what's still baffling me.
I noticed that the 220ohm resistor was creating a "foor" that no frequency could fall below, and I could see that it was creating a second apex in the curve, but there was no way in a million years I would have guessed that you could plug the 220 ohm resistor into the same formula to determine that lower apex. It just doesn't make any sense to me, since that resistor is performing an entirely different function in the circuit. Weird.
Actually I may have an explanation, but its a bit wacky, and since I've been having trouble communicating my other points in this thread, I don't know even I should even try. I'm probably digging a hole with this one. My mathematics terms are a bit sketchy at this point in my life, but here goes:
EDIT: Mohiz came to a similar conclusion while I was typing this, but I'm going to post my analysis also since I had already finished typing it before I saw his...Let's call the 1K resistor "R1" and the 220 ohm resistor "R2" and the capacitor "C1", like in Azilt's graph.
Ignore R2 for a moment.
R1 and C1 form voltage divider that varies with frequency according to the value of C1. Together, they can be used to calculate the rolloff point off the filter. They are inversely related, in that, if you double R1 you must half C1 in order to get the same rolloff point, and vice versa.
Now, let's ignore C1 for a moment and look at the relationship between the two resistors. R1 and R2 form a voltage divider with a set value. They can be used to calculate the level which the signal cannot fall below. C1 only controls frequencies above this "floor" by adding to the bottom leg of the voltage divider that is initially set by R2. R1 and R2 are directly related, in that if you double R1 you must double R2 in order to get the same frequency "floor", and vice versa.
Where the rolloff curve set by R1 and C1 meets the "floor", created by R1 and R2, is a second apex, or the "leveling off" point. I imagine that there is some formula you could use that involves all three variables to calculate that apex, and it would seem like it would be necessary. But, it's not necessary, and here's the trick...
Since R1 and C1 are inversely related in the same way that R1 and R2 are directly related, you can essentially plug R2 and C1 into the same formula that describes the relationship between R1 and C1 to get a answer that describes that lower apex independent of R1. It works because all 3 terms share the same mahtematical relationship in the same voltage divider. This explains why you can manipulate R2 and C1 independently of R1, even though R1 is essential in creating the filter.
Say you half the value of R2. That means that the relationship between R2 and C1 doubles and the frequency of the lower apex increases. But that also means that the value of voltage divider formed by R1 and R2 decreases. So, whatever value you plug in for R1, the lower apex is going to be the same.
It's just confusing as hell because when you are just thinking in terms of electrons in a wire there should be no reason why R2 controls the frequency of anything without R1! You can't even calculate a voltage divider without R1 ...but the math works because the relationship of the terms allows for that simple substitution. R1 and C2 sets the initial rolloff point of the filter and the relationship between R1 and R2 sets the slope of the curve. If you take away R1, there is no curve, and no filter, but that doesn't mean you can't calculate the apex of the lower part of the curve intependant of R1. We just can't talk about it as the "technical" -3db rolloff point anymore, because it's not. It just describes the general location of the apex.
At least that makes some sense to me.
Quote from: MohiZ on April 15, 2009, 08:08:29 AM
Only one question: why this isn't properly explained in these articles? I hope you can make out the hand-written equations :P
Yes, really!
It's all so clear now. I've seen these particular implementations of filters all over the place, but no adequate explanations. I scoured the internet during the course of this thread and could only find either basic descriptions of RC filters, or complex mathematics that went way over my head. I'm really surprised this has never been discussed here before. It seems pretty important to understanding many of these stompbox circuits. I'm really glad we finally all came to the same conclusions. This stuff is fascinating!
This thread has actually encouraged me to go back to college and take some electronics courses. Maybe I'll stop being a lazy musician and make something of my life after all.
Thanks to everyone who contributed to solving this little puzzle. :)
i guess the thing about calculating a time-constant or a cut-off frequency from R&C , is it all has to do with when the Cap in question has an equal impedance with another element of the nearby circuit, so R1 or R2.
What I'm trying to suggest here is, at one frequency, the impedance of C is equal to R1, and at another frequency its equal to R2, and I believe this works for more complicated filters in general. Thoughts?
Quote from: aziltz on April 15, 2009, 09:56:05 AM
i guess the thing about calculating a time-constant or a cut-off frequency from R&C , is it all has to do with when the Cap in question has an equal impedance with another element of the nearby circuit, so R1 or R2.
What I'm trying to suggest here is, at one frequency, the impedance of C is equal to R1, and at another frequency its equal to R2, and I believe this works for more complicated filters in general. Thoughts?
I follow you on the general idea, but you can you give an example of a more complicated circuit where you think the RC calculation would work like this? I'm just not so sure this little trick substitution that works so perfectly with r2 and c1 would apply in all situations.
Quote from: earthtonesaudio on April 15, 2009, 08:42:00 AM
Quote from: Projectile on April 13, 2009, 07:35:26 PM
Look, you can't build an RC filter like this:
(http://img13.imageshack.us/img13/1083/rc1l.jpg)
I know I'm splitting hairs here, but that's incorrect. Real-life op-amps have low, but non-zero output resistance. So while you're correct that you can't build an RC filter without an "R," you will never find an op-amp without output resistance, so in practice, you can build a filter like that.
Carry on...
I'm aware of that. Funny, I originally had a tangent where I explained that the output impedance of an opamp could act as R2, but I omitted it before posting because I thought it just muddled the issue and distracted from the general point I was trying to make. The graphic doesn't specify that it is in the feedback loop of an opamp anyway, and when I did describe that circuit in the feedback loop of an opamp I think I mentioned that the opamp had a low output impedance just so that nobody would call me on the fact that an opamp's output in fact has an impedance.
If we are going to really split hairs that much, then I would still be calling the AMZ article incorrect, but I'm just assuming that certain things were omitted for the sake of brevity.
Quote from: Projectile on April 15, 2009, 10:13:56 AM
I follow you on the general idea, but you can you give an example of a more complicated circuit where you think the RC calculation would work like this? I'm just not so sure this little trick substitution that works so perfectly with r2 and c1 would apply in all situations.
like our example above, i think since 720 and 3.2k are far enough apart, you can describe things separately.
i'm not sure its universal, and I won't make a claim to that. it just appears that at high/low frequency approximations the gain can definitely be simplified, as MohiZ showed. i just wanted to suggest that there's something special happening at each combination of C and Rx (1 or 2). I think the R2/C cut-off (3.2kHz) relates to the High Frequency limit MohiZ took. Can anyone flesh that out?
other filter arrangements to try would be, R1+C as the upper leg and R2 as the lower leg (High Pass + Low Shelf?)
or R1||C as one leg and R2 as the other leg, in either a Low or High Pass arrangement.
If i get some time tonight I'll simulate these.
It's amazing to think that we thought about the same thing at the same time! This is great stuff, I wouldn't have even thought about it without this topic. Now we can create all kinds of interesting filters instead of just the usual low-pass or high-pass.
The key definitely lies in simplifyin these things. Think about how much we had to simplify this tone stage for instance; first we took apart the tone control's relation to the op-amp and even after that the tone control itself had to be broken to pieces before we gained full understanding about this. I agree that the approximations work so well because the two corner frequencies are far enough apart.. Otherwise there would be more cross-talk. Aziltz's view that something interesting happens at the points when the impedance of the cap equals one of the resistors' impedances makes sense. The cap's impedance is the only one that changes due to frequency, and it changes in a non-linear way too.
Nevermind...
Cool stuff, I'm glad this thread turned out well :)
I've seen that filter arrangement a lot but never understood what it did until now.
I think this also explains why the TS has the funky S taper pot for the tone control. Like I said earlier the portion of the tone pot between the 220R resistor and the negative input of the opamp comes in to play, just not in the way I thought it did. Basically it increases the size of R1 from Projectile's drawing. If you look at the response when the tone pot is set in the middle so R1 = 10K + 220R the graph is basically flat. Transferring this to the TS the opamp is essentially just a buffer at this point, it's not boosting anything. You don't get any real change in the response until you reduce R1 to about 2K, so if you used a linear pot the whole central portion would be useless.
hey guys, if you're interested, i started a new thread with some simulation data for the TS as a whole, comparing the clipping amp output to the tone control and so on.
http://www.diystompboxes.com/smfforum/index.php?topic=75775.0
This is an awesome thread! Many thanks! I joined just so I could pursue an issue with something very similar that is not really addressed in plain terms anywhere that I can find.
Right now I am working on the guitar electronics only, and want to experiment with some on-board circuits. Maybe a box itself is in my future, but right now I am just trying to learn the intricacies.
I believe this is ultimately related to my quest to find a generalized way to map corner frequencies of the guitar's tone circuit. For simplicity's sake, this is in general what I am working with:
(http://i593.photobucket.com/albums/tt16/Caspar98/simplecircuit.jpg)
All I would like to do is figure out how to determine the -3db corner frequency on paper. Approximations are fine. I thought I knew what i was doing and then realized it did not work at all like I assumed it should. The corner frequency is inversely proportional to the value of C1, but seems proportional to the value of R2.
Since this seems to be very similar to the paradox that Projectile had questioned in this thread, can anyone give me insight on how I go about calculating The corner frequency of the circuit? I know it has to rely somehow on the rest of the circuit interacting, I just can't grasp how.
Quote from: ScottB on August 13, 2009, 05:25:03 PM
(http://i593.photobucket.com/albums/tt16/Caspar98/simplecircuit.jpg)
R2 can be thought of as a 'shelving resistor'. It limits how much the circuit can attenuate the higher frequencies. You can think of the combination of R2 and C1 as a signal capacitor with an adjustable 'quality'. So your 3dB point is always the same and is a function of R1 (the series component) and C1 (the shunt component). The maximum amount of attenuation is a function of R1, R2, and C1 (near the 3dB point). As you go up in frequency it reduces to just the ratio of R1 and R2.
--john
Thank you John, but I am still confused so let me break this down into digestible chunks.
chunk 1
If I understand you, and assuming we turn R2 completely off and remove it from the circuit, the Fc will follow the traditional rule of 1/(2*Pi*R1*C1). Is that correct?
chunk 2
Now, in the case of the value of R1 you mentioned the series component, did you mean only the top "half" of R1 depending on where the wiper is? Or did you mean the entire value of R1 top to bottom?
chunk 3
Does the Impedence of L1 (the guitar pickup) have any real effect? I assume not if the answer to the above is to only use the series portion (or top half) of R1.
chunk 4
Regardless of the above answers, I am still not clear in how R2 limits attenuation.
Thank you for the response
Quote from: ScottB on August 13, 2009, 09:23:49 PM
Thank you John, but I am still confused so let me break this down into digestible chunks.
chunk 1
If I understand you, and assuming we turn R2 completely off and remove it from the circuit, the Fc will follow the traditional rule of 1/(2*Pi*R1*C1). Is that correct?
chunk 2
Now, in the case of the value of R1 you mentioned the series component, did you mean only the top "half" of R1 depending on where the wiper is? Or did you mean the entire value of R1 top to bottom?
chunk 3
Does the Impedence of L1 (the guitar pickup) have any real effect? I assume not if the answer to the above is to only use the series portion (or top half) of R1.
chunk 4
Regardless of the above answers, I am still not clear in how R2 limits attenuation.
Thank you for the response
Chunk 1. Yes
Chunk 2. Top half, depending on where the wiper is.
Chunk 3. Yes, it will. But will depending greatly on the pickup characteristics. For basic understanding of the circuit, let's just assume it''s a perfect voltage source.
Chunk 4. The capacitor will (idealy) reduce to zero impedance with high frequency. If R2 is zero then all the signal will be shorted to ground. With any non-zero setting of R2, there is a fix amount of resistance in that path so therefore it can not attenuate any further.
Oh! Now I see! that turns out to be much simpler than I was making it out to be.
Thanks John, this is plenty to think about now. I'm not sure how exactly to calculate the limit of the attenuation due to the shelving resistor but I do see how that works. It clears up some major confusion I had.
Quote from: ScottB on August 14, 2009, 01:04:30 AM
Oh! Now I see! that turns out to be much simpler than I was making it out to be.
Thanks John, this is plenty to think about now. I'm not sure how exactly to calculate the limit of the attenuation due to the shelving resistor but I do see how that works. It clears up some major confusion I had.
For the limit, assume the capacitor is a short and then you have a simple voltage divider (ignoring other factors like source and load impedance). So when R2 is equal to the setting of R1 (between the wiper and top) the attenuation is 1/2.
This is excellent stuff. I've scoured the web and asked in various forums, it turns out I should have just come here first. Thanks!
When the shelving resistor ratio (thanks for that, it opened my search up) limits the attenuation aka a simple voltage divider, obviously it has no effect on lows because the cap looks like an open to them. The very highs obviously work the opposite, so at 1/2 ratio half of the highs are missing the cap completely.
Now, thinking of a simple bode plot, assuming an ideal class 1 on paper, where does this ratio start its effect? Is it dependent on the corner frequency and simply reduces the slope (from -6db/octave to say -3db/octave) or is it independent and only related to the cap properties itself (basically leveling off the slope at some higher frequency, creating a sort of S curve on the bode plot)?
Quote from: ScottB on August 14, 2009, 11:51:39 AM
This is excellent stuff. I've scoured the web and asked in various forums, it turns out I should have just come here first. Thanks!
When the shelving resistor ratio (thanks for that, it opened my search up) limits the attenuation aka a simple voltage divider, obviously it has no effect on lows because the cap looks like an open to them. The very highs obviously work the opposite, so at 1/2 ratio half of the highs are missing the cap completely.
Now, thinking of a simple bode plot, assuming an ideal class 1 on paper, where does this ratio start its effect? Is it dependent on the corner frequency and simply reduces the slope (from -6db/octave to say -3db/octave) or is it independent and only related to the cap properties itself (basically leveling off the slope at some higher frequency, creating a sort of S curve on the bode plot)?
Just think of the capacitor as a frequency-dependent resistor. Add it's reactance to R2 and compute the voltage divider. It sounds like you have the correct idea (S-curve).
Here's the TS tone control section, including the 1k/.22uF cap response:
(http://www.greene-pedals.com/GEAD/The%20Technical%20Page_files/tnctl2.jpg)
Here's what happens when you put a 100 ohm resistor in series with the .22uF cap. You can see the 'shelving'.
(http://www.greene-pedals.com/GEAD/The%20Technical%20Page_files/newtone.jpg)
That is exactly what I envisioned, excellent!
Is there any general formula for the "shelf"? What I mean is there any rough guide to help envision what frequency the shelf "corners" in a similar way that you can roughly calculate the initial corner frequency of the filter circuit?
And in my simplified diagram, if R1 is turned all the way up, ie: having no series component, the tone control still works. Is that because the impedence of the pickup itself is still present and actively determining a filter circuit corner frequency (for a humbucker that would be roughly 10kohms on average I guess).
Quote from: ScottB on August 14, 2009, 05:23:18 PM
That is exactly what I envisioned, excellent!
Is there any general formula for the "shelf"? What I mean is there any rough guide to help envision what frequency the shelf "corners" in a similar way that you can roughly calculate the initial corner frequency of the filter circuit?
And in my simplified diagram, if R1 is turned all the way up, ie: having no series component, the tone control still works. Is that because the impedence of the pickup itself is still present and actively determining a filter circuit corner frequency (for a humbucker that would be roughly 10kohms on average I guess).
Yes, now you start getting into the second order effects. If the pickup was a perfect source, the cap would have no effect.
Man this is so cool!
Iafter more research, especially here, I decided I needed to bite the bullet and learn one of these tools. I was a little intimidated at first but I went ahead and ripped LTSPICE, played with it a little while and before long I was graphing frequency response curves all over the place!
I highly recommend this to anyone who seems to need some other way to understand what is going on. I now see the relationship between the shelf resistor and the tank resistor. Different diagrams behave slightly different or almost not at all, but in each case I couldn't guesstimate what the differences would be. Now I see it visually in a graph.
I'm not sure how to model the pickup exactly; I used a generic AC source and a 1M load to model the marshall amp. I tried a parallel resistor load to simulate the pickup impedence but it seems to make little difference if any regarding the tank resonance. Doesn't matter too much, the concepts are now much clearer. There are a few things I question but that may be the limitations of the simulator, or it may be the limitations of my model.
Thanks!
It is cool.
I enrolled in college for an electronic engineering tech degree, all because of this forum. It starts in a few weeks. I'm excited!
Quote from: Projectile on September 07, 2009, 02:46:49 AM
It is cool.
I enrolled in college for an electronic engineering tech degree, all because of this forum. It starts in a few weeks. I'm excited!
sounds like a good investment, I hope you enjoy it!