Hello,
After I've tested a lot of transistors for noise, i liked very much a BC337.
My only problem is now that i have a strong ticking (that i had anyway with any transistor that i tested...and, what's important, i used ONLY alkaline batteries):
http://www.aronnelson.com/DIYFiles/up/fsh.mp3 (http://www.aronnelson.com/DIYFiles/up/fsh.mp3) (i used a rocktron rampage as distortion)
What can i do?
I searched the forum, but i don't find anything very clear...
Thx a lot!
I believe ticking is a common problem associated with this circuit.
Thx eggman6...i hope that i will resolve it because i like it a lot!
???
???
The ticking is LFO leakage into the audio path. You could try getting rid of it by using a capacitor/resistor pair configured as a highpass filter. Try a 0.01 or 0.027 at the end of the audio path with a 100k resistor to ground immediately after it.
Like this: (ignore the square in/pulse out bit)
(http://i210.photobucket.com/albums/bb292/frequencycentral/squarepulse.jpg)
Basically it should defeat the LFO by not allowing frequencies through below the HPF cutoff.
EDIT: I had similar ticking in a tube tremolo I designed ("Bipolar Disorder"), and this is how I solved it.
Thx a lot frequencycentral!
I have two questions regarding the filter:
1) the 10nF capacitor must to be electrolytical or not?
2) where i must to put the filter: before the output jack (easy) or after the LFO (very hard...i have no ideea where is that...)?
Thx a lot again!
Ok just looking at the schematic here: http://www.freeinfosociety.com/electronics/schemview.php?id=1125
C4 and R14 are actually configured as a HPF. You should try increasing the value of C4, maybe just add another 0.15 in parallel. No it doesnt need to be polarised. You need to strike a balance between stopping the ticking at the LFOs highest speed and cutting the bottom end off your guitar signal.
Also, look at C7 - try anothyer 0.15 in parallel there too.
Get some crocodile clip jumper leads and play about.
I hope this sorts it for you! I haven't built the FSH as I have all that stuff available in my modular synth.
Thx a lot again!
I will try what you said tonight...
Hope it works out for you, just keep upping the cap value until optimum.
did it work?
I have exactly the same problem with my build... would be nice to solve the problem somehow.
Sorry, but i hadn't time to test...i was gone from my home for a couple of days...maybe tomorrow...
I've tested what frequencycentral suggested to me above, but i still have some ticking.
Even i replaced the 100nF capacitor from the output with a smaller one (47nF, 33nF, 22nF, 10nF and...1nF) the ticking is still audible and, of course, i lost almost all of the bottom end :-[
???
that's a pitty. I hope someone can help us soon.
Hi, if the problem is due to LFO ticking "bleeding through" to the audio path, there are several things that you can do.
I found the following tips on this site when I was looking to cure LFO ticking in an UltraFlanger. The following tips were written by various people (Mark Hammer and others) and were used to address deticking LFO's in general but I'll try to rewrite it to make it appropriate for deticking the FSH LFO:
http://www.diystompboxes.com/smfforum/index.php?topic=61250.msg480708#msg480708
There are a few separate matters to be addressed (quotes from others slightly rewritten to adress the FSH LFO):
The LFO and audio path need to be decoupled from each other. You should run something like a 100R resistor from the shared +9v source to the V+ pin of the LFO opamp chips (especially IC5 which generates the LFO), and run a 10-100uf cap to ground from that pin. This will give each chip a small reservoir of current to hold them over in case one of the other chips makes a sudden demand for current. Those sudden demands are what you hear as ticking.
The FSH is using a 741 opamp for the LFO path (IC5) and 3080 OTA's for the (filter) audio path (IC1 IC2 Q1 Q2). This will let you decouple them separately, instead of having whatever the LFO does impact directly on the OTA's in the same hunk of epoxy.
It is also a smart idea to use a low-current op-amp for the LFO. Why? Because the "standard" opamp LFO produces a square wave. What you hear as ticking is the sudden current draw as the square wave rises, placing noise on the power line. The decoupling reduces the extent to which that spike "bleeds" into other parts of the overall circuit. Using a low-current op-amp reduces the magnitude of that spike in the first place. Think of the decoupling as being like ear muffs, and the low-current op-amp as being like smaller speakers. Use of either will keep the volume of undesirable sounds down, but use of both will reduce it even further.
You will note in glancing through schematics of commercial phaser, flanger, chorus, and tremolo pedals that many of them will use either an LM358 or a TL022 dual op-amp for the LFO, and use different op-amps elsewhere in the audio path. They use those particular op-amps because they draw less current. There are other functional equivalents, I'm sure. I just wanted to explain why you often see those ones.
Another strategy for deticking involves a slight change to the LFO design itself. Nicholas Boscorelli explains this nicely in one of his Stompboxology ( http://moosapotamus.net/IDEAS/stompboxology/stompboxology.html ) newsletters, and you can sometimes see this in commercial designs. The gist is to change that initial square wave into what is more or less a trapezoidal wave. What we hear as ticking depends on the suddenness of the rise-time. If the rise-time can be slowed down even just a wee bit, the current draw of the LFO is essentially distributed over a long enough period of time (in electronic, not human, terms) that it does not produce disturbances elsewhere in the circuit.
So, reduce the amount of current required to produce a square wave, make the rise-time of the square wave less sudden, and isolate the chip requiring the sudden current draw from the rest of the circuit. That's how you get a tick-free modulation pedal.
"In my Small Stone (russian) I have a 1uF capacitor between + and - input of the comparator opamp which is the one that produces the square wave of the LFO. This seems to detick the Small Stone, probably, by reducing the rise time of the square wave."
QuoteIn my Small Stone (russian) I have a 1uF capacitor between + and - input of the comparator opamp which is the one that produces the square wave of the LFO. This seems to detick the Small Stone, probably, by reducing the rise time of the square wave
Clever- Is this also an issue with the DIY (Tonepad) Small Stone version?
Sorry, I do not know, I have only a "Russian" one.
WOOOOOOOW...thx a lot flo for your advices...i will try them as soon as possible...
let me know if that works. i am now noticing a tick too.
it's not that bad, if there is an easy fix though, i'd like to try it out.
thanks again
> "thx a lot flo for your advices"
You're welkom but I can't take the credit for the wisdom of it. They are just advices from others on this forum (like Mark Hammer: Thanks!) that I collected. I only adjusted them a bit for you to "fit" the FSH. Try a search on "ticking LFO" / "deticking LFO" and you'll find the same advice... Good luck fixing it!
I've decoupled IC5 using a 100 ohms resistor and a 47uF capacitor, but no noticeable effect :(
I will try to got some TL022, but it's almost mipossible here...
Perhaps you can try a TL061 low power single opamp (although it has a high slew-rate). It has the same pinout as the 741.
http://www.datasheetcatalog.org/datasheet/SGSThomsonMicroelectronics/mXvwry.pdf
1 - Offset null 1
2 - Inverting input
3 - Non-inverting input
4 - VCC-
5 - Offset null 2
6 - Output
7 - VCC+
8 - N.C.
The +8V -8V power-supply of the "audio path", Q1 Q2 IC1 IC2, is already decoupled from the rest via R35 R36 C12 C14.
Just to check, you did try to decouple the "rest" by putting a 100ohm resistor, together with a 10uF to GND, between:
- C11 and anything that wants +9V
- C13 and anything that wants -9V
Like this:
+9V / -9V
|
>
> 100 ohm
>
|
| 10uF
|------||-------- GND
|
|
opamp pin7 / pin4
Just for "fun", although I do not have high hopes for this one, try to put a 1uF capacitor (not an elco) between:
- The + and - inputs of the 741 opamp
- The output of the opamp and GND.
Just try this by touching the opamp pins with the capacitor's legs while the circuit is on.
Also:
The opamp IC5 is used as a "relaxation oscillator" with an uneven duty-cycle so it outputs a pulse for the S&H circuit. You can try to reduce the pulse "slope" a bit by putting a 0.1nF up to 100nF capacitor between the gate of Q5 and GND or -9V. The problem is, the Sampe&Hold circuit wants a "sharp" pulse to sample the noise adequately so you can not make that C too large. Find a compromise.
http://en.wikipedia.org/wiki/Relaxation_oscillator
I one of these does not work on its own, try them all together!
Wooooow again...you're a very kind man...thx a lot again!!!
In the layout that i used (http://www.tonepad.com/getFile.asp?id=96 (http://www.tonepad.com/getFile.asp?id=96)) i replaced the jumper from the right upper side with a 100ohms resistor and i placed after it the 47uF capacitor...it's ok?
I was referring to the single 741 opamp of the LFO (IC5) in:
http://www.freeinfosociety.com/electronics/schemview.php?id=1125
Sorry for the confusion, I should have been referencing LFO IC3b of Tonepad's version.
So in my previous posts: 741 and IC5 are the same a Tonepad's IC3b. Well, you'll figure it out...
The LFO opamp IC3b is part of a DUAL opamp. So a low power LM358, TL022, TL062 DUAL op-amp for IC3 will be fine NOT a 741 or TL061 SINGLE opamp!
> "i replaced the jumper from the right upper side with a 100ohms resistor and i placed after it the 47uF capacitor...it's ok?"
That's the red trace going to pin8 of IC3b, right? If the capacitor was placed:
- One leg on the 100 ohm resistor connected to IC3b pin8
and
- The other leg on GND,
then its ok.
Try the same with the yellow trace that goes to IC3b pin 4 to decouple the -9V of the LFO.
Thx again...there is any trick about how to set the 100K trimpot?
It must to be set to 0V on the "wiper"?
I have no idea about the trimpot, sorry...
Quote from: gigimarga on August 03, 2008, 04:29:56 PM
Thx again...there is any trick about how to set the 100K trimpot?
It must to be set to 0V on the "wiper"?
Connect a meter set to 'voltage' to the wiper, trim the wiper until the meter reads 0 volts.
Thx a lot!
Oh wow, Thanks very much for your help FLO.
tried the LM358, but that didn't solve the ticking... I will work on it, when I find the time.
Quote from: frequencycentral on August 03, 2008, 05:47:12 PM
Quote from: gigimarga on August 03, 2008, 04:29:56 PM
Thx again...there is any trick about how to set the 100K trimpot?
It must to be set to 0V on the "wiper"?
Connect a meter set to 'voltage' to the wiper, trim the wiper until the meter reads 0 volts.
I did it, but after i've set the voltage of the wiper to 0V i couldn't find any position of the another trimpot to work... :(
Try exchanging Q5 and Q6... I have had this problem and after changing the transistors it worked again. so maybe one or both of them are broken.
Thx returnable...my FSH works, but bot if i set the wiper of the 100K trimpot to 0V...that was your problem too?
yeah, I was turning the trimpots without the Sample&Hold effect becoming audible, no matter which setting. After exchanging the Transistors everythings fine. I set the 100K Trimpot to 0V and the other one was fairly easy to set up.
Did you manage to solve the ticking problem yet? I haven't had the time for experiments yet.
Thx a lot returnable...i will replace Q5 and Q6 as soon as i will have some time for that.
I've replaced IC3 with a TL062 and it seems that the ticking was serious lowered (thx a lot flo for the tip!), but i did it at 2AM and i coudn't test it at a fairy high volume.
I replaced IC3 with a TL022 and that lowered the ticking, so that it's okay at lower volumes, but at higher volumes, it's still audible.
Quote from: gigimarga on August 11, 2008, 02:46:23 AM
Thx a lot returnable...i will replace Q5 and Q6 as soon as i will have some time for that.
I've replaced IC3 with a TL062 and it seems that the ticking was serious lowered (thx a lot flo for the tip!), but i did it at 2AM and i coudn't test it at a fairy high volume.
OK...thx TL062 solved the problem of ticking for me almost perfect...thx a lot all!!
Cool, then I'll get one as soon as possible too :)
Great research, I will have to try a TL062. That seems like an easy fix to try.
damn, after exchanging IC3 with a TL062, the S/H doesn't really work anymore. The Filter is fine though, only in very high filter sounds you could hear a little ticking...
So now I will have to fix the S/H part. I wonder if I ever finish this project... :icon_rolleyes:
I just tried a TL062 in my FSH-1 and the ticking is still there. It seems a little more quiet, but it's still unacceptable.
In my FSH, after using the TL062, i still have the ticking, but i think it's at a reasonable level... ???
Okay, now I found the time to work on this project again...
I read in another thread(don't really remember which one it was), that was also about a ticking LFO, that soldering a 100Ohm resistor in series with the power supply could help.
I tried it and it helped! at least a little. the ticking is even less, but still there.
So maybe changing the resitor value to 150 or 200 Ohm would work? I'll try it in the next few days and let you know.
It does tick even with TL062 but acceptable.
That UP/DOWN switch is useless I guess..
Mine ticks too. Even used shielded cables for all the off board wiring. Ugh.
I've gotten the Up/Down switch to work. It takes a lot of playing around with this pedal to get a feel for it. When you switch between the 3 different functions (S+H - Auto Wah Up + Auto Wah Down) the knobs almost always need to be re-tweaked - you can't really leave any of the knobs that the different functions share in the same position (i.e. you can't necessarily click between S+H to Filter Up or even from Filter Up to Filter Down without resetting some things or else you get no or a weak effect).
I would love to see a copy of the manual for the original if anyone has one. I know what attack & decay are but they don't seem to respond like I would expect. I've also never owned an autowah or envelope follower before, so maybe that's part of it. I'm more familiar with attack & decay in the context of an ADSR or AR type envelope for a VCF or VCA in a synth - is the envelope on an autowah an AR envelope or does it take the envelope from the audio signal? Also do attack and decay act like a slew generator in S+H mode? Doesn't really sound like it but they definitely do effect that setting as well.
EDIT - just realized the original didn't have the attack & decay settings - these mods documented anywhere?
Here I found some information about FSH-1 and mods:
http://www.generalguitargadgets.com/index.php?option=com_content&task=view&id=115&Itemid=26 (http://www.generalguitargadgets.com/index.php?option=com_content&task=view&id=115&Itemid=26)
On the other hand when i use mine at a higher volume i can hear the ticking even in the filter mode!!!!
So, I think that there are big problems with the design...
Yeah, depending on the settings (especially with the Range set high in up mode), I get more or less ticking as well.
Noticed that when playing in up mode, I get better response if I dial the Range setting down a bit but in down mode I get better response with it about 3/4 of the way up. Still getting a feel for setting Attack & Decay along with how hard I play. In theory I know how those settings should work, just not in this specific context, and find myself still trying to figure it out, which I will the more I play with it. It all works, though. Just wish someone could figure something out with the ticking.
Quote from: returntable on September 02, 2008, 06:26:40 PM
I read in another thread(don't really remember which one it was), that was also about a ticking LFO, that soldering a 100Ohm resistor in series with the power supply could help.
I tried it and it helped! at least a little. the ticking is even less, but still there.
So maybe changing the resitor value to 150 or 200 Ohm would work?
This helps in limiting the current for the circuit that makes up the LFO. This is a bit crude method (but hey, if it helps...) and can be improved by only decoupling the power supply for the LFO circuit as was described by me at the beginning of this thread:
http://www.diystompboxes.com/smfforum/index.php?topic=69638.msg559774#msg559774
http://www.diystompboxes.com/smfforum/index.php?topic=69638.msg560013#msg560013
http://www.diystompboxes.com/smfforum/index.php?topic=69638.msg559997#msg559997
The main problem is that the sharp change in the LFO signal when it changes direction creates a sharp current/voltage spike in the circuit that is fed into the audio-path.
Try a multiple approach:
- Change the LFO opamp to a low-power one like the TL062.
- Decouple the LFO power supply from the rest of the circuit.
- Apply some current limiting (for instance with a small resistor) to the power supply of the LFO.
- Reduce the steep slope of the LFO circuit's square wave part, perhaps by adding a capacitor at a strategic place. I had good results in a ticking SmallStone by placing a 1uF capacitor between the LFO comparator part of the LFO opamp's + and - inputs...
Good luck with it!
Quote from: flo on September 03, 2008, 07:20:42 AM
- Reduce the steep slope of the LFO circuit's square wave part, perhaps by adding a capacitor at a strategic place. I had good results in a ticking SmallStone by placing a 1uF capacitor between the LFO comparator part of the LFO opamp's + and - inputs...
Where i must to insert the 1uF in Tonepad's schematic? The capacitor must to be electrolytical or not?
Thx a lot!
He mentions this here (it's not electrolytic): http://www.diystompboxes.com/smfforum/index.php?topic=69638.msg559997#msg559997
I'm using TL082 chip and shielded wires but like I said still getting ticking. Was waiting for someone else to try the resistor and cap to ground to isolate the LFO. Sounds like folks possibly tried that without much luck. Seems like no one has tried slewing the pulse wave for the S+H yet - maybe that would do the trick (or at least in combination with the isolation and low noise chip).
Quote from: gigimarga on September 03, 2008, 10:16:55 AM
Quote from: flo on September 03, 2008, 07:20:42 AM
- Reduce the steep slope of the LFO circuit's square wave part, perhaps by adding a capacitor at a strategic place. I had good results in a ticking SmallStone by placing a 1uF capacitor between the LFO comparator part of the LFO opamp's + and - inputs...
Where i must to insert the 1uF in Tonepad's schematic? The capacitor must to be electrolytical or not?
Thx a lot!
Quote from: flo on September 03, 2008, 07:20:42 AM
I had good results in a ticking SmallStone by placing a 1uF capacitor between the LFO comparator part of the LFO opamp's + and - inputs...
One question - since when you initially suggested this you had not looked at the FSH schematic and were referencing a 741 and a different IC # in the flanger circuit you'd fixed - would this 1uf cap go between pins 5 & 6 (IC3B) or pins 2 & 3 (IC3A) or even both? And this would literally be just touching each lead of the cap directly to those leads on the IC, right (I believe you initially said you could even test this before wiring it in by just touching the caps leads to the chips inputs).
Thanks!
Quote from: flo on September 03, 2008, 07:20:42 AM
This helps in limiting the current for the circuit that makes up the LFO. This is a bit crude method (but hey, if it helps...) and can be improved by only decoupling the power supply for the LFO circuit as was described by me at the beginning of this thread:
Yes, I realized that too, but it actually does help quite a lot. Not completely, but maybe a raised value will... I'll try that.
The Problem with your Tip in the beginning of the Thread is, that it seems to me that the the LFO should already be decoupled in the Tonepad Schematic.
But maybe I've got it all wrong since I still have to learn a lot.
But If my try doesn't work, I'll try your other suggestions. Except for the different OP Amp, I've already changed that with only little success.
Quote from: gigimarga on September 03, 2008, 10:16:55 AM
Where i must to insert the 1uF in Tonepad's schematic? The capacitor must to be electrolytical or not?
Put the capacitor between pin 5 and pin 6 of IC 3b. No guarantees that it will work because it is a different LFO circuit than the SmallStone one. If it does not improve anything remove it again. It will not break anything... ;)
Elco or not does not matter.
Quote from: fogwolf on September 03, 2008, 01:06:02 PM
would this 1uf cap go between pins 5 & 6 (IC3B) or pins 2 & 3 (IC3A) or even both?
Between pins 5 & 6 (IC3B).
Quote from: fogwolf on September 03, 2008, 01:06:02 PM
And this would literally be just touching each lead of the cap directly to those leads on the IC, right (I believe you initially said you could even test this before wiring it in by just touching the caps leads to the chips inputs).
Correct, just touch the chip's pins with the capacitor leads: One on pin 5 and one on pin 6 of IC 3B.
But also combine this with the other suggestions:
- Change the LFO opamp to a low-power one like the TL062, LM358 or a TL022.
- Decouple the LFO power supply (even more) from the rest of the circuit.
- Apply some current limiting (for instance with a small resistor) to the power supply of the LFO.
- Reduce the steep slope of the LFO circuit's square wave part, perhaps by adding a capacitor at a strategic place. (well, that's covered now ... :) )
Quote from: returntable on September 03, 2008, 04:50:02 PM
The Problem with your Tip in the beginning of the Thread is, that it seems to me that the the LFO should already be decoupled in the Tonepad Schematic.
Yes, the design already shows some decoupling: 470 resistors and the 220uF capacitors for the 9V and 8V power supplies.
But it is possible to add some more dedicated resistor-capactor power supply decoupling for the power-supply of IC 3.
Your current limiting resistor is also a good suggestion that seems to work for you. Put one in the positive - and one in the negative power supply lines for IC 3 after the decoupling.
Quote from: returntable on September 03, 2008, 04:50:02 PM
... I'll try your other suggestions. Except for the different OP Amp, I've already changed that with only little success.
Try a combination. All the "little success" adds up... :)
Quote from: flo on September 04, 2008, 03:54:33 PM
Yes, the design already shows some decoupling: 470 resistors and the 220uF capacitors for the 9V and 8V power supplies.
But it is possible to add some more dedicated resistor-capactor power supply decoupling for the power-supply of IC 3.
Your current limiting resistor is also a good suggestion that seems to work for you. Put one in the positive - and one in the negative power supply lines for IC 3 after the decoupling.
All right, I will try that in the next few days and let you know whether it wored or not.
Oh and Flo, thank you so very much for all your help with this problem, I really appreciate it!
from the PCB we can see the clock circuit shares the same gnd leg as the filter ... try cutting the gnd trace which passes under C14 and run a second wire to the common ground between batteries to the filter cap ground ... another, you can add a huge cap between the rails of the audio filter circuit - ie., on downside of 470r filter resistors, between those two resistors ... another, you can try feeding the batteries on the other side of the 470r filter resistors - this sometimes works ... make sure your batteries are fresh / power source able to supply ample current to absorb the spikes without producing ground bump internally ... I didn't go too in depth to check for other spike leaks on the board
Good one about the trace cut! I didn't look at the PCB...
I've been following this thread and was curious if the original FHS-1 had ticking issues as well? I haven't had the opportunity to try one.
Quote from: Eb7+9 on September 07, 2008, 10:53:00 PM
from the PCB we can see the clock circuit shares the same gnd leg as the filter ... try cutting the gnd trace which passes under C14 and run a second wire to the common ground between batteries to the filter cap ground
I've never had to do this before - what is the best way to cut a trace on an already etched & populated board?
I use a sharp knife and cut the trace several times at the location where I want it cut, then "scrape" the groove a bit wider to make sure the connection is completely cut.
Do not forget to "run the second wire to the common ground..." so that both parts of the trace are still connected to ground.
What you are trying to accomplish here is a "star-ground" (google it...), this should reduce the LFO having an impact on the audio-path in the filter circuit via the common ground trace.
Silly question perhaps but which schematics is being referred to when talking about C14? Tonepad's schematics uses no such indicators... I assume we are takling about the ground trace connected to IC3 pin4.
Quote from: flo on September 08, 2008, 07:14:52 PM
I use a sharp knife and cut the trace several times at the location where I want it cut, then "scrape" the groove a bit wider to make sure the connection is completely cut.
Do not forget to "run the second wire to the common ground..." so that both parts of the trace are still connected to ground.
What you are trying to accomplish here is a "star-ground" (google it...), this should reduce the LFO having an impact on the audio-path in the filter circuit via the common ground trace.
Silly question perhaps but which schematics is being referred to when talking about C14? Tonepad's schematics uses no such indicators... I assume we are takling about the ground trace connected to IC3 pin4.
Okay, I will try star grounding this thing in the next days...
The trace cutting is a good idea, but the question from Flo is really appropriate: which track exactly is to be cut?
I think we should take the Tonepad schematic as a reference, since most of us built that.
I've built one of these using a PCB layout supplied by General Guitar Gadgets. I have the same issue.
I'm hardly an expert on the matter. However, I did noticed that since my charge pump is offboard that the further away from the board I moved it, the less pronounced the ticking was. Maybe using some shielded wiring on the switches would help remedy this. ???
All right, I have solved the ticking Problem for my build at least in the Filter Mode.
Like in the Echo Base Thread esplained, I soldered a 1uF Electrolytic Cap between Pins 7 and 3 of IC3(the S&H LFO). the ticking in Filter Mode is completely gone now.
The S&H doesn't work exactly how it should, but maybe I have to adjust the trimpot value again, but I wasn't able to test that yet. I will report about that tomorrow...
Woow...very good news!!!
Looking at the schematic it seems to me that you put the 1uF cap from the output of the LFO to the ground. Am I right?
I'm not sure if this was mentioned earlier in the thread, but going to low power BiMOS-based op amps may help ticking problems. Since they usually draw a lot less current per amplifier than a normal BJT/FET based op amp, they won't pump the rails as hard and don't dump as much current into the grounds. This, in conjunction with proper decoupling, should help quite a bit. Also, if you run the LFO on a completely separate op amp, this can help to prevent bleedthrough problems.
Quote from: gigimarga on February 07, 2009, 04:32:11 AM
Looking at the schematic it seems to me that you put the 1uF cap from the output of the LFO to the ground. Am I right?
exactly.
Quote from: Cliff Schecht on February 07, 2009, 04:41:00 AM
I'm not sure if this was mentioned earlier in the thread, but going to low power BiMOS-based op amps may help ticking problems. Since they usually draw a lot less current per amplifier than a normal BJT/FET based op amp, they won't pump the rails as hard and don't dump as much current into the grounds. This, in conjunction with proper decoupling, should help quite a bit. Also, if you run the LFO on a completely separate op amp, this can help to prevent bleedthrough problems.
Thanks for your ideas, but they have been mentioned before and didn't solve the problem completely.
The 1uf cap to ground off the output of the LFO will make the output pulses into a trapezoid shape rather than square, hopefully they should still switch the FET in the S/H section.
This might already of been mentioned, but adjusting the the value of the 3k3 resistor in the LFO will change the width of the pulses. Making it smaller will narrow the pulses which might reduce the ticking, of course if you make them too narrow the S/H won't work. Could be worth trying though.
Quote from: slacker on February 07, 2009, 08:10:44 AM
The 1uf cap to ground off the output of the LFO will make the output pulses into a trapezoid shape rather than square, hopefully they should still switch the FET in the S/H section.
This might already of been mentioned, but adjusting the the value of the 3k3 resistor in the LFO will change the width of the pulses. Making it smaller will narrow the pulses which might reduce the ticking, of course if you make them too narrow the S/H won't work. Could be worth trying though.
This also reduces the "error" seen by the sample and hold circuit. If you hold the "S&H pin" high too long, the capacitor will take the sample but also starts to allow some of your signals "wiggle" through. This can sound cool sometimes, other times it's undesirable. If you make the pulse-width wide enough, you can get a sort of track-and-hold type effect.
Quotetrack-and-hold
I like it!
It isn't a true track-and-hold circuit, but you can get similar effects. If you make the sample capacitor really small, it will take the sample really fast but also allows a lot more "jitter" through because such a small capacitance can't easily block a lot of fast moving transients (less averaging is happening). So when you're in "sample" mode, the sample circuit grabs a value but also allows in some signal, which gives a cool addition to any filter. It will start to jump all over the place in conjunction with your signal transients and frequency (i.e. a low frequency gives a slow moving filter).
A true track-and-hold will take an input and reflect in on the output until the gate of the T&H circuit is pulled high, at which point it grabs a value based on where your signal is at in its swing. Korg implemented this on the MS-20 synth and it gives some really cool sounds!
Quote from: slacker on February 07, 2009, 08:10:44 AM
This might already of been mentioned, but adjusting the the value of the 3k3 resistor in the LFO will change the width of the pulses. Making it smaller will narrow the pulses which might reduce the ticking, of course if you make them too narrow the S/H won't work. Could be worth trying though.
Sounds Interesting... I will try to put a Trimpot in that position al well...
Thanks Slacker :)
All right, re-adjusting the Trimpots worked, now both effects are working like they're supposed to :)
Finally... that was the hardest troubleshooting I ever had to do.
Thanks again for all the help.
PS:
I had a problem before, mine's getting a severe volume drop when effect is on. This doesn't happened the first time I tested it. So I reviewed Tonepad's layout and compare it to the schematic and JD's layout.concerns the FET orientation. I was using Fairchild's BF245A then, and its pinout facing flat side is G S D. In the layout, it is oriented backwards.
Assuming "G S D" is the standard pinout of BF245A. From the schematic, the SOURCE of Q2 was supposedly in junction with the 2 - 100K, 12K, and 3.3K resistors but in the layout they're connected to Drain. And the Source goes to +8V where Drain is supposedly the one to be connected. This has to be interchanged.
Now it's Q5, from the schematic, it's Drain goes to the 10K trimpot and Source goes to the Gate of Q6 together with the 0.05/0.047 cap. But in the layout the Source was connected to the trimpot while the Drain goes to the Gate of Q6. Again, this has to be interchanged.
JD's FET connections is the correct one.
After fixing, my unit returns to normal now.. :)
I build an FSH-1 from the Tonepad layout last year for a friend of mine. I 'cured' the ticking by killing the LFO signal when its not needed (if I remember correctly, I did it by grounding one leg of the speed pot). This means that the ticking in bypass and in filter mode are gone, but it's still there in Sample & Hold mode. But, there it is not really that much of a nuisance (gives a nice kind of percussive noises). All in all, it seems that there is something wrong with the Tonepad layout. if I ever build an FSH again, I will attempt to make my own board, and isolate the LFO circuit as much as possible. I remember reading from someone on this forum who build a separate board for the LFO with good results ...
Quote from: returntable on February 10, 2009, 02:59:38 PM
All right, re-adjusting the Trimpots worked, now both effects are working like they're supposed to :)
Finally... that was the hardest troubleshooting I ever had to do.
Thanks again for all the help.
Really?..Are you sure? coz I tried it too and it doesn't even cures the ticking. Instead, it amplifies the pulses and destroyed the sample and hold mode. Yeah, I readjust the trimmers too with no
difference in the sound.! :icon_evil:
Quote from: DimebuGG on February 15, 2009, 07:30:52 AM
Quote from: returntable on February 10, 2009, 02:59:38 PM
All right, re-adjusting the Trimpots worked, now both effects are working like they're supposed to :)
Finally... that was the hardest troubleshooting I ever had to do.
Thanks again for all the help.
Really?..Are you sure? coz I tried it too and it doesn't even cures the ticking. Instead, it amplifies the pulses and destroyed the sample and hold mode. Yeah, I readjust the trimmers too with no
difference in the sound.! :icon_evil:
really? but you added the Capacitor from the LFO Output to ground, right?
Quote from: returntable on February 15, 2009, 07:51:27 AM
really? but you added the Capacitor from the LFO Output to ground, right?
Yeah, a 1uF from pin 7 to ground(pin 3)..I used TL062 in the LFO..it really destroyed the S&H effect...
PS:
Ooops, to everyone BTW, does Xotic's version ticks too?
Okay, sorry DimebuGG and everyone else, the Pedal doesn't work correctly.
Last time, I tested it by running my WSG through it and into my Soundcard.
Today I tested it with guitar and bass through a loud amp, and the s&h doesn't work correctly with these instruments
(but it did work with my WSG)
plus at high volumes you could still hear the ticking in filter mode.
So, now I will try WormBoy's suggestion of killing the LFO when S&H is not needed and if that doesn't help I will build an extra board for the LFO.
Damn, this pedal sucks. I really start to hate it.
Quote from: returntable on February 15, 2009, 04:25:58 PM
Okay, sorry DimebuGG and everyone else, the Pedal doesn't work correctly.
Last time, I tested it by running my WSG through it and into my Soundcard.
Today I tested it with guitar and bass through a loud amp, and the s&h doesn't work correctly with these instruments
(but it did work with my WSG)
plus at high volumes you could still hear the ticking in filter mode.
So, now I will try WormBoy's suggestion of killing the LFO when S&H is not needed and if that doesn't help I will build an extra board for the LFO.
Damn, this pedal sucks. I really start to hate it.
maybe like this:
(http://i303.photobucket.com/albums/nn156/DimebuGG/fsh.jpg)
possible ground loops may occur..
The strange thing is I have mine working in S&H mode, but filter seemingly does nothing. I did have to substitute the resonance pot with a 1M because I couldn't get a 2M. I also have a strong ticking. A 10 uF electrolytic helped some, but still noticeable. I am going to try a TL022 or similar to see if that helps. I wonder if maybe the switch thing could activate a diode so that the LFO can't send out anything when in filter mode. Have the cathode oriented towards the LFO so that it blocks the LFO signal when in filter mode, hmm. I just might have to try it when I get some time. S&H mode sounds great, though. Strange, seems most have a problem with S&H and not filter mode.
Just want to chip in... I had this up on the breadboard using TL022s (good choice as they have a very low current draw). Initially pretty noticable LFO tick. Removed decoupling caps from the +ve side of the supply, leaving just the two 220uF caps on the -ve side of the supply. No ticking.
Also worth trying bigger caps for the 150nF off the collector of Q3 - this should help damp some of the feedthrough from the LFO. The Gristleizer has a similar thing - a pair of low-pass filters designed to take the edge off the LFO tick.
Quote from: flo on August 02, 2008, 08:22:46 AM
Hi, if the problem is due to LFO ticking "bleeding through" to the audio path, there are several things that you can do.
I found the following tips on this site when I was looking to cure LFO ticking in an UltraFlanger. The following tips were written by various people (Mark Hammer and others) and were used to address deticking LFO's in general but I'll try to rewrite it to make it appropriate for deticking the FSH LFO:
http://www.diystompboxes.com/smfforum/index.php?topic=61250.msg480708#msg480708
There are a few separate matters to be addressed (quotes from others slightly rewritten to adress the FSH LFO):
The LFO and audio path need to be decoupled from each other. You should run something like a 100R resistor from the shared +9v source to the V+ pin of the LFO opamp chips (especially IC5 which generates the LFO), and run a 10-100uf cap to ground from that pin. This will give each chip a small reservoir of current to hold them over in case one of the other chips makes a sudden demand for current. Those sudden demands are what you hear as ticking.
The FSH is using a 741 opamp for the LFO path (IC5) and 3080 OTA's for the (filter) audio path (IC1 IC2 Q1 Q2). This will let you decouple them separately, instead of having whatever the LFO does impact directly on the OTA's in the same hunk of epoxy.
It is also a smart idea to use a low-current op-amp for the LFO. Why? Because the "standard" opamp LFO produces a square wave. What you hear as ticking is the sudden current draw as the square wave rises, placing noise on the power line. The decoupling reduces the extent to which that spike "bleeds" into other parts of the overall circuit. Using a low-current op-amp reduces the magnitude of that spike in the first place. Think of the decoupling as being like ear muffs, and the low-current op-amp as being like smaller speakers. Use of either will keep the volume of undesirable sounds down, but use of both will reduce it even further.
You will note in glancing through schematics of commercial phaser, flanger, chorus, and tremolo pedals that many of them will use either an LM358 or a TL022 dual op-amp for the LFO, and use different op-amps elsewhere in the audio path. They use those particular op-amps because they draw less current. There are other functional equivalents, I'm sure. I just wanted to explain why you often see those ones.
Another strategy for deticking involves a slight change to the LFO design itself. Nicholas Boscorelli explains this nicely in one of his Stompboxology ( http://moosapotamus.net/IDEAS/stompboxology/stompboxology.html ) newsletters, and you can sometimes see this in commercial designs. The gist is to change that initial square wave into what is more or less a trapezoidal wave. What we hear as ticking depends on the suddenness of the rise-time. If the rise-time can be slowed down even just a wee bit, the current draw of the LFO is essentially distributed over a long enough period of time (in electronic, not human, terms) that it does not produce disturbances elsewhere in the circuit.
So, reduce the amount of current required to produce a square wave, make the rise-time of the square wave less sudden, and isolate the chip requiring the sudden current draw from the rest of the circuit. That's how you get a tick-free modulation pedal.
"In my Small Stone (russian) I have a 1uF capacitor between + and - input of the comparator opamp which is the one that produces the square wave of the LFO. This seems to detick the Small Stone, probably, by reducing the rise time of the square wave."
I solved the ticking by grounding leg 3 of speed to ground when bypassed but i experience a strong pop when i engagge the effect. Any ideas?