So Mark suggested something like this. At least that's how I read it.
If it's incorrect Mark please say.
The idea being that when played with a less dynamic style the effects sound normal amplitude but when played harder it triggers the envelope detectors variable resistance and creates deeper notches in the effects.
By deeper notches I guess you mean more amplitude.
Shout up If you guys think it's something you guys think you'd want to build too.
Rich
(http://i.imgur.com/ge5CBaj.png)
Think of it this way, Rich: the LFO is always working, but the pick attack determines what the LFO is doing. If you're playing hard, you get to hear the LFO sweeping phase-shift stages, and if you back off a bit, it settles into tremolo.
So, it's very much like your notion of morphing from one effect into another via envelope control, but the two morphed effects share a common LFO.
In many respects, it would be a less jarring transition. And in other respects it could also be something that appears to play a trick on the ear, because the modulation rate would remain constant, but the listener would mistake the constant rate for a constant effect, potentially resulting in a "Wait, did I just hear what I think I heard?" reaction.
But yes, your diagram captures the signal flow perfectly. Consider it the aural equivalent of what gets called "trompe-l'oeil" in the visual arts. https://en.wikipedia.org/wiki/Trompe-l%27%C5%93il
PSYCHOLOGY!
Thanks for the clarification Mark.
I got it now.
I can be done in a similar fashion to the EnvoBlender I'm sure. However I fear it won't be as clear cut as one might think since you always get some bleed/blend of the two signals.
Once I'm done with the EnvoBlender I'll put my thinking cap on and research some likely candidate circuits to work with.
It might be an idea to use an op amp lfo. Some I've seen have dual wave options, triangle,square.
I also plan to look at AnotherJims soft delay and preamps over the coming weeks. Those little cd4007ub circuits sound mint.
Thanks again and all opinions and ideas greatfully received.
Dig in
Rich
Ps Mark did you listen to some of the other demos on the tube?
Nice idea. It feels to me like there must be a way to use one combined circuit rather than a separate phase-shifter and tremolo in parallel. I mean, aren't they pretty similar, fundamentally (ok, not *that* similar). But with a bit of tweaking?
The phase-shift stage is basically a differential amp. If the cap in the standard op-amp phase-shift were a resistor, the stage would be a tremolo, wouldn't it? (since the resistor to ground is variable).
That would change the problem into cross-fading between something that acts like a capacitor and something that acts like a resistor.
Dunno. I'm not there yet, I realise, but I'm sure there's a way to do it just out of reach.
T.
Maybe a flux capacitor Tom?
Aha, ETI tech tips had a similar idea I put on paper back in the 1980s if anyone can find it!
Quote from: Kipper4 on April 04, 2017, 02:50:19 AM
Maybe a flux capacitor Tom?
I've got one here in the workshop somewhere. I think I left it behind the box of dilithium crystals.
T.
Quote from: StephenGiles on April 04, 2017, 03:09:23 AM
Aha, ETI tech tips had a similar idea I put on paper back in the 1980s if anyone can find it!
It was CMOS Fuzz/Tremelo in ETI September 1982 Tech Tips, I can't find my copy though.
Thanks Stephen.
I'll look for it.
>It feels to me like there must be a way to use one combined circuit rather than a separate phase-shifter and tremolo in parallel.
If you take an op-amp phase shift stage (Phase 45, 90, etc, etc.) and short the feedback resistor between output and (-) input, the dry signal is removed and you get a variable low pass filter. This can sound quite nice and might be suitably tremolo-y.
Quote from: Kipper4 on April 04, 2017, 04:47:01 PM
Thanks Stephen.
I'll look for it.
It's here! http://www.americanradiohistory.com/ETI_Magazine.htm
Excellent thanks guys.
As Baldock says
"I have a cunning plan."
Quote from: samhay on April 04, 2017, 05:31:57 PM
>It feels to me like there must be a way to use one combined circuit rather than a separate phase-shifter and tremolo in parallel.
If you take an op-amp phase shift stage (Phase 45, 90, etc, etc.) and short the feedback resistor between output and (-) input, the dry signal is removed and you get a variable low pass filter. This can sound quite nice and might be suitably tremolo-y.
That's exactly the sort of thing I had in mind. Thanks.
That Fuzz/Tremolo actually does a hard switch from one to the other, if I'm reading it right. It'd be nice if you could get a softer fade between the two effects based on average level. So if you were to replace the feedback resistor in the op-amp phase shift stage with an OTA-acting-as-a-resistor, you'd be able to cross fade from phase shifter to lowpass.
I know the OTA-as-resistor thing works best if one end of the resistor you're replacing is grounded. Does it also work if that's a virtual earth not a real ground point?
Still, at this point, the circuit is perhaps getting complicated enough that there isn't much to be gained over a two-circuits-in-parallel approach.
Tom
For Instance.
What about if I used something like this to effectively make R15 very small (negligable)
Would that do the trick of nearly a short or just lower the gain.
Or do I need to completely remove R15 for the effect?
I still havent looked at the article. Just supposing......
(http://i.imgur.com/XpRcxXb.png)
That could work.
If it's easier to light an LED rather than make it dimmer (with the signal envelope) you could move the LDR so that it is in series with R14.
Quote from: samhay on April 05, 2017, 12:31:33 PM
That could work.
If it's easier to light an LED rather than make it dimmer (with the signal envelope) you could move the LDR so that it is in series with R14.
Either is easier light or dark. It all depends on led orientation and to which rail, with the envelope I have in mind.
Good idea Sam.
Appreciate the input. I'll bare this in mind when I come to the breadboard stage.
Rich
Don't forget that in 1982, ETI were paying £20 for any old drivel to put in Tech Tips. I took as much advantage of that as I could. However, I did have the Cmos Fuzz/Tremelo working..........but I was so much older then, I'm younger than that now!!!!!
Quote from: StephenGiles on April 04, 2017, 05:38:24 PM
Quote from: Kipper4 on April 04, 2017, 04:47:01 PM
Thanks Stephen.
I'll look for it.
It's here! http://www.americanradiohistory.com/ETI_Magazine.htm
Noted and printed off Stephen Thanks.
Inverters hmm intresting. I've only used them in one distortion.
Looking like a whole new world of possibilitys.
Cheers Rich
I've been diving into some research for this. Part of that turned up this.
(http://i.imgur.com/7xXXpFp.png)
Credit to Tom
This is going to feature in the early breadboard stages because of the low parts count and lack of crosstalk.
Next will be putting it on the breadboard with an envelope detector controlling the crossfade voltage divider (pot in the above).
Sounds so simple right.......
Have you checked out Tom's actual VCA implementation? It replaces that pot with a control voltage.
The pic you have is his single-gang pot cross-fader.
http://electricdruid.net/single-vca-crossfader/
Yer I did see that Eric but I didnt breadboard it since I assumed it was intended to be a fast acting crossfader. I did watch the video that they made of the module with it in though and it sounds good.
and ota's scare me. So I'll keep it analogue for now.
Rich
Quote from: ElectricDruid on April 05, 2017, 07:13:01 AM
Quote from: samhay on April 04, 2017, 05:31:57 PM
>It feels to me like there must be a way to use one combined circuit rather than a separate phase-shifter and tremolo in parallel.
If you take an op-amp phase shift stage (Phase 45, 90, etc, etc.) and short the feedback resistor between output and (-) input, the dry signal is removed and you get a variable low pass filter. This can sound quite nice and might be suitably tremolo-y.
That's exactly the sort of thing I had in mind. Thanks.
That Fuzz/Tremolo actually does a hard switch from one to the other, if I'm reading it right. It'd be nice if you could get a softer fade between the two effects based on average level. So if you were to replace the feedback resistor in the op-amp phase shift stage with an OTA-acting-as-a-resistor, you'd be able to cross fade from phase shifter to lowpass.
I know the OTA-as-resistor thing works best if one end of the resistor you're replacing is grounded. Does it also work if that's a virtual earth not a real ground point?
Still, at this point, the circuit is perhaps getting complicated enough that there isn't much to be gained over a two-circuits-in-parallel approach.
Tom
I noticed the hard switch between the two as well, which I'm assuming would not be too pleasing. What I'm envisioning is a sort of VCA-based biasing being applied to a transistor operating as a linear amplifier, which is amplifying/attenuating the signal you want to fade in. Maybe OTAs are the way to do it?
Have you a link to something similar to what you're proposing.
I'm not to familiar with OTA's and vca based biasing being applied to a transistor operating as a linear amplifier.
Are we talking like in some compressors/ gates. If so s there a link to a schematic?
Welcome aboard the good ship Stompbox.
Rich
Quote from: Kipper4 on April 07, 2017, 04:16:46 PM
Have you a link to something similar to what you're proposing.
I'm not to familiar with OTA's and vca based biasing being applied to a transistor operating as a linear amplifier.
Are we talking like in some compressors/ gates. If so s there a link to a schematic?
Welcome aboard the good ship Stompbox.
Rich
Good to be here! Just brainstorming, doing some research on a schematic to reference now..not as familiar with compressor/gate circuits as I'd like to be, but I can see the same concept being relevant here.
I will try and explain: think of a set-up similar to one of the JFETs in JFET switching circuits. Now think of taking that gate terminal and using it to attenuate or allow (not amplify) whatever signal you're passing through. That gate would be varied by a (narrowly-ranged) voltage generated by a VCA which is proportional to the attack of the raw guitar signal.
I could be crazy; I'll have to breadboard it and share the results/schematic with you.
I think I get where your coming from.
I'll be interested to see how you get on.
Even if it doesn't work sometimes just sharing our ideas creates sparks of intrest enough to spur others on to ideas that might be appropriate.
Like electric druids earlier post and Stephen Giles subsequently sharing the fuzz trem.
It's all good.
Thanks
Quote from: EBK on April 07, 2017, 12:28:06 PM
Have you checked out Tom's actual VCA implementation? It replaces that pot with a control voltage.
The pic you have is his single-gang pot cross-fader.
http://electricdruid.net/single-vca-crossfader/
One thing I couldn't see very well is what pin the pnp transistors control voltage goes to.
Can someone enlighten me please.
Thanks
Rich
Pin 1
Thanks Eric
That would be the amp bias input then according to the datasheet.
Tolerance of upto 2ma. Scary.
http://www.ti.com/lit/ds/symlink/lm13700.pdf
I'm used to outputing upto 7volts from an envelope detector. Although I'm sure it can be padded down to within a range of voltage who's currant can fit. I guess a BAR or divider would help.
Not my field really so I could be barking up the wrong tree.
I note however that the cross fader (lm13700) is expecting a 0-5v control voltage.
I'm not quite sure if the lesser voltage will only partially cross fade the input signals.
Is that how that work?
Cheers
I believe that circuit coverts an input voltage to a control current. You can do the same. No need to worry. :icon_wink:
That's exactly what's going on - it converts an input voltage to an output current, which is what the 13700 expects at its control input.
I'd post a decent reference for that part of the circuit, but oddly I can't find one. Anyone? It's not like it's some amazing thing I came up with myself...it must have been around for 40 years at least (since the advent of op-amps).
Tom
I knew I'd seen this. I printed this off recently.
http://www.geofex.com/Article_Folders/VCA%20Applications.pdf
> it converts an input voltage to an output current
> It's not like it's some amazing thing I came up with myself...
The 5K input resistor actually does the V/F conversion (and the source must be able to deliver 1mA).
The opamp, PNP, and other 5K just bend that around to go "down", as needed for the OTA Iabc pin.
You can put 10V into the "0V-5V" input before the OTA smokes. It may survive a bit more (but you can't complain if it doesn't). Scale your input to never go much over 5V, or limit the input voltage.
Thanks Paul.
this may be a goer then with an ota crossfader. I'll breadboard it at some point with the envelope detector.
Quote from: Kipper4 on April 08, 2017, 03:32:43 AM
Thanks Paul.
this may be a goer then with an ota crossfader. I'll breadboard it at some point with the envelope detector.
What are you using for the envelope detector, or have I missed something ?
Probably something of this ilk Stephen. At this time it's just a series of ideas in my head.
http://www.diystompboxes.com/smfforum/index.php?topic=117156.msg1087463#msg1087463
Quote from: Kipper4 on April 08, 2017, 10:32:08 AM
Probably something of this ilk Stephen. At this time it's just a series of ideas in my head.
http://www.diystompboxes.com/smfforum/index.php?topic=117156.msg1087463#msg1087463
You may want to consider how the envelope detector is triggered - presumably by the leading edge of the note played?
Is there another way?
Can I suggest that you should convert the half-wave rectifier into a full wave rectifier? It'd reduce ripple in the envelope follower and give you a better result. Costs you another op-amp, I admit.
http://www.play-hookey.com/analog/feedback_circuits/full-wave_rectifier.html (http://www.play-hookey.com/analog/feedback_circuits/full-wave_rectifier.html)
T.
It's entirely possible I may even go a step further with the envelope. Thanks Tom.
I saw this here
http://www.deeptronic.com/electronic-circuit-design/envelope-detector-circuit-with-separate-attackrise-and-decayrelease-time-settings/
I got to thinking how malleable it was.
I haven't breadboarded this yet. I've just redrawn it for my envelope archive.
I'm curious to know if you guys think I should reference the caps, decay pot and other stuff to Ground rather than +V/2 (Vb my drawing).
(http://i.imgur.com/6dyTTRd.png)
What am I missing? Any improvements before I bread it? Of course all opinions are welcome.
Thanks guys
Rich
Oops copy and paste error. My drawing.
C5 c7 need to be 10uf NOT 1uf.
There's a lot on that schematic that I don't understand. That leads me to believe that it's too complicated!
Firstly, I don't understand how the first op-amp is supposed to produce full-wave rectification. I'll have to simulate it and have a look.
Secondly, I can't believe you need two more op-amps and so many diodes and electrolytic caps just to do variable attack and decay. Although I suppose you've got to compensate both the diodes, so maybe you do. Just seems over-complicated to me, and makes me think "there must be a simpler way...".
Aside from throwing a lot of "stuff" at the problem, it seems to work fine, so that's all good.
I don't understand it that well, but I'd have thought you've done the right thing going to Vb. That's the equivalent of the centre "Ground" in the original circuit.
Tom
Fair enough I get where your coming from Tom.
So based on your reccomendation
http://www.play-hookey.com/analog/feedback_circuits/full-wave_rectifier.html
I drew this.
since the diagram was a block diagram with brief explanation, I'm not sure it's a good reperesentation
so I'm looking for clarification
Thanks
Rich
(http://i.imgur.com/cc386rE.png)
Scratch that. I think I found it over at Rod Elliots.
Only difference is the diodes are reversed. fig4 here
Right?
http://sound.whsites.net/appnotes/an001.htm
I've breadboarded fig 4 and it works ok.
I did buffer the front end and put an inverting op amp led driver on the back end with a gain of ~100-200.
I tried the diodes oreintated both ways and concluded that it inverts the led action.
Quick update.
I've been testing full wave rectifiers. Heres some ghetto lab notes.
(http://i.imgur.com/E5uuBih.png)
link
http://imgur.com/E5uuBih
edit images edited
Re Fig D
I might need to buffer the V/2 Vb as the rest of the circuit is unsettling the input buffers voltages.
I hooked up an amp to the buffer output (via a cap) and noticed its getting very splatty(tech term) on harder played noted.
fig D hardly responds to lightly played notes but hard strummed ones extinguish the led which might be a useful feature in a project where the second (phaser) effect is only supposed to come in when played hard.
I also will stick the cro probes on it and see if i can spot anything useful. DC wise at least.
I might have to play with the Merlins Engineers Thumb recifier next.
In all of those schematics, you're rectifying with reference to Vb, and then smoothing with reference to Ground.
That doesn't seem right. If your signal is all "positive" wrt to Vb, then you need to have your envelope referenced to Vb too, no? Otherwise you're suddenly adding on a Vb-sized offset at the bottom. I can believe that might get "splatty" (to use the technical term).
What do you think?
Aside from that, you've still got the "changing attack time changes envelope level"problem mentioned in the article you posted, which is why they'd used *quite* so many op-amps. But I still think there must be a better way... I just can't quite *actually find it* yet...;)
Still, only a question of time, right? How old are we?!? Years to go yet!
Tom
Years to go yet Tom I'm only in mid 50's.
What do you think of using the engineers thumb rectifier in combination with your 13700 cross fader?
I'll look at referencing the charge cap etc to Vb.
Is there anything useful in this?
https://www.dropbox.com/sh/2aedsoafmaa3bc7/AACiigi5aVYTjrPP1ybQMO_oa?dl=0
Thanks Stephen I'll have a good read.
and this lot from Aphex which they removed from their website!!!
https://www.dropbox.com/sh/gwftwz5hfx1ijvn/AAC3pUtghFVp2qe2HeCu-flpa?dl=0
Thing with precision rectifiers is you can only get Vb to near +supply (if positive rectifier) or Vb to 0v (if negative rectifier). They make more sense with bi-polar supplies when a positive rectifier can do 0v to near +supply (if some DC gain added).
There are trick designs using CMOS op-amp, taking advantage of the fact that these don't mind the input going negative. So the input is AC coupled and 0v referenced. These think they have bi-polar supply, but as they can only produce positive output, anything negative is ignored (sounds like description of someone with an MBA!). The output can range 0v up to near +supply if some DC gain is built in.
I've messed with precision rectifiers quite a bit and not sure if they are worth it.
I go for diode charge pump half wave. They are rectifiers and voltage doublers. Ground referenced output and perfect for single supply amps that can work in and out to 0v like the 358. Ideally, I'd precede it with an amplifier that can swing closer to the +supply than a 358 can, because it can only use the swing above Vb (but must have swing below Vb to prime the pump capacitor).
If you want full wave, an inverted signal can be fed to a second positive charge pump and blended with the first.
Unlike a precision rectifier, they do blank input below the diode threshold - but do you want every little noise from the guitar to kick the envelope?
Quote from: anotherjim on April 15, 2017, 08:18:20 AM
These think they have bi-polar supply, but as they can only produce positive output, anything negative is ignored (sounds like description of someone with an MBA!).
:icon_lol: nice one.
Thanks Jim.
I'm thinking I need a rectifier that only really reacts to hard strumming.
Since the concept is that the phaser notches only kick into the mix when strummed harder.
I've done a fair bit of messing with rectifiers too. I'd like to get something in the parish of a 5v pp swing so I can experiment with Toms lm13700 single vca
Here
http://electricdruid.net/wp-content/uploads/2015/06/VCBigMuffTonecontrol.jpg
Form this article
http://electricdruid.net/single-vca-crossfader/
I'll go back over the charge pump rectifiers.
I'll find a link to my first concept idea so you can get a better idea.
Thanks
Something's are worth taking your time over. I'd rather not, in the future wish to redesign it.
Rich
The cmos idea you talk about Jim is it the way he mentioned here
http://www.diystompboxes.com/smfforum/index.php?topic=113273.40
Sorry I missed your post Stephen.
Looks like I have some more reading to do. Thanks.
Rich
Quote from: Kipper4 on April 15, 2017, 04:31:38 PM
Sorry I missed your post Stephen.
Looks like I have some more reading to do. Thanks.
Rich
I couldn't remember if I'd passed that Aphex stuff your way.
I think the DOD440 has a charge pump rectifier.
http://www.generalguitargadgets.com/pdf/ggg_ef440_sc.pdf
Yes, the DOD is a perfect example. D1 and it's position right before D2 is vital. The Range pot ( the pump pre-amp gain) is a good addition as you can't really make the envelope detection with fixed performance suit everything.
True that Jim
Its on the breadboard having some treatment right now
(http://i.imgur.com/qP3impk.png)
That looks right. One thing is you can get more pump action if the amp driving the diodes was an 072, because it can swing closer to +supply than the 358 can. It's probably only 0.5v better, and may not make much practical difference, but it's something if you do need a bit more envelope range.
Thanks for the tip Jim
I'm getting a 5~7v pp swing at the led.
I've made a few changes, no doubt there will be more to come.
I've changed the gain of Ic2B. With the test instrument i can get the led only flashing when strummed hard.
Next up is that Attack pot, not sure how effective it is, maybe i'll try something 5x or 10x bigger.
(http://i.imgur.com/N0mwjJV.png)
Heres a mock up.
If you spot anything that wont work please chip in.
(http://i.imgur.com/1RmsE1q.png)
Will the phase of the effects affect the outcome?
Will the 2N5087 pnp work? Does the circuit care as long as it's PNP?
Bare in mind both Tremolo and Phaser will be run by the same LFO.
Is it unwise to use the other half of the lm13700 as a tremolo? As in RG's OTA applications?
link.
http://www.geofex.com/Article_Folders/VCA%20Applications.pdf
Thanks
with IC1A DC coupled to IC2B, R7 becomes redundant. I'm not sure about your D3 direction when it has ground hanging off the anode.
however - jim - if using an lm358, the output can swing to 0V, so if the opamp sinks the led current instead, won't the available range/swing/brights be increased?
Welcome aboard Stephen.
Good spot on R7.
You're correct D3 is reversed and goes to +9v as in reply 60
I got so many drawings in the folder it's easy to pluck the wrong one out. My apologies.
Otherwise no one sees any issues?
I also think D3 will be redundant later on. At the moment it's talking to me about what the envelope is doing.
Yes led's talk..... in my breadboard world......
They also listen, by the way. :icon_wink:
http://www.analog.com/en/analog-dialogue/raqs/raq-issue-108.html
Quotehowever - jim - if using an lm358, the output can swing to 0V, so if the opamp sinks the led current instead, won't the available range/swing/brights be increased?
I mean the op-amp driving the diode pump. That one doesn't need to swing completely to 0v, but does need a symmetrical swing so it works the pump capacitor (C2 in most recent scheme) efficiently. After the diodes, I would definitely prefer a 358 in order to follow the envelope voltage all the way down to zero.
Looking over some phaser schematics for the project I'm beginning to wonder if a 2 stage phaser is going to cut it especially as its the effect that kicks in when strumming gets harder.
What do you guys think?
Obviously a two stage OTA based would be a smaller pcb footprint than a 4 stage.
Here's a previous project by a forumite
http://www.diystompboxes.com/smfforum/index.php?topic=77241.
Also not forgetting the causality 4
http://www.diystompboxes.com/smfforum/index.php?topic=80456.0
Bare in mind
So far there's an envelope detector to control the mix of the two effects
The tremolo
The phaser (a single lfo will control speed of both effects.)
The OTA based single vca cross fader.
Some buffers.
The power supply
I Have another drawing for a tremolo based on RG's VCA application notes. I'll post it up later.
Thanks
Rich
Goddam wasted my day breadboarding the lm13700 vca with no sucess.
Rant over..
heres my drawing im working from.
(http://i.imgur.com/OSchbxV.png)
I've managed to get the op amp crossfader version working and responding to the envelope.
Theres still some tweeks and breadboard nonsense to sort but if I cant get the ota version working at least I know I can go old school.
Yip yip......
In the OTA, why Q3? I don't know origin of this. Maybe it adds some log/exponential kink to the sweep, but should there still be some resistor into the OTA control pin to set a current limit?
Q3 is a current source. Pin 1 of the OTA expects a control current.
To be more accurate, it is part of a current source.
Well yes, but so is the op-amp with a series resistor, which is the usual arrangement in OTA control. Will Q3 just cut off when control on R15 moves below Vb?
Quote from: anotherjim on April 18, 2017, 06:58:53 AM
Well yes, but so is the op-amp with a series resistor, which is the usual arrangement in OTA control.
Would you mind sketching that up or providing a link to an example?
OP-amp (often just a buffer) > Resistor (somewhere about 10k per control pin) > OTA control pin.
Opamp input voltage change = OTA control pin current change.
Dod FX-25 already mentioned uses envelope cap direct to feed resistor into the control which doubles as the cap bleed off.
One Rich refers to...
(http://i210.photobucket.com/albums/bb292/frequencycentral/Causality4PhaserRev10.jpg)
I'd love to know where the current transistor idea comes from. A bi-polar supply synth circuit with exponential CV control?
A simplified version of this?
(http://electro-music.com/forum/phpbb-files/rs20_781.png)
"I'd love to know where the current transistor idea comes from. A bi-polar supply synth circuit with exponential CV control?"
yes +12v -12v
Synth yes.
The op amp version works. Albiet with just clean guitar (no tremolo) and a plugged in phaser to the second differential input.
See the Electric druid site for details of the cross fader.
Using a version of the DOD440 envelope detector.
More on this later.
Some of the values in the detector have changed but essentially this is whats on the breadboard and in the video.
Except D4 the vactrol led is reversed and goes to gnd via the 2k2 CLR. So that it lights up when strummed harder.
As drawn It will go dark.
(http://i.imgur.com/N2yBDlj.png)
I just tried it with a tremolo and phaser and it works as intended. It can be a little tricky setting the envelope gain so It functions as you want but it's still early days.
I can see this being useful for any 2 effects of your choice with provisos.
I've added a few bells and whistles.
Envelope sensitivity pot. Must be a BAP
also a pot to adjust the gain of the effect that kicks in.
Some adaptation to envelope gain.
Corrected the vactrol led in the drawing. Changed the vactrol to a ready made model.
Today is a good breadboarding day.
#CheersDino.
#Get me being all modern......
Had my wife not been admitted to hospital for emergency treatment on dangerously infected leg ulcers I would have been able to speak from the breadboard, but unlikely for the time being as current regime of hospital/work/hospital seems open ended at the discression of Axa PPP's willingness to pay!
I'm sorry to hear that Stephen.
I hope she recovers soon.
I dislike hospitals. I also hope you get AXA to cough up.
Take care of yourself my friend.
Wishing Mrs Giles to get well soon.
I wish my breadboard spoke to me.
It would have said "Your wasting you're time tonight boyo, might as well pack it in."
I would have gone to bed a lot earlier last night.
Rich
Thanks for your good wishes.
Quote from: Kipper4 on April 19, 2017, 02:55:41 PM
I'm sorry to hear that Stephen.
I hope she recovers soon.
I dislike hospitals. I also hope you get AXA to cough up.
Take care of yourself my friend.
Wishing Mrs Giles to get well soon.
+1
+1!! Hope she'll get well soon!! :(
me too on the get wells.
Thank you all.
Because of the uneven number of op amps. I changed the input buffer to an N channel fet type.
Which meant adjusting the envelope gain again and the sensitivity pot more in line with the impedence. Or at least thats how I imagined it. So I changed it to much smaller and It's better.
I may well change the envelope to a darlington envelope.
I also lowered the bias point of the envelope to around 2v. Which should give me more swing.
I'm not sure if the led driver is going to work off the breadboard it too may need a biasing resistor to the +input pin.
Also on the cards is to try using the darlington envelope detector with an Expandora led driver.
Thanks for that one Dino.
More later.
Rich
Following something RG said.
http://www.diystompboxes.com/smfforum/index.php?topic=117220.0
I've hatched a new cunning plan.
I'm fishing for opinions.
Might this work with Toms OTA crossfader (Seen here
http://electricdruid.net/wp-content/uploads/2015/06/VCBigMuffTonecontrol.jpg
)
(http://i.imgur.com/BEqkSNq.png)
Quote from: Kipper4 on April 22, 2017, 04:35:09 AM
Following something RG said.
http://www.diystompboxes.com/smfforum/index.php?topic=117220.0
Some great discussion of OTAs and current mirrors in that thread, but it could have used some of Mark Hammer's award-winning analogies to jazz it up a bit more. :icon_wink:
QuoteI've hatched a new cunning plan.
I'm fishing for opinions.
I can't help but think that if you push that op amp buffer over so that it replaces the left hand transistor in that mirror that it is essentially where you started. In other words, I think you're using more parts than you need to (something I do in every single circuit I design, by the way :icon_wink:).... But, I haven't had my coffee yet.
Never mind the thread Eric. What dya think of what i've come up with? You think it will work?
Maybe you could give my tired brain a quick boost and tell me the max voltage you expect across C67.
My hunch is that this will not work, but again, coffee.
I'll expose my thinking a bit here, but no judging me if I'm flat out wrong right now.
When your voltage is at 0 (C67 fully discharged), your current will be (9 - 0.7 - 0)/47 = 0.177A, enough to destroy the OTA utterly.
I'm going to be of no help whatsoever and say - kipper, the ET env diodes are pointing 'backwards', and Merlin has used non-polar caps there. you might need to check the voltage generated and its polarity there, or just use poly caps and don't think about it. arse-about always confusels me.
I agree with Eric. I think the 47R in that current mirror isn't big enough. When that transistor starts conducting, the OTA is toast.
I've been wrong before though. But it never saved an OTA.;)
Tom
I had to go look at a milliamperes to amps converter. Ain't google marvellous.
From what I can gather the current would be in the 100x too large.
0.002=2ma.
Theoretically I therefore surmise I'm gonna need a resistor at least 5k resistor right?
Should I swap the 47R to 5k or use a series 10k to be sure on the current mirror output.
Thanks for any guidance you can offer.
I appreciate it.
Rich
No time for a thorough answer, at the moment, but a few quick things:
1. This one would be good to fully test without the OTA, so you can take some actual current measurements, to be safe. :icon_wink:
2. It may be a good idea to throw some overvoltage protection on your rail here (that current is based on a voltage that is essentially referenced to your positive supply -- I'm not using enough words here to explain this properly)
3. There will be some temperature dependence here, but I don't know if it would matter (this one just randomly jumped to the front of my engineering mind while I was writing the others)
Hopefully, I'll have time to take a better look later.
Well thanks Eric.
I already have my envelope detector on the breadboard and a box of breadboard transistors. Random bunch of raggle taggle things though they are.
Puts me in mind of going to see Bob and the wailers back around 1979.
I'll stick the DMM on the current mirror output and have a gander.
Cheers.
Rich
I use a regulated 9v power supply so most of the time. Don't encounter issues but I guess it might be wise to use a Zener over the power supply cause you never know who's gonna plug what into it.
I know zilch about temperature dependance. In fact I have two left feet I can't dance.
:)
Going back to what I said about too many parts, I think this may be better (ignore the values):
(https://cdn.instructables.com/FL9/YN1I/HEXSGAHU/FL9YN1IHEXSGAHU.MEDIUM.jpg). It's similar to Tom's original design.
Benefit of this arrangement is that your control voltage goes into the non-inverting op amp input, giving you that high impedance buffering.
I'll sketch it up and give you the relevant math once my kids are in bed. :icon_wink:
power supply 9v1
envelope output 2v4 no signal
envelope output with strong strum upto 7v
with a 2n5087 pnp
R3=10k (ref last posted diagram )
same output as above
let me go look at your previous calcs and see if i can figure it out
Still willing to bet its gonna be too much though.
(9-0.7-2.4)/10 000 = roughly 6/10 000 0.0006 might be too small
with R3= 2k2
6/2200 =0.0027 its still gonna make the lm13700 fry.
with R3= 4k7
6/4700 = 0.0012ma seems much more ballpark right?
Quote from: Kipper4 on April 22, 2017, 07:24:22 PM
(9-0.7-2.4)/10 000 = roughly 6/10 000 0.0006 might be too small
with R3= 2k2
6/2200 =0.0027 its still gonna make the lm13700 fry.
with R3= 4k7
6/4700 = 0.0012ma seems much more ballpark right?
Yes, it does to me.
Let's just say I'd start well on the safe side and work back, rather than the other way around.
T.
The negative side of your electrolytics are connected to ground, and so, there may exist the possibility of the positive side going lower than 2.4V due to leakage, etc. As a quick and dirty fix, you can bump up the negative side of the caps to 2.4V.
I just tried the analogue op amp version crossfader with a normal channel (chorus)
and an enveloped fading channel (Distortion)
It's all gone a bit 80's hair metal.
fun though.
I'll come back to the OTA version later I think.
Phaser untested in breadboard trials.
Meantime.....
HammerTime. Can't touch this.........
(http://i.imgur.com/v6n7Qnf.png)
Crikey! He's firing off ideas like a catherine wheel! The chap's on fire!
T.
I have some hammer time observational questions - are you missing a blocking cap before the threshold pot, connected to Vr and the Q13, or is it in the wrong position on the dia (C77)?
and - is Q1 biasing high enough from the Q5 emitter, which would be around 3V?
don't burn the kippers, catherine.
Well That's how I have the blocking cap on the breadboard and so far I havent had any audable problems.
The reason I put it there is because of the different biasing on the envelope follower. Maybe i need another between the input buffer and the sens pot too.
Would you think it will cause an imbalance/instability?
I havent tested the phaser portion with Q1 and Q5 in circuit so I dont know the answer to the biasing question mate.
Still burning.
ahh, I see. if'n youse takes the pot to ground instead of Vr, it should be okey dokey.
Diagram fixed I missed off a cap between sens pot and buffer.
good Spot D.A.