Author |
Message |
rogerlatur
Joined: Dec 22, 2012 Posts: 118 Location: france
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Wed May 22, 2013 12:41 pm Post subject:
|
|
|
There are several ways to do voltage control.
You can control a resistance with voltage and you can control a current with voltage. There are even ways of controlling capacitance with a voltage.
For a resistor, it is common to use what is called a "vactrol". This is a device which contains some kind of light emitting device, like a lightbulb or an LED which shines on an LDR (Light Dependant Resistor). The LDR then is the resistance that is changed when the amount of light hitting it changes. There are also devices like the H11F1, H11F2 and H11F3 which are "opto-isolators" of a sort. These devices contain an LED and an MOS transistor. The light from the LED strikes the MOS transistor and changes the resistance (or conductivity) of the transistor. Usually, the light emitting device is controlled by a current which is controlled by a voltage...
Controlling a current is most often done using an OTA (Operational Transconductance Amplifier). This is a device that is similar to an operation amplifier, but it has an input called Iabc which stands for Amplifier Bias Current. Changing the Iabc current value changes the gain of the OTA and thereby changes the current output of it. Currents can also be controlled by using a transistor or even a diode (as in transistor ladder filters and diode ladder filters).
Changing a capacitance with a voltage is often done with a transistor switch that turns on and off rapidly. This switch is used with some fixed capacitor value. The more "on time" the switch has, the closer to the actual capacitor value the circuit will present. Lower on times cause the capacitance value to decrease. For audio, you want the switching rate to be above the ability of human ears to hear. The rate will remain constant, but the duty cycle is what is controlled by voltage. This is done with a PWM (Pulse Width Modulation) circuit.
What I have written here is just an overview, there are likely other ways to do these tasks, but these are the ones that I've personally seen as most common.
HTH _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
beep
Joined: May 05, 2013 Posts: 105 Location: Germany
|
Posted: Wed May 22, 2013 2:49 pm Post subject:
|
|
|
This is very interesting. Do you have experiences with the H11F1? Is it similar to a vactrol? _________________ Is the future obsolete? |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Wed May 22, 2013 2:57 pm Post subject:
|
|
|
I've played with them. They are in fact similar to a vactrol, but much much faster, so there's no vactrol lag that a lot of folks like (however, you can put the lag into the driver circuit or the output circuit). The reason is that an LDR is fairly slow in responding to changing light intensity, where as the MOS transistor is mondo fast in comparison. I used a current source to control the LED to make sure I didn't blow it. You might be able to get away with a resistor and an opamp, but I have no idea how that affects the transfer function curve (linearity). Something to experiment with. _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
beep
Joined: May 05, 2013 Posts: 105 Location: Germany
|
Posted: Thu May 23, 2013 2:31 am Post subject:
|
|
|
cool! I'm happy if it works. gonna stock alot of them, instant CV
they look beautiful. white DIL-6-packages.
"Hey maan, what's that mysterious white chip on the synth board"? _________________ Is the future obsolete? |
|
Back to top
|
|
|
-minus-
Joined: Oct 26, 2008 Posts: 787
Audio files: 13
|
Posted: Thu May 23, 2013 3:23 am Post subject:
|
|
|
I wasn't aware of the voltage controlled capacitance method. Thanks!
I think a good simple method of voltage control of resistance, particularly for beginners, is making your own 'vactrols'.
Just tape together a superbright LED facing a light dependant resistor(LDR). You'll need a resistor on the LED so you don't burn it out, and the tape needs to be light proof.
A better method would be to use some heat shrink tubing to join the two. They are really fun to make! Connect the LDR where a pot would go and feed a voltage source to the LED. Some LDRs are faster than others but for bent instruments it may not matter all that much.
Here:
http://ruinwesen.com/blog?id=1325 |
|
Back to top
|
|
|
analog_backlash
Joined: Sep 04, 2012 Posts: 393 Location: Aldershot, UK
Audio files: 21
|
|
Back to top
|
|
|
rogerlatur
Joined: Dec 22, 2012 Posts: 118 Location: france
|
Posted: Sun Jun 02, 2013 2:29 am Post subject:
|
|
|
JovianPyx wrote: | There are several ways to do voltage control.
You can control a resistance with voltage and you can control a current with voltage. There are even ways of controlling capacitance with a voltage.
For a resistor, it is common to use what is called a "vactrol". This is a device which contains some kind of light emitting device, like a lightbulb or an LED which shines on an LDR (Light Dependant Resistor). The LDR then is the resistance that is changed when the amount of light hitting it changes. There are also devices like the H11F1, H11F2 and H11F3 which are "opto-isolators" of a sort. These devices contain an LED and an MOS transistor. The light from the LED strikes the MOS transistor and changes the resistance (or conductivity) of the transistor. Usually, the light emitting device is controlled by a current which is controlled by a voltage...
Controlling a current is most often done using an OTA (Operational Transconductance Amplifier). This is a device that is similar to an operation amplifier, but it has an input called Iabc which stands for Amplifier Bias Current. Changing the Iabc current value changes the gain of the OTA and thereby changes the current output of it. Currents can also be controlled by using a transistor or even a diode (as in transistor ladder filters and diode ladder filters).
Changing a capacitance with a voltage is often done with a transistor switch that turns on and off rapidly. This switch is used with some fixed capacitor value. The more "on time" the switch has, the closer to the actual capacitor value the circuit will present. Lower on times cause the capacitance value to decrease. For audio, you want the switching rate to be above the ability of human ears to hear. The rate will remain constant, but the duty cycle is what is controlled by voltage. This is done with a PWM (Pulse Width Modulation) circuit.
What I have written here is just an overview, there are likely other ways to do these tasks, but these are the ones that I've personally seen as most common.
HTH |
Very interesting and very useful! Thank you so much.
I am glad I asked here.
I had really no idea about these methods, nor did I know about the H11F1.
At the moment I see a lot of potential in many circuits and I ignore if it will remain a dream or not. It will depend how far I go with experimenting with the methods you are mentionning here.
The first thing I would like to try is to implement CV control on a digital delay to be able change its time, which parameter is currently controllable via pot.
I guess it is less complicated to achieve here because there is a pot, so I can control the signal from here? I will start experimenting carefully. The most important is that I do not destroy anything. Maybe I will even try first on another circuit I have no problem with if it then does not function anymore after my operation! |
|
Back to top
|
|
|
rogerlatur
Joined: Dec 22, 2012 Posts: 118 Location: france
|
Posted: Wed Jun 05, 2013 10:49 am Post subject:
|
|
|
To be sure (DIY noobie + my english being far to be perfect, I want to be sure I ask exactly what I need) !
If I take a parameter on a module being already controllable via pot (like the time parameter on a delay module or the freq on an osc) I will have to go through one of the ways described earlier (res, cap, ota ...) to get this same parameter controllable via CV (the CV arrives via banana cable from modules generating voltages - like a signal generated by a seq for instance) ?
I will have to identify to which part of the circuit or to which wire the pot is connected to, then pick (split) this signal and connect it to the output of an OTA. The input of the OTA beeing wired to the banana jack (which gets the seq CV signals) ? Do I understand the principle or am I miserably wrong |
|
Back to top
|
|
|
gdavis
Joined: Feb 27, 2013 Posts: 359 Location: San Diego
Audio files: 1
|
Posted: Wed Jun 05, 2013 6:33 pm Post subject:
|
|
|
rogerlatur wrote: | To be sure (DIY noobie + my english being far to be perfect, I want to be sure I ask exactly what I need) !
If I take a parameter on a module being already controllable via pot (like the time parameter on a delay module or the freq on an osc) I will have to go through one of the ways described earlier (res, cap, ota ...) to get this same parameter controllable via CV (the CV arrives via banana cable from modules generating voltages - like a signal generated by a seq for instance) ?
I will have to identify to which part of the circuit or to which wire the pot is connected to, then pick (split) this signal and connect it to the output of an OTA. The input of the OTA beeing wired to the banana jack (which gets the seq CV signals) ? Do I understand the principle or am I miserably wrong |
There's no simple answer to your question, it depends entirely on the circuit you're trying to control and requires an understanding of how that circuit functions. If you post a schematic maybe someone can give you more specific details.
It's possible that the existing pot simply provides a control voltage, but there are many other possibilities as well. _________________ My synth build blog: http://gndsynth.blogspot.com/ |
|
Back to top
|
|
|
rogerlatur
Joined: Dec 22, 2012 Posts: 118 Location: france
|
Posted: Tue Jun 18, 2013 7:31 am Post subject:
|
|
|
gdavis wrote: | rogerlatur wrote: | To be sure (DIY noobie + my english being far to be perfect, I want to be sure I ask exactly what I need) !
If I take a parameter on a module being already controllable via pot (like the time parameter on a delay module or the freq on an osc) I will have to go through one of the ways described earlier (res, cap, ota ...) to get this same parameter controllable via CV (the CV arrives via banana cable from modules generating voltages - like a signal generated by a seq for instance) ?
I will have to identify to which part of the circuit or to which wire the pot is connected to, then pick (split) this signal and connect it to the output of an OTA. The input of the OTA beeing wired to the banana jack (which gets the seq CV signals) ? Do I understand the principle or am I miserably wrong |
There's no simple answer to your question, it depends entirely on the circuit you're trying to control and requires an understanding of how that circuit functions. If you post a schematic maybe someone can give you more specific details.
It's possible that the existing pot simply provides a control voltage, but there are many other possibilities as well. |
I understand.
I would really like to add the functionality on a Boss DD5 Digital Delay pedal to control its delay time via CV and this is the main reason why I first asked here more generally.
It might be difficult as it is a digital delay ? I searched online for circuit bending on DD5 but I did not find CV of delay time. Though I found circuit pictures but not the DD5.
I made a new thread here with the exact mod question.
Thank you very much. |
|
Back to top
|
|
|
yogi
Joined: Jun 26, 2008 Posts: 29 Location: Maryland
|
|
Back to top
|
|
|
rogerlatur
Joined: Dec 22, 2012 Posts: 118 Location: france
|
|
Back to top
|
|
|
rogerlatur
Joined: Dec 22, 2012 Posts: 118 Location: france
|
Posted: Wed Jul 10, 2013 1:08 pm Post subject:
|
|
|
I tried with JFET to CV a simple osc. I made it on breadboard to test control voltage and the tracking is sometimes weird depending how/where I connect D/S/G, but it works. I will show here when I have time how I got finally the best results and to allow you to comment the best way to achieve this. It might help others here too. |
|
Back to top
|
|
|
elektrouwe
Joined: May 27, 2012 Posts: 143 Location: Germany
|
Posted: Thu Jul 11, 2013 7:15 am Post subject:
|
|
|
rogerlatur wrote: | I tried with JFET to CV a simple osc.... |
Salut Roger, I see no good reason to use a JFET in this application(VCO), because you can get a VCO with much better linearity when it is based on a VC current source instead of using a VC resistor. JFETs as VC-resistors do a
much better job in VCA and VCFs, so better waste your time with JFET controlled VCA/VCFs and use a simple BJT current source for your oscillator |
|
Back to top
|
|
|
rogerlatur
Joined: Dec 22, 2012 Posts: 118 Location: france
|
Posted: Sun Jul 14, 2013 4:07 am Post subject:
|
|
|
elektrouwe wrote: | rogerlatur wrote: | I tried with JFET to CV a simple osc.... |
Salut Roger, I see no good reason to use a JFET in this application(VCO), because you can get a VCO with much better linearity when it is based on a VC current source instead of using a VC resistor. JFETs as VC-resistors do a
much better job in VCA and VCFs, so better waste your time with JFET controlled VCA/VCFs and use a simple BJT current source for your oscillator |
Thank you very much for the advice.
I am doing this breadboard thing to help me understand/learn what is doing what and how it functions (incl. limitations), comparing solutions etc, so your advice of course helps me to move in the right direction when it comes in further steps to build some own modules. Cool !
I would like to try with vactrols now, but did not find a concrete example with a simple osc where a pot is already controlling the freq / pitch (this I have done on breadboard with a very simple sawtooth osc and a pot wired to plus/ground and sent to opAmp - this is the one I took to test for CV control with JFET after I read yogi post) and where I could add a vactrol to add CV functionality (an external voltage would be sent to an input, allowing to control the freq via ext. CV, in combi with the pot).
I searched the web but did not really find where an example to start (also a lot of old broken links in forums where people are seeking for a solution too). What I miss is what to connect to where (which vactrol pin), knowing that it might be different depending on the project.
I thought I would ask first here, so there is no risk I destroy the only vactrols I have (a couple of VTL5C3/2, quite expensive) but depending on reactions I might as well just try, like I did after looking at the mfos page about JFET, though I had to wire them in a complete different way.
With the vactrol I am not sure at all which pins to wire, not even after I looked at their datasheets like I use to do to look at applications where we get all infos about wiring.
Next step would be to have it track 1v/oct. For me it all means "challenge", because I am new to all this.
Vactrol CV:
Maybe I should send my signal from banana input getting voltage to VTL5C3/2's pin2, getting the CV and then pin 1 to ground (I see a cathode marked in the datasheet) ? And I could try with pin 3,4,5, but I do not want to harm the vactrol.
If I look in more detailled doc like here:
http://pdf1.alldatasheet.com/datasheet-pdf/view/84389/PERKINELMER/VTL5C3.html
I have no idea how in the applications to voltage control anything via vactrol either.
Maybe you know a schematic where we see the use of vactrol to extend the functionality of a simple module like an oscillator for instance, where originally only a pot is controlling the pitch, but with the vactrol it is then possible to control the pitch via external CV as well?
This might help me to understand where the vactrol is wired and how it interacts. |
|
Back to top
|
|
|
gdavis
Joined: Feb 27, 2013 Posts: 359 Location: San Diego
Audio files: 1
|
Posted: Sun Jul 14, 2013 3:15 pm Post subject:
|
|
|
In input of the vactrol is an LED. You need to convert the control voltage into a current in an appropriate range for the LED. Might be as simple as putting a resistor in series depending on how your control voltage is sourced.
The output is a resistor whos resistance varies depending on the intensity of the LED. This resistor goes wherever you need a variable resistance. Pins 3,4 and 5 of a VTL5C3/2 basically act like a pot.
Finding a current range that makes for a good continuous response might be a little tricky, that's where the performance curves in the data sheet come in but you might still need a trimmer to compensate for variations from one vactrol to another. _________________ My synth build blog: http://gndsynth.blogspot.com/ |
|
Back to top
|
|
|
rogerlatur
Joined: Dec 22, 2012 Posts: 118 Location: france
|
Posted: Mon Jul 15, 2013 8:47 am Post subject:
|
|
|
gdavis wrote: | In input of the vactrol is an LED. You need to convert the control voltage into a current in an appropriate range for the LED. Might be as simple as putting a resistor in series depending on how your control voltage is sourced.
The output is a resistor whos resistance varies depending on the intensity of the LED. This resistor goes wherever you need a variable resistance. Pins 3,4 and 5 of a VTL5C3/2 basically act like a pot.
Finding a current range that makes for a good continuous response might be a little tricky, that's where the performance curves in the data sheet come in but you might still need a trimmer to compensate for variations from one vactrol to another. |
I understand !
I will make some tests. Thank you !!! |
|
Back to top
|
|
|
rogerlatur
Joined: Dec 22, 2012 Posts: 118 Location: france
|
Posted: Tue Jul 16, 2013 9:58 pm Post subject:
|
|
|
I made all tests and realized after all worked*** fine with the JFET and the Vactrol (more or less depending on resistor) that the CV was still working after taking both circuit parts (JFET or Vactrol) away !!!
Pot still wired, changing the pitch too.
External incoming voltage (sequence or whatever), to resistor, to opAmp input: CV works the same as with JFET or Vactrol (it means these 2 did nothing actually).
I am quite confused now !
***Worked fine means the small saw oscillator which reacted to pot (wired between ground and plus) is also reacting to incoming voltage. The tracking does not seem OK though, but I am now confused about the way we described CV and voltage which needed to be transformed to current.
I am sorry for always asking.
Again I think I need to make more basic circuits before asking about CV because there is something I do not get here.
Before I posted my questions here I did test and never got CV because the resistor between incoming voltage signal and opAmp was missing and though I never heard any pitch difference before.
Then I asked and read about the different ways to achieve CV.
Now I do not know what would be the difference as I do not get any difference with or without adding the different components. It works the same, so I should be happy so (and see in a future step how to achieve 1v/oct tracking), but I guess you see what is confusing me.
Maybe it works with or without JFET/Vactrol because this simple osc example is driven by the pot ? Or maybe I asked the wrong way and it works anyway always because I have already a voltage incoming and did not explain enough that I wanted to add to CV anything while sending an external voltage signal in ?
I want to be able to add CV so an oscillator can react to a sequence for instance. Or an LFO. Or add CV to a delay, so its time parameter can change depending on incoming voltage (a for seq or whatever) for instance. I understand it will depend on the original circuit, but it really looks like I am missing the point here. |
|
Back to top
|
|
|
gdavis
Joined: Feb 27, 2013 Posts: 359 Location: San Diego
Audio files: 1
|
Posted: Wed Jul 17, 2013 12:00 am Post subject:
|
|
|
rogerlatur wrote: | I made all tests and realized after all worked*** fine with the JFET and the Vactrol (more or less depending on resistor) that the CV was still working after taking both circuit parts (JFET or Vactrol) away !!!
Pot still wired, changing the pitch too.
External incoming voltage (sequence or whatever), to resistor, to opAmp input: CV works the same as with JFET or Vactrol (it means these 2 did nothing actually).
I am quite confused now !
***Worked fine means the small saw oscillator which reacted to pot (wired between ground and plus) is also reacting to incoming voltage. The tracking does not seem OK though, but I am now confused about the way we described CV and voltage which needed to be transformed to current.
I am sorry for always asking.
|
If I'm understanding this correctly, it sounds like the oscillator is already voltage controlled with a pot generating the control voltage as I mentioned might be the case in my first reply in this thread. So ya, to have another CV input all you likely have to do is apply the CV through a resistor to the opamp input. This will sum the input with the voltage from the pot. The value of the resister will affect the scaling and depends on the feedback resistor of the opamp. JFET or vactrol is not an appropriate solution in this case.
It would help a lot if you included circuit diagrams in your posts instead of trying to describe in words. You're getting a mixed bag of replies because you're making us guess a lot about what you're working with and it's making things more confusing. _________________ My synth build blog: http://gndsynth.blogspot.com/ |
|
Back to top
|
|
|
rogerlatur
Joined: Dec 22, 2012 Posts: 118 Location: france
|
Posted: Wed Jul 17, 2013 9:21 am Post subject:
|
|
|
gdavis wrote: | It would help a lot if you included circuit diagrams in your posts instead of trying to describe in words. You're getting a mixed bag of replies because you're making us guess a lot about what you're working with and it's making things more confusing. |
I asked too generally how to CV anything (an osc, a delay module) and decided to try with a simple oscillator circuit, but I see exactly what you mean.
Thank you very much gdavis for the advice + precious help !!! I will do so in the future. |
|
Back to top
|
|
|
gdavis
Joined: Feb 27, 2013 Posts: 359 Location: San Diego
Audio files: 1
|
Posted: Wed Jul 17, 2013 11:09 am Post subject:
|
|
|
I'll show you some examples. (Lets see if I can do this before I have to leave)
Take the Yusynth VCO:
http://yusynth.net/Modular/EN/VCO/index.html
In the upper left are the CV inputs for frequency. They all go to a summing opamp. The "coarse" and "fine" inputs actually come from pots that produce a voltage, the 1V/oct inputs can be used with any CV source. Notice the 1V/oct inputs have 100k resistors, but the pot inputs have other values to control the amount of scaling. You can pretty much have any number of inputs you want here (within reason).
Now lets look at the Yusynth ADSR:
http://yusynth.net/Modular/EN/ADSR/index_new.html
Notice the "attack", "decay" and "release" pots. They are all variable resistors in the timing path of the circuit, not generating control voltages. The most straight forward way to make this voltage controlled is probably a vactrol since the output acts as a variable resistor, so you can just drop that in place of the pots.
There's a key point here I want you to notice: a JFET is not a very good choice because the gate voltage is relative to the JFET source pin (Rds is proportional to Vgs). If you just drop the JFET in place of the pot to vary the resistance, the source is not ground but the signal your trying to control, so you don't have accurate control over Vgs which is what is controlling the circuit.
Finally, take a look at the Yusynth diode VCF:
http://yusynth.net/Modular/EN/EMSVCF/index.html
Look at Q12 near the middle. This controls the resonance by shunting it to ground. When it's off, the feedback bypasses the JFET and goes to the base of another transistor in the filter. When it's on, the feedback is basically shorted to ground so it no longer influences the transistor.
What makes the JFET work here is the fact that the JFET source pin is connected to ground, not the signal being controlled, so you have a solid Vgs to control. _________________ My synth build blog: http://gndsynth.blogspot.com/ |
|
Back to top
|
|
|
rogerlatur
Joined: Dec 22, 2012 Posts: 118 Location: france
|
Posted: Fri Jul 19, 2013 10:29 am Post subject:
|
|
|
Thank you so much gdavis for these examples and the explanations !
I will take time to check the schematics.
Your blog looks very interesting !
DIY = amazing !!! |
|
Back to top
|
|
|
rogerlatur
Joined: Dec 22, 2012 Posts: 118 Location: france
|
Posted: Sat Jul 20, 2013 1:10 pm Post subject:
|
|
|
To show how ignorant I am and how important it is to post schematics or pictures of what we are referring to before asking (like gdavis adviced):
When I remarked the CV worked on my circuit (a simple oscillator, see picture at the bottom) even without JFET or Vactrol, I thought (I am sincere. I know it sounds stupid, but I am learning out of these mistakes) that it worked because the pot in the circuit (driven by power) was generating voltage (I mean I understood it this way and here again we see that posting more concrete questions with examples might save time and confusion. Mea culpa !), so today I disconnected the pot and thought I will then try with Vactrol to see what happenĀ§. But the incoming CV was working !
The resistor R between incoming voltage and opAmp input, which I added while trying with Vactrols in previous tests and did not have at the beginning in the circuit (this is the reason why I asked here how to add CV because all my tests were not successful), is essential and allows CV in my case.
Now I wil try with this concrete simple example to understand a couple of important points:
Measurements in red color.
When I measure at 1, I see the voltage from the incoming CV input. Okay.
When I measure at 2, there is no voltage shown on my tool.
And now I think I am missing here something essential in DIY. In this thread voltage and current were mentionned quite often. Is it exactly here the case we talk about since the beginning ? Having to transform voltage into current or current into voltage depending on the circuit ?
At 2, if not a voltage is it then a current (you see how ignorant I am I did not know we could have current without voltage !!!!!! People here will fall down from their chairs, I know and I am reasy sorry !).
I measure a current in amperes here not changing, but what kind of signal else should I then be able to read here ?
Thank you very much for your explanations !
|
|
Back to top
|
|
|
gdavis
Joined: Feb 27, 2013 Posts: 359 Location: San Diego
Audio files: 1
|
Posted: Sat Jul 20, 2013 10:49 pm Post subject:
|
|
|
What you are measuring is correct. With the + input of the opamp grounded and negative feedback (in this case through the capacitor between the opamp output and - input), the - input is a "virtual ground". It's not actually connected to ground, but the circuit behaves in a way to maintain that point at the same potential as ground.
Do a google search on basic opamp circuits, there are a lot of good explanations out there, better than I could do here. Wikipedia isn't bad: http://en.wikipedia.org/wiki/Operational_amplifier _________________ My synth build blog: http://gndsynth.blogspot.com/ |
|
Back to top
|
|
|
|