voltage contrlled resitor circuit

[Thunderchild]

owing to the cost of integrated voltage controlled amplifiers i wish to make my own using an op amp but have the problem of a voltage controlled resistor. only method i can think of is controlling a motor on a pot or by using a light bulb and photo resistor but surly thre is a simpler way of doing it ? :?: :?: :?: :?: :?: :!:

 

[williB]

owing to the cost of integrated voltage controlled amplifiers i wish to make my own using an op amp but have the problem of a voltage controlled resistor. only method i can think of is controlling a motor on a pot or by using a light bulb and photo resistor but surly thre is a simpler way of doing it ? :?: :?: :?: :?: :?: :!:

i think what you are lookink for is a transfer resistance device..
A mosfet should do..
or a regulsr fet..

 

[Thunderchild]

ok so how do i go about controlling a mosfet like a resistor as i have never dealt with fets

 

[williB]

what is the expected current Flow..?
you may be better off with an analog switch to select different resistors if it is for an op amp ..

 

[audioguru]

FETs cause terrible distortion when used as a variable resistance in an opamp circuit. The severe 2nd harmonic distortion is converted to much less 3rd harmonic distortion if you add half the signal at the drain to the gate. Then you are left with some distortion and a very slow response caused by the coupling capacitor for the half signal fed to the gate.

Even the H11F1 opto-FET causes severe distortion unless the signal across its FET is extremely low. It is advertised as having very low distortion.

There is a method of using a dual opto-transistor as a low distortion gain control device. The distortion that is optically transferred in the 1st one is cancelled by the optically transfered signal in the negative feedback by the 2nd one. They are monolythic on a single chip so they are exactly the same. The circuit is on the broken hard drive of my old computer. :lol:

 

[Thunderchild]

as this is going to substitute a gain resistor the current is minimal and i could do with a practical circuit. distortion is not a severe issue but the less the better it is not going to directly control audio so will not be heard as it is just to control another circuit.
ps this is to put in the volume section of a theremin so it's output it will not sound but control the volume of the sound. of course severe distortion will make it more difficult to play

 

[Pyroandrew]

Don't mock the LED-LDR invention . It's fantastic, I remember swindleing top marks for a lab project at Uni using one of those. It is cheap, easy to make, easy to use, but not amasingly linear, but that is easily worked around. If your after a quick and easy bodge, the LED/LDR solution is for you.

Alternatevely, why not just use a digital pot. As linear as you like. No distortion and pretty much analogue, I'm pretty sure not many people can tell when a 0.07dB increace in gain happens. I've made a super high quality AGC using an op-amp and a digital pot before, (and a bit of control circuitry to get the 'A' bit of AGC).

You can buy voltage controlled resistors (old audio ones are basically big wippers with a small motor), you can also get more modern opto-isolators/couplers that give the same result as the LED/LDR (but better), but they are large and/or expensive.

It is quite easy to interface a simple steady voltage signal to the inputs of a digital pot, there must be hundreds of ways. But remember, if your after a quick and easy bodge, the LED/LDR solution is for you (sand off the domed part of the LED, and then superglue it to the LDR once you've sanded that as well to get a good key, plaster it in tipex or insulation tape, and there you are, A voltage controlled resistor (not really but as close as you'll get it's actually a current controlled resistor )).

 

[eblc1388]

FETs cause terrible distortion when used as a variable resistance in an opamp circuit.

Hi Audio,

Does the above also applies to CMOS bilateral switches connecting various resistors to ground to varies the opamp gain?

 

[Pyroandrew]

FETs cause terrible distortion when used as a variable resistance in an opamp circuit.

Hi Audio,

Does the above also applies to CMOS bilateral switches connecting various resistors to ground to varies the opamp gain?

An array of MOSFETS with various resistors to ground would work. You could have a binary addressing system also (8 transistors and 8 resistors i.e. 500, 1k, 2k, 4k, 8k, 16k, 32k and 64k ohms). You'd need to have the op-amp in a non-inverting arrangement, the 64k would be the LSB and the 500 ohm one would be the MSB. One a downer the amplifier would have to be onesided.

You could also have the same idea used with analog switches, this way you could use an inverting or a non-inverting op-amp arrangement, and there are no polarity issues.

 

[audioguru]

Hi L. Chung,
Cmos transmission gates like the CD4016, 4051, 4052, 4053 and 4066 are pretty good when they have no less :lol: than about a 1k load. Maxim have improved ones.

 

[mstechca]

I can't remember too well, but I think I was able to drive a motor with a single power transistor, but you might want to try this:

Take a transistor suitable for power amplifiers and power consumption. I think a TIP31 will do, but do NOT use a 2N3904, as it will overheat.

Connect the collector to +ve, emitter to motor, other end of motor to ground, and base to source. I think a resistor between base and collector helps, but PLEASE do not go under 100 ohms or I can guarantee that with at least a 12V supply, your transistor will heat. and if you want to get more cranky, drive it with 120V and I can guarantee your transistor will malfunction. LOL

sorry, I got a little carried away. :roll:

 

[eblc1388]

Hi L. Chung,
Cmos transmission gates like the CD4016, 4051, 4052, 4053 and 4066 are pretty good when they have no less :lol: than about a 1k load. Maxim have improved ones.

Thanks. I am thinking of circuit using a group of resistors with resistance from 10K upwards. At +/- 8V supply to the CMOS, I believe the gates would behave nicely.

 

[Optikon]

FETs cause terrible distortion when used as a variable resistance in an opamp circuit. The severe 2nd harmonic distortion is converted to much less 3rd harmonic distortion if you add half the signal at the drain to the gate. Then you are left with some distortion and a very slow response caused by the coupling capacitor for the half signal fed to the gate.

Can you quantify "terrible" distortion and also can you explain the origin of the 2 harmonic distortion?
Also, what mode of operation are you assuming for the FET (linear "ohmic" region?)

 

[Roff]

FETs cause terrible distortion when used as a variable resistance in an opamp circuit. The severe 2nd harmonic distortion is converted to much less 3rd harmonic distortion if you add half the signal at the drain to the gate. Then you are left with some distortion and a very slow response caused by the coupling capacitor for the half signal fed to the gate.

Can you quantify "terrible" distortion and also can you explain the origin of the 2 harmonic distortion?
Also, what mode of operation are you assuming for the FET (linear "ohmic" region?)

Here are sims of the two methods Audioguru is talking about. Note the differences in linearity of the IV curves. Asymmetry around the voltage and current axes will cause even harmonic distortion. Symmetrical nonlinearity will cause odd harmonic distortion.
Ignore the 1 Meg resistor in the circuit without feedback. I forgot to take it out, but it has no effect and can be ignored. It is needed in the circuit with feedback. Lower values might be required, depending on bandwidth requirements.

 

[audioguru]

Thanks Ron, that's what I was talking about.
If you keep the signal low, the distortion is reduced. If the signal is really low then it is buried in circuit noise and you don't notice the distortion. :lol:

 

[Thunderchild]

those diagrams look impressive so how do i use the fet in my circui then ?

 

[Roff]

those diagrams look impressive so how do i use the fet in my circui then ?

What do you need to do (gain, bandwidth, signal amplitude, available power supplies, etc.)?

 

[Optikon]

Thanks Ron, that's what I was talking about.
If you keep the signal low, the distortion is reduced. If the signal is really low then it is buried in circuit noise and you don't notice the distortion. :lol:

The one without feedback looks bad..
The second plot (with feedback) doesnt look too bad!

Looks like you swept Vg & Vd. For a given Vg, Id versus Vd looks pretty linear right through zero, just like a resistor should be. In the sense of using it as a resistor, the slope of those lines should remain constant and looks like it does (for a given Vg). So in the feedback case, it is still not clear to me how this causes distortion. Now if Vg is varied along with Vd then that makes for a non-linear resistor of course! But I don't think that's how the intended application works.

The derivative of those plots would really show how well of a resistor the Jfet makes.

Now if the linearity might fall apart with large Vd and/or large Id due to other causes.. I wonder if a network of a bunch of tiny(operating points) but linear resistors makes an equivalent "big" linear resistor.

 

[williB]

switcher cad has an FFT option..


https://www.electro-tech-online.com/custompdfs/2005/07/DS2890.pdf
also maxim has a varible resistor IC , but you would need to control it with a uController..

 

[audioguru]

This circuit will work for audio if you keep the signal level low. A negative contol voltage allows it to pass signals and it attenuates more if the control voltage becomes closer to ground. :lol: