Take the 2-minute tour ×
Electrical Engineering Stack Exchange is a question and answer site for electronics and electrical engineering professionals, students, and enthusiasts. It's 100% free, no registration required.

I want to create parallel op-amp circuits/channels and use the value from a single potentiometer to adjust the gains of each uniformly by making each op-amp see a 'copy' of the resistance to set the feedback gain - the amps in question are TL072's and I'm looking to make at least 4 seperate channels.

Is this possible - do any circuits come to mind?

share|improve this question
    
In addition to the answer below, you can also get multi-channel potentiometers. They are called 'stacking' potentiometers or 'stacking pots'. –  Zuofu Mar 14 at 5:39
1  
Sorry I should have mentioned the 'at least 4 separate channels' would possibly go as high as 16 so I don't think a mechanical solution would be very feasible. –  norlesh Mar 14 at 5:49
    
@Zuofu Good suggestion, the only problem might be virtually all of the stackable pots are chassis mount only, which may or may not be an issue with the OP depending on whether this is a one-off project or a product. –  tcrosley Mar 14 at 5:50
    
How closely do the gains of the multiple channels need to be matched to each other? In other words, how much "gain error" can you tolerate? –  Dave Tweed Mar 14 at 11:55
    
What frequency range for the signal? How much resolution do you need in the gain? And as Dave asked, how closely must the gain of each of the channels match? There are many ways to approach this, but without any specs there is no way to know which solutions are acceptable or most appropriate. Some of this really should have been obvious. –  Olin Lathrop Mar 14 at 13:39

5 Answers 5

If you are working at audio frequencies I would recommend using a monolithic Voltage Controlled Amplifier (VCA), with the analog control voltage coming from a single potentiometer, and perhaps a buffer driver for low impedance to drive the higher impedance gain control inputs of your 4 channels.

A single channel VCA is the SSM2018 from Analog Devices. The 4-channel SSM2154 is obsoleted. I am sure there are similar devices from other vendors.

If you are working with anything higher than audio frequencies then you should look at the vendors' selections of Variable Gain Amplifier (VGA) and Programmable Gain Amplifier (PGA) devices. However these are likely to be set by a 6-bit to 8-bit digital control and not via an analog control voltage that would easily be obtained from a potentiometer.

A simple approach is to use a JFET shunt VCA topology. This picture is courtesy of Elliot Sound Products (http://sound.westhost.com/articles/vca-techniques.html)

http://sound.westhost.com/articles/vca-f7.gif

National Semiconductor AN32 from 1970 has a JFET as an attenuator at the input to an op-amp. In this configuration you would use a fixed gain to gain-up your output to a maximum and attenuate it down at the final stage using the control voltage. The image below would be the final stage, after your gain stage.

enter image description here

EDIT: Final schematic was deleted

share|improve this answer
    
Your last circuit doesn't control the gain, it controls the output level by varying the gain. Not what the OP asked for at all. –  Dave Tweed Mar 14 at 11:52
    
Yes, you're right, good point. If the input goes up, the last circuit I have drawn will compensate by gaining down.. so probably not what the OP really wants. What's the best practice here in this situation on the site? Should I edit the answer and remove the circuit? cheers.. –  Brian Onn Mar 14 at 12:29
    
Yes, feel free to modify your answer at any time. That's the goal of this site -- to incrementally improve answers so that they are more generally useful. After you do that, I'll delete my comment. –  Dave Tweed Mar 14 at 12:59
    
I removed the last schematic. No need to delete comments, the comment flow maintains continuity. –  Brian Onn Mar 15 at 4:09

If you consider the following device: -

enter image description here

Now consider the action of analogue switch (whose control input is driven by the LTC6992 i.e. it is being toggled on and off at 1MHz). You should recognize that you can control an analogue signal amplitude by feeding it through the "toggling" analogue switch. The higher the duty cycle, the more the signal gets transferred to the output of the switch: -

enter image description here

In effect you are pulse width modulating the analogue input signal and the average level of the signal is determined by the input signal and the mark-space ratio of the PWM. It's a fairly accurate 2-quadrant analogue multiplier.

After the analogue switch you need a fairly simple low pass filter to remove the high frequency switching artifacts. So one PWM chip operating up at 1MHz, several analogue switches (1 per audio channel) and several op-amp based sallen-key filters to remove the HF noise and this should work. 1 potentiometer connected to the input of the LTC6992 will control the amplitude of several signals by the same factor.

share|improve this answer

One way to do this is to use a EEPOT (Electrically-Erasable Potentiometer). They come in all values from 1K to 1M, and have up to 6 pots in one package. Here is the datasheet for the AD5253/5254, which is a quad EEPOT. You would obviously set all the pots to the same value in your case.

You will need to have a microcontroller to set the resistance value, as the EEPORT runs off of an I2C bus. Virtually all microcontrollers these days have I2C. You wouldn't even need to have a fancy user interface, just up and down buttons that would raise and loser the resistance value. If you needed to set the EEPOT to an exact value, you could add a serial interface to a PC using a UART to USB converter, like this one, which would allow you to set the resistance via a terminal program on the PC like RealTerm.

share|improve this answer
    
Thanks I had been thinking of something like that (although the devices I had looked at were only single channel), but was really hoping I could keep everything analog. –  norlesh Mar 14 at 5:46

So after some more research I found you can make a voltage controlled resistor using a FET and also this instructable that uses an LED and a photocell and resistive opto isolators also fit the bill. There is also an 'active resistor' or 'active load' used in integrated circuit designs but all the material I could find on the circuits were well beyond me at present.

share|improve this answer

The problem with VCAs and voltage-controlled resistances is achieving perfect tracking of the gain in each channel. You don't say how accurate you need the channels to be matched to each other; if somewhere around 0.5 dB variation between channels is acceptable, you have a wide range of options to choose from.

The SSM2018 mentioned in another answer is very very good, but its datasheet notes (p.11):

A 25°C temperature change causes a 8.25% increase in the gain constant, resulting in a gain constant of 30 mV/dB.

The much cheaper (dual) LM13700 features 0.3dB gain tracking between different amplifiers according to its datasheet.

However if you need much better gain matching between channels, take a look at an MDAC - Multiplying DAC - which is a switched attenuator, with the gain setting applied by changing the code on the digital inputs. So you would need to read the pot in the ADC built into a microcontroller and write the appropriate code out to the MDAC for all channels.

The gain matching achievable is limited by the precision of the MDAC, and you may be able to achieve higher bandwidth than devices specialised for audio use.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.