Has anyone ever used a digital potentiometer to change the gain on an op amp? My thought is this (crude?) tactic would add noise to the signal. For some applications it may be acceptable though. Interested to hear your thoughts...
ezcomp 
Op amp + digital potentiometer = PGA?
Author
Message
ezcomp Lennart 
Nope, what I have used with good results are Light Dependant Resistors. They have very good noise values both static and dynamic. Large range 20M-R100. LED can be controlled either by a pot connected to DC or digitally.
I built a bass preamp, using digital pots to control LED's. Since I'm very concerned with quality of signal digital pot's are not trusted to touch my signal. Besides they can't handle +/- 18 volts level.
LDR's are virtually noiseless, they also have a "memory behaviour" that make them switch a bit slow which is really good for audio. LDR's where used a lot in classical vintage effects for guitar, then along came IC's and changed both price and size. Unfortunately sound quality declined dramatically. Can be compared to valve vs solid state amps.
I use 72 LDR's in my preamp, almost exclusively controlling gain of op-amps. As you can see I'm a great fan of LDR's :lol:
dmcinnes 
EDIT: deleted text as later comments were much more relevent than mine!
Darren
Last Edited: Thu. Jul 12, 2007 - 10:20 AM
Guillem Planisi 
Level: Raving Lunatic
Joined: Fri. Jun 11, 2004
Posts: 2405 View posts
Location: Vilanova (Barcelona)
@ Lennart: Just an stupid question. If valve bassed amp's give easily 10% of distortion, why are they considered better than solid state amps that give about 01% or less distortion?
Also the output transformers are another passband filter that adds phase and amplitude variations to the distortion.
Guillem.
PS: I know about the second order harmonics (one octave) in front of third order harmonics (and another) introduced by BJT's (not by MOSFET's), but 10% distorion means that you don't hear what you are playing anymore.
Lennart Guillem,
I can't explain this is a scientific way, except that as you also describe valve distorsion have other harmonics considered more pleasant to the ear. In a guitar amp you often WANT distorsion. When solid state amps arrived in market many good amps went out of production, nowadays things have changed back so that any amp considered high quality use valves. Many attempts have been made to simulate valve distorsion, no one so far have done it with great success.
OTOH it seems that also audiofiles love/prefer to build valve units, hard to believe they have 10% distorsion.
jgmdesign 
I do not recommend changing the gain of an opamp "on the fly" The simple and most important reason is because you change the frequency response of the op-amp circuit.
When I needed to create an optical agc for a fiber optic video link I designed I used a VACTROL and varied the input signal to a fixed gain wide bandwidth opamp. The optical signal was sensed via active rectifier.
There are though times when you do want to change the gain of an op-amp reat time. Like in ATE. Take a look at the AD524 from Analog Devices. It is expensive but it's possibilities are extrodinary.
Lennart Just a small add to my reply after reading jamdesigns post.
I do not change feedback resistor but input resistor value just as jamdesign describe.
bobgardner The 'emulator' Pod boxes from Line 6 claim to have transfer functions that simulate the most popular guitar pickups, amps, speaker cabinets... all done with DSPs
Lennart Line 6 are doing a great job, closest to the goal so far as I see it. They even try to model the transformer "impuruties" mentioned by Guillem, needed to get a "living sound".
DSP's have evolved fast over last decades, so the future probably belongs to them. 8)
jgmdesign Their price has also dropped into the hobbyist spectrum as well. I have been using the TMS320xxx family professionally for a couple of years now and it is amazing how cheap they have gotten
ezcomp Thanks for the hints guys. I will probably just try to avoid this technique.
ka7ehk
I've used them just fine with op-amps.
There CAN be noise (when gain is changed) if you are not careful. Run the control signal at as low a rate as you can. Slow everything down with RC filters. It can work very well.
One of the things to watch for is to minimize the DC component of the input signal. That is because every gain change results in a voltage step at the output. Some applications will be OK, but with audio, it will cause clicks.
Jim
jgmdesign Quote:
LDR's are virtually noiseless, they also have a "memory behaviour" that make them switch a bit slow which is really good for audio. LDR's where used a lot in classical vintage effects for guitar, then along came IC's and changed both price and size. Unfortunately sound quality declined dramatically. Can be compared to valve vs solid state amps.
I use 72 LDR's in my preamp, almost exclusively controlling gain of op-amps. As you can see I'm a great fan of LDR's
72 LDR's WOW!!
I have used scores of LDR's myself. As I said before I used a VACTROL in an optical AGC circuit and it worked great.
Check them out at:
http://optoelectronics.perkinelm...
Keep in mind that changing the gain as opposed to varing the amplitude of the input signal can change the harmonics response charachteristics of the circuit. This is important in audio circuits.
Good Luck!!
Jim
Lennart Quote:
I have used scores of LDR's myself. As I said before I used a VACTROL in an optical AGC circuit and it worked great.Check them out at:
http://optoelectronics.perkinelm... ... L5C+Series
I bought one of those about ten years ago when I first started my experiments with LDR's.
Since they are both large footprint and expensive I started rolling my own and eventually managed to pack four sets of LDR-LED into same size as VACTROL.
I tried to post a picture of my preamp but apparently .bmp is not an allowed extension so I had to give that up. A very good feature are that you get a galvanicly isolated connection between LED and LDR so AGND and DGND can be totally isolated from each other.
Jepael How about MAS9116 volume control chips? Handles at least 15 volts, and I think 18 volts was in specs. Has only 5V digital and analog supply pins, and needs an external opamp, but it was the opamp gain we were setting here anyway.
It might not be the best audio volume control chip there is, but it is one of the few that can handle over 12 V audio signals.
- Jani
Lennart Quote:
How about MAS9116 volume control chips?
Seem to have fairly good audio specs. 0.01 THD is rather ok...
Have you used it Jani?
Claims to do noiseless gain changes but I have read that before and been disappointed when it comes to reality with HIGH volumes.
jgmdesign Lennart:
I agree that they are a tad large, and expensive, but the devices have become much smaller and the price has also become tolerable.
Jim
Lennart Quote:
devices have become much smaller and the price has also become tolerable.
Yes, I know. Here is a company that make rather small ones.
http://www1.silonex.com/audiohm/...
Still prefer my quad packages :wink:
afit_intern nothing comes close to the sound of old tube amps. I wouldn't even think about using anything digital in my signal chain. I hate even using protools in the studio (although its inevitable). All my guitar gear is vintage where I can afford it (I have a decent collection of old amps and effects), or it is a very very accurate replica. I'm almost done building an old Marshall replica, and just finished a boost pedal. Obviously this stuff is so outdated but nothing has the same sound, feel, etc...
afit_intern I've played tons of Line 6 stuff and they make some decent (best digital stuff I've played) effects, particularly the little delay/echo unit. Nothing comes close to an old tube tape echo unit though!
barnacle
But Afit, all you're saying is you prefer the distortion modes of valve equipment :D
Neil
afit_intern Distortion takes on a whole new meaning when talking to EEs or guitarists. As a guitarist I'd say the clean tones of a Fender amp of my favorite era are really warm and deep.. lots of character, harmonics, etc.. From an engineering standpoint this is undesired distortion and probably a bit of noise. The overdrive I hear when playing my favorite Marshall stack with no master volume is completely different, but is still just distortion. Now distortion to most guitarists means a really high gain tone you can get from an crappy effects pedal or solid state amp (and a lot of over hyped modern tube amps).
barnacle
Can't say I disagree with you :)