Digital pots, better or worse than analogue pots ?
Opinions welcome.
I am about to start on a mixer with digital pots and controlled via USB from a PC.
Opinions welcome.
I am about to start on a mixer with digital pots and controlled via USB from a PC.
Moved to Parts
well I guess we should qualify what BETTER means?
I have heard some solutions that use Digital pots that sound very good. the channel to channel tracking tolerances are much tighter then conventional pots. yes as a DIY'er you could carefully select a well balanced pot that would work well. but with time pots get dirty and scratchy and the tracking over its range maybe good in some spots but not as good in others.
Digital pots have their issues too. static sensitive. requires the use of a microcontroller to work. and can jump from one setting to the next under certain "glitch" situations. requires the use of additional power supplies etc etc. they get pretty complex to implement well.
If you are building the ultimate no holds barred pre-amp. then the digital pot solution has advantages. But for the average DIY'er. a carefully selected pot i think is a much easier solution to implement well!
YMMV
Zc
I have heard some solutions that use Digital pots that sound very good. the channel to channel tracking tolerances are much tighter then conventional pots. yes as a DIY'er you could carefully select a well balanced pot that would work well. but with time pots get dirty and scratchy and the tracking over its range maybe good in some spots but not as good in others.
Digital pots have their issues too. static sensitive. requires the use of a microcontroller to work. and can jump from one setting to the next under certain "glitch" situations. requires the use of additional power supplies etc etc. they get pretty complex to implement well.
If you are building the ultimate no holds barred pre-amp. then the digital pot solution has advantages. But for the average DIY'er. a carefully selected pot i think is a much easier solution to implement well!
YMMV
Zc
Without offering an opinion on the digital vs. analog issue, the PGA2310, 2311, 2320 volume controls from TI seem to be the best parts in this application, there's a Cirrus 8-channel part too, CS3308/18. They have quarter- or half-dB steps and have some gain as well as attenuation. The digipots I've looked at often have worse THD. You need to make sure the source is <600R or the THD suffers.
If you achieve the stated performance in practise these parts should be transparent.
If you achieve the stated performance in practise these parts should be transparent.
For the ultimate, these two aren't your only choices, and may not be your best. A "digital pot" might mean a mechanical rotary switch attentuator - very low distortion, very good tracking (with well-matched resistors), and for "digital control" could possibly be turned by a motor. The only disadvantage I can think of is high cost of the multiposition rotary switches. Also for digital control would be a set of telecom/reed relays switching appropriate fixed-resistor attentuators, and likely controlled by a microcontroller. This would be the most complex design, but it sounds like a "fun" project.
I knocked up a test circuit and used Microchips MCP4231.
They sound very good and adjust ok without any clicks.
Making a 6 channel mixer taht is controlled via USB from a PC.
The PC has a screen with 6 faders and a bargraph display, the mixer sends back audio peak detection for the bargraph.
They sound very good and adjust ok without any clicks.
Making a 6 channel mixer taht is controlled via USB from a PC.
The PC has a screen with 6 faders and a bargraph display, the mixer sends back audio peak detection for the bargraph.
It is digitally controlled via SPI (Serial bus) and it has 129 discrete settings.
If your suggesting teh audio is not in digital format then that is true.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
I have recently been working with the PGA2310 and I'm quite impressed with it. as long as you stay within the recomended parameters (low source impedance, not too low load impedance, supply voltage, layout considerations) then the performance is good. it is definitely more complicated to implement but if you can overcome the challenges there are several advantages. Zero Cool already mentioned the superior channel matching (+/-0.05dB matching according to spec sheet) that's way better than any dual ganged pot. Also, once you have digital control you can do things that could be really useful for a mixer like store and recall settings - like a snapshot, or control several chips together creating groups. remote control. many possibilities that conventional pots aren't suited for.
-Joel
-Joel
Figured I would bump this rather than create a new topic....
What are peoples opinions these days with digital pots? There are a few applications where digital pots would be useful - Bass/Treble tone controls for instance, balance setting, etc. Digital would mean I could save or recall the settings per channel.
There are also some interesting automation applications for music creation being able to automate stuff from MIDI.
The cost of these IC's vs high quality pots is also fairly attractive TPL0102-EP ( http://www.ti.com/product/TPL0102-EP/description ) is $3.61 (cnd) in qt1 which isn't too bad.
Analog has some nice ones with 1024 steps in a reasonable price range too.
What are peoples opinions these days with digital pots? There are a few applications where digital pots would be useful - Bass/Treble tone controls for instance, balance setting, etc. Digital would mean I could save or recall the settings per channel.
There are also some interesting automation applications for music creation being able to automate stuff from MIDI.
The cost of these IC's vs high quality pots is also fairly attractive TPL0102-EP ( http://www.ti.com/product/TPL0102-EP/description ) is $3.61 (cnd) in qt1 which isn't too bad.
Analog has some nice ones with 1024 steps in a reasonable price range too.
I went with pure analog pot with motor drive. I don't have direct experience with digital pot, BUT in here, we are talking about slight distortion due to voltage coef of capacitors and even resistors already. There are no semiconductor devices that I know of don't have voltage coef. BJT has early voltage, MOSFET is even worst. There are compensation methods in semiconductor, but it is perfect? ONLY true passive component that known to have no voltage coef can truly not introducing distortion.
There might be other reasons for this, but I am not an expert. The voltage distortion alone already convinced me not to try that already. We are not talking about building any stereo here, we are here to build something that money is hard to buy, to be better(hopefully) than even the high end equipments. digital pot is not my way of starting this.
JMHO
There might be other reasons for this, but I am not an expert. The voltage distortion alone already convinced me not to try that already. We are not talking about building any stereo here, we are here to build something that money is hard to buy, to be better(hopefully) than even the high end equipments. digital pot is not my way of starting this.
JMHO
I avoid parts which contain internal opamps due to opamps in my experience having relatively poor PSRR. TI has some older National parts which don't contain opamps (LM197x) which look to be worth a try. They need external opamps but at least we can pick our own poison 🙂
I'm getting -154dBC 2nd harmonic (and no 3rd, 4th or higher) with a MOSFET switching circuit at 2KHz, but that's only a dB or two away from analyzer residual, so it might be less than that. I'm willing to bet that your pot is worse!
Yes, voltage is a problem. Switch current instead ;-)
- Status
- Not open for further replies.