Deglitching a DAC

On a sunny day (Tue, 17 Jul 2012 16:37:34 -0400) it happened Spehro Pefhany wrote in :

Sample output after switch change, store, etc.?

I'll ask the designer, he's a colleague.

Message off into the ether.

What kind of logic are you using to drive the DAC? Is there any relationship to the driving edge speed, or does it depend only on the internal update signal after the data are in?

Sure, that's the idea.. but the details of ppm-level S&H are where it gets "interesting".

Best regards, Spehro Pefhany

It's doing pretty much what the data sheet says it should do, unfortunately. The glitching happens with the update.

Best regards, Spehro Pefhany

I can certainly send you some DACs, and some nice voltage references, too. You will need a uP and some displays and encoders or something, too.

One could also do it as a feedback system, where a 24-bit delta-sigma ADC would measure the output and servo it to a target value. Some of those cheap d-s adcs are amazingly good.

I have some LCD displays & encoders.

I have some 12x1 LCD displays like these:

I have some 16x2 LCD displays like these:

I have some similar to these Rotary Encoders:

I am learning to program Atmel processors on a Arduino Mega 2560. I just haven't been able to find the DACs. We used some dual 18 bit at Microdyne to set video output levels & DC offset. A lookup table gave accurate .1 dB steps over a range of 0 to -63 dB.

I'll send you an email for my address.

Generally doesn't work. There are certain code transitions that produce large spikes, such as the half-scale and the quarter-scale transitions. It is very dependent on the internal construction of the DAC, some are really HORRIBLE, some are much better.

Jon

Sure. But I have an idea for a gated deglitcher, but it wouldn't handle big voltage steps. It might work if Speff can digitally slew-rate-limit the dac steps.

Still thinking about it.

Instruments for the past 15 years or so. Less is more, no T/H is used. The idea there is to uniformize the glitch energy, or make it code independent, and shift its spectrum to the sampling frequency where the so-called anti-imaging LPF can eliminate it. This can't be done with code-dependent glitch energy since quite a lot of it remains withing the signal band. See US5646620...guess it's still protected though.

A simple slew-rate limiter circuit might make a pretty good deglitcher. Sort of like a lowpass filter, but with more precise settling behavior.

You've got to re-sample it outside the chip. There is no way to push things down to ppm levels on the same chip with the digital stuff on there. Mainly because there's lead inductance, chatter across the substrate, finite metal layer conductivity, inductive coupling, and all this fun stuff. I'd consider a diff-output DAC followed by a differential S&H. Decouple and bypass the supplies really well. Ferrite beads and 0402 or 0603 caps are your friends here.

Is the glitch also noticable when you change one bit? Adding more analog stuff often means more error & noise. Could this be handled in the digital domain by converging to the new value by flipping one bit at a time?

injection

Instruments for the past 15 years or so. Less is more, no T/H is used. The idea there is to uniformize the glitch energy, or make it code independent, and shift its spectrum to the sampling frequency where the so-called anti-imaging LPF can eliminate it. This can't be done with code-dependent glitch energy since quite a lot of it remains withing the signal band. See US5646620...guess it's still protected though.

Yep. Trick is to make desired slew-rate, yet have precise settling. ...Jim Thompson

The big problem will be charge injection in the s/h switch.

That's where the dual-gate MOSFET comes in. They work great.

Cheers

Phil Hobbs

The irritating thing is that it's directly related to the switches.. and it ought to be possible to match that stuff on the chip to very close tolerances. Maybe they'd have to trim it a bit.. but 10-20mV glitches on a chip (datasheet claim, so ideal conditions) on a chip with ~100nV of RMS output noise and 1ppm linearity seems... a bit much.

Yes, it's dependent on the number of switches changing.. so as few as one and as many as 19 switches for a single LSB output change.

Nope. Although I suppose I could declare some codes personna (numbera?) non grata and avoid the worst of them at the expense of a bit or two resolution. 8-(

injection

Instruments for the past 15 years or so. Less is more, no T/H is used. The idea there is to uniformize the glitch energy, or make it code independent, and shift its spectrum to the sampling frequency where the so-called anti-imaging LPF can eliminate it. This can't be done with code-dependent glitch energy since quite a lot of it remains withing the signal band. See US5646620...guess it's still protected though.

Not sure slew rate is the right way- the glitches look to have about the same time-voltage area above nominal as below, so a linear filter might be better.

Sure looks nice in simulation.