The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> Solutions for dielectric absorption?
https://designers-guide.org/forum/YaBB.pl?num=1294093795

Message started by tonyc on Jan 3^rd, 2011, 2:29pm

Title: Solutions for dielectric absorption?
Post by tonyc on Jan 3^rd, 2011, 2:29pm

Hi All,

I'm new to the board. I found the paper on dielectric absorption from a Google search and was seriously impressed. I could tell right away this forum is a "hidden jewel" for experienced circuit designers.

My question on dielectric absorption is really a business development one. I developed a solution to the problem from a lot of work in the battery field. Unfortunately the company I did the work for fizzled a few years back.

It recently occurred to me that my solution to the "warburg impedance" problem in electrochemistry is equally valid for the "da" problem in electronics.

Does anyone have any idea what applications would benefit most from a solution to the "dielectric absorption" problem? The paper mentions sample-and-hold and switched-capacitor ADCs but obviously the problem can't be that significant given the level of sales of those types of components.

Thanks in advance for all replies!

Title: Re: Solutions for dielectric absorption?
Post by vivkr on Jan 4^th, 2011, 5:19am

As far as I recollect, the problem of dielectric absorption starts to play a role in switched-capacitor circuits only when you start to get to ever higher precisions, and even there, it really depends a lot on the the type of circuit you are looking at.

My guess would be that you would need to look at Nyquist-rate ADCs targeting >= 14b resolution. There are not that many players in that segment, and most of those designs are likely limited by other things rather than by dielectric absorption.

Nevertheless, there is a very good paper by Fattaruso, Tewksbury et. al. in the IEEE JSSC, Dec. 1990 issue. You might want to take a look at that.

Vivek

Title: Re: Solutions for dielectric absorption?
Post by rfmagic on Jan 4^th, 2011, 6:34am

Hi tonyc,

can you post the paper that you are reffering to?

Thanks

Title: Re: Solutions for dielectric absorption?
Post by tonyc on Jan 4^th, 2011, 9:10am

Vivek,

Thanks for the reply, even though it's not very encouraging. I'll try to get a copy of the paper, appreciate the reference.

rfmagic,

Sure, the paper I was referring to is available from this site:

http://www.designers-guide.org/Modeling/da.pdf

that's what brought me here in fact.

Title: Re: Solutions for dielectric absorption?
Post by tonyc on Jan 5^th, 2011, 6:26am

vivkr wrote on Jan 4^th, 2011, 5:19am:

My guess would be that you would need to look at Nyquist-rate ADCs targeting >= 14b resolution. There are not that many players in that segment, and most of those designs are likely limited by other things rather than by dielectric absorption.

Vivek

Vivek,

Do you know who the players are in this area? I'd like to survey the market and see where things are at and what types of applications they are used in. If more resolution at Nyquist rate enables a new application, it might be valuable.

I did do some research and see what you mean, DA is a factor in these circuits but not the only one. But it does appear to be the limiting factor as far as resolution goes though from a preliminary review.

Title: Re: Solutions for dielectric absorption?
Post by vivkr on Jan 6^th, 2011, 6:12am

Tony,

The big players that are well-known are:

Analog Devices
Maxim
Texas Instruments
Linear Technology

There are surely other companies which are also in the business including several fabless ones, but those are the ones which dominate the market if I recall correctly.

I could not tell you whether the present resolution levels are limited by DA. I was under the impression that it is primarily down to matching for the Nyquist ADCs. One can use calibration etc. but pushing beyond 16 bits of resolution is probably not possible.

There are several effects which would come into play once you start to get to those resolution levels, but usually the device loses its attractiveness beyond a certain point. Even if you could make an ADC that good, say 16bits @ 100 Msps (there are some by the way), the problem is then the significant power consumption, and the difficulty involved in providing signal/clock that are pure enough to be able to actually get the maximum performance out of such designs, something that makes them less and less attractive.

However, that is only my opinion and maybe wrong. You would do well to look around a bit.

Vivek

Title: Re: Solutions for dielectric absorption?
Post by rf-design on Jan 7^th, 2011, 12:57am

Hi Tony,

is the solution at the center of the business a technology based or circuit based solution?

DA is know long time and for instance the reason why integrated high resolution ADCs (>20Bit) are working with SC techniques instead of relative long time continuous integration techniques like Dual/Multi-Slope.

DA lead to INL far below 20bit for integrated caps. In SC Circuits the clock period is relative short in relation to the wide spread time constants of the ionic transport mechanism. So only a small part of the slow ions are moved. The good part is that the signal relations withing the SC circuit are all using the same time periods. So in effect the matching of DA time constants would affect linearity.

But in SC circuits the active devices (MOS) show a similar effects which is because of the nonlinear transition phases at the clock edges the differential overdrive voltage of a MOS diffpair lead to nonlinear trap response. That hurt more than DA in poly caps now. And is getting worse with scaling. So also here circuits solutions are designers best friend.

Reiner

Title: Re: Solutions for dielectric absorption?
Post by tonyc on Jan 7^th, 2011, 6:10pm

Vivek,

Thanks again, I am continuing to look into it and your input is appreciated.

Reiner,

The solution at the center of the business is circuit based. (I think by technology you mean process/materials, and if so, no).

You've got me interested in looking at very high resolution ADCs based on integration techniques like Dual/Multi-Slope. They are slower of course but that may be ok is some apps.

The high speed of the SC ADCs may preclude my circuit solution practically speaking. But I don't know enough yet about them to know for sure, it's a concern I have for now.

If you have any links on either for more follow-up I'd appreciate them.

Tony

Title: Re: Solutions for dielectric absorption?
Post by vivkr on Jan 11^th, 2011, 5:44am

rf-design wrote on Jan 7^th, 2011, 12:57am:

DA is know long time and for instance the reason why integrated high resolution ADCs (>20Bit) are working with SC techniques instead of relative long time continuous integration techniques like Dual/Multi-Slope.

DA lead to INL far below 20bit for integrated caps. In SC Circuits the clock period is relative short in relation to the wide spread time constants of the ionic transport mechanism. So only a small part of the slow ions are moved. The good part is that the signal relations withing the SC circuit are all using the same time periods. So in effect the matching of DA time constants would affect linearity.

Reiner

Reiner,

I know that DA and its effects are known for a long while (one does read it about it even if very briefly in undergraduate physics), but I was unaware that it was the reason for using SC techniques instead of continuous-time (CT) techniques. I would have imagined that CT techniques would anyway suffer from way too much inaccuracy that one could not even get to 10 bits, let along 20 bits of accuracy, because one would be trying to match not 2 identically laid out components but trying to keep stable timeconstants.

For instance, one can use CT techniques today in making CT delta-sigma modulators, which also have poorer control of timeconstants and poorer accuracy but I never heard of DA being the major reason there. It would be interesting to see if there are any papers (obviously older ones) that discuss this limitation as the reason for using SC solutions.

Obviously, your comments about the effects of DA cancelling out in SC solutions due to use of a common clock are spot on.

Regards,
Vivek