Designer’s Guide Community

baohulu

Community Member

Offline

Posts: 43
china

is there any difference between TH and SH
Mar 18^th, 2006, 10:17pm

IN some papers, they use track and hold amplifier, but others use sample and hold amplifier, is there any difference between them?

in the input of the pipeline ADC, most papers present a SH and a first stage in one square as the first part after the input, but in these papers, there is no particular SHA circuit presented, I want to know whether the SH is an independent part in the pipeline or just be merged with the first stage? And if the SH is merged with the first stage, is there any difference between the merged and the independent?

thanks

Back to top

IP Logged

huber

Community Member

Offline

GO BEARS!

Posts: 45
Los Angeles

Re: is there any difference between TH and SH
Reply #1 - Mar 19^th, 2006, 5:44pm

The terms THA and SHA are used interchangably in the literature. There's no real difference I am aware of.

It is common practice these days to merge the THA of a pipeline ADC with the first pipeline stage. This has the advantage of contributing less noise for a given power consuption, but sampling becomes trickier. Because the input is continuous-time, you have to worry about timing skews between the sampling capacitor and the sub-ADC. Sometimes special clock phases are used in the merged first stage to mitigate timing skews, and often the input switches will be bootstrapped for good tracking linearity. Aside from these minor differences (and perhaps a scaling factor), the merged first stage is usually identical to the other stages.

-Dan

Back to top

IP Logged

baohulu

Community Member

Offline

Posts: 43
china

Re: is there any difference between TH and SH
Reply #2 - Mar 19^th, 2006, 9:04pm

hi, huber
could you recommend some papers about what you have said that the special clock phases in the first stage merging with the SHA.

And just in the circuit with independent SHA, there are two systematic diagram, one is the input directly go to both the sub-ADC and the SHA, the other is that the input goes to the SHA, then the output of the SHA goes to both the sub-ADC and the sampling capacitors ( both the structures have been in the Gray's papers). If it is the former case, the slew still exist even there is a SHA, only in the later case the SHA help decrease the slew in the ADC and the sampling capacitors. Do I have some misunderstanding in these cases?

thanks

Back to top

IP Logged