The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Mixed-Signal Design >> Even order harmonics in a pipelined ADC
https://designers-guide.org/forum/YaBB.pl?num=1230483119

Message started by roland on Dec 28^th, 2008, 8:51am

Title: Even order harmonics in a pipelined ADC
Post by roland on Dec 28^th, 2008, 8:51am

Hi friends,
I'm testing my pipelined ADC, even order harmonics (2nd and 4th) are higher than the 3rd harmonics. What factors may cause the rise of these even harmonics, how can i figure it out by some kind of test? Has anybody any experiences dealing with this problem?

Title: Re: Even order harmonics in a pipelined ADC
Post by ywguo on Dec 28^th, 2008, 10:03pm

You'd better check something like

signal source,
the input driver on the test board,
asymmetry in the track and hold,
the nonlinearity of the sampling switch.

Title: Re: Even order harmonics in a pipelined ADC
Post by roland on Dec 30^th, 2008, 10:09am

Thanks guo! I've checked the signal source, it's not the bottleneck after bandpass filtering.And I'v simulated again the T/H together with board level compenents and bondwires considering some asymmetry about 10%, it do not induce such large even order harmonics.
However, I fond the power and ground bounce will increase 2nd harmonic. By reducing the digital current, the 2nd harmonic is suppressed. But the 4th harmonic is affected by nothing! and it is almost as high as the 2nd harmonic, about -80db. Which factors will affect the 4th harm more than the lower order harmonics?

ywguo wrote on Dec 28^th, 2008, 10:03pm:

You'd better check something like

signal source,
the input driver on the test board,
asymmetry in the track and hold,
the nonlinearity of the sampling switch.

Title: Re: Even order harmonics in a pipelined ADC
Post by vivkr on Dec 30^th, 2008, 11:02pm

Hi Roland,

Perhaps a silly question, but how did you simulate the T/H? Was it with the same sort of
signal (high-speed sine)? Also, do you have any ESD diodes etc. at the inputs which you
may need to consider?

I would also check if the layout of the pipelined ADC is well-made (symmetrical) although I think you have already checked that. Normally, the tracking switches should
be the only source of such nonlinearity, but it may depend a bit on your T/H architecture.

What scheme do you use? Could you post a simple picture showing the switching arrangement?

Regards,

Vivek

Title: Re: Even order harmonics in a pipelined ADC
Post by wave3x on Jul 7^th, 2009, 2:36am

Hi, maybe DC offset of your OTA is another problem, you may check your behavior model and find how much DC offset you can tolerant in your system.