The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Measurements >> Phase Noise and Jitter Measurements >> good scope for SERDES evaluation?
https://designers-guide.org/forum/YaBB.pl?num=1320956727

Message started by carlgrace on Nov 10^th, 2011, 12:25pm

Title: good scope for SERDES evaluation?
Post by carlgrace on Nov 10^th, 2011, 12:25pm

Hope this is the right forum for this. My organization is looking at the possibility of designing high-speed serial links to integrate with out image sensors. We would need something like 1 - 10 GB or more eventually. Really this is just a what-if at this point. I'm not an expert in serial comms so I'm wondering what would be a good scope for testing our homemade SERDES? Like, for 10 GB/s links, how fast a scope do we need? Are there specific models designed for this? Hopefully someone can point me in the right direction before I do too much research and fall down the rabbit hole.

Carl

Title: Re: good scope for SERDES evaluation?
Post by raja.cedt on Nov 11^th, 2011, 12:46am

hello,
for testing at 10GBps if you have 10G BERT, 10G PRBS, signal generator which can support up to 10G and sampling scope is fine i guess for validating internal loopback and external loop back.

Thanks,
raj.

Title: Re: good scope for SERDES evaluation?
Post by carlgrace on Nov 11^th, 2011, 10:16am

So I got a little more information. It seems I need a scope with at least 1.5X the bandwidth of the bitrate (based on third harmonic of 1010101 pattern). So a storage scope with 15 GHz input bandwidth would work. Can I do the BERT with an FPGA based solution?

Title: Re: good scope for SERDES evaluation?
Post by loose-electron on Nov 11^th, 2011, 10:20am

Homemade SERDES?

LOL!

Why don't you get in touch with the manufacturers representatives for Agilent (HP) Tektronix, LeCroy, etc and ask for their reccomendations.

If its at Bezerkly, then you may be able to cut a deal with them for a free loan or similar.

Title: Re: good scope for SERDES evaluation?
Post by carlgrace on Nov 11^th, 2011, 10:39am

Loose-electron,

You're not wrong about begging... but we borrow so much stuff from Tek and Agilent they complain they are "charitable organizations". I talked with a guy at Agilent today this morning and I think I know what we need. It's really not that expensive after all.

And, no, not homemade SERDES, just homemade serial transmitters. We're running out of pads! And as you know IP is OUTRAGEOUSLY expensive unless it comes from the foundry.

Carl

Title: Re: good scope for SERDES evaluation?
Post by loose-electron on Nov 14^th, 2011, 10:14am

carlgrace wrote on Nov 11^th, 2011, 10:16am:

At those data rates you are going to have to set up a parallel data path in something at a lower frequency, and then recombine the signals to make the higher data rate in a specialized device.

Why don't you look around at whats out there already for SerDes xmitters or optical drivers?

At these data rates there is going to need to be pre-emphasis in the Xmit however.

I presume this is research at Cal?

Title: Re: good scope for SERDES evaluation?
Post by carlgrace on Nov 16^th, 2011, 2:13pm

Thanks for the help. I think we have enough info for our grant proposal.

Well, it is for research in Berkeley, I guess you could call it "Cal Adjacent". I work for the University of California, but not for Cal.

Title: Re: good scope for SERDES evaluation?
Post by loose-electron on Nov 21^st, 2011, 2:27pm

ah, got it....