Hi It’s *lots* easier to apply it in software than to do the fun and games with the delay line. One of the biggest issues with the delay line approach is finding one that is good enough ( stable, 0.1 ns resolution, wide enough range, low noise ….) to do the job. Bob > On Nov 10, 2020, at 4:07 PM, Matthias Welwarsky <time-nuts@welwarsky.de> wrote: > > Joe, > > I mentioned those spikes in an earlier, private email as well, I think Bob is > correct, there's something going on with the sawtooth correction. > > Since you're already reading the quantization error correction data from the > Ublox TIM_TP message, you might want to try "digitally" applying it. Leave the > delay line in bypass and just subtract the correction data from the analog > interpolator reading you get for the corresponding pulse. You need to properly > scale it to picoseconds, though. > > Regards, > Matthias > > On Dienstag, 10. November 2020 19:13:26 CET Joe & Gisela Noci wrote: >> Bob, Thanks for that. I am going to dig! >> >> Joe > > > > > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there.