Nice board! Will give you a lot of experience. See forward if you have more to report. What 2ppm/degC resistors do you use? Do you have any previous experience with them? Like noise , jumps could have its own thread. I think I have seen 5-10uV jumps on most of my 10V refs based on buried zeners. On some it has been years between and some very often. For some I have only seen it during temperature scans over a span of a couple of °C. Sometimes at low temperature and sometimes at higher temperatures. Of course the LM399 is special with its fixed heater. On one DMM based on LM399 I can have minutes or sometimes hours with 1.5ppm shift in reading but when it goes back for a long time. This have been ongoing for 20 years for that particular DMM. Lars Från: Ian Johnston Skickat: ‎onsdag‎ den ‎25‎ ‎november‎ ‎2015 ‎23‎:‎29 Till: Discussion of precise voltage measurement Hi all, Thought you guys might be interested in a small PCB I made in order I could test, select & check pre-age (if there is such a thing!) my LM399's. 10 channel. Each channel has it's own chopper op-amp (self bias the LM399), 2ppm/degC tempco resistors throughout, 2x2 sockets for the LM399's (same footprint). Test points for each channel. Jumper link on the power rail to disable each channel individually (op-amp and heater power). Holes around each LM399 footprint to help isolate any thermal stress. Here:- http://www.ianjohnston.com/temp/lm399select.jpg So far it's working out great.......i've already discovered one of my LM399AH's (sitting in ch.4 in the photo) is somehow unstable. It's output jumps around by about 10uV's. Need to investigate........ The idea is that I can test, select and pre-age LM399AH's before soldering them into my target PCB's. I understand that soldering can/will reset any aging.......but I figured it might help. I did a pre-test on those Ebay LM399's I bought and they seem to work ok.......still to properly evaluate.....but I do notice that the new ones's I bought from Digikey are all giving over 7Vdc, the Ebay ones all well under 7Vdc. Cheers! Ian. -- IanJohnston.com /"because it feels good to make stuff"/ [Sent by MDaemon Mail Server at IanJohnston.com] _______________________________________________ volt-nuts mailing list -- volt-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts and follow the instructions there.

Hi, Sasumu RG2012L-223-L-T05 from Digikey - 408-1636-1-ND 0805 package. £2.90 each in low qty. My first experience of them....... I was also trying 5ppm/degC resistors but they drifted like hell......... I haven't done the sums yet so either the 2ppm ones are good, or the 5ppm ones aren't so good.......there was too much drift to simply be 2 versus 5 ppm/degC. The jumps/noise one the one LM399 is every second.......haven't investigated yet. Ian. On 26/11/2015 15:46, Lars Walenius wrote: > Nice board! Will give you a lot of experience. See forward if you have more to report. > > > What 2ppm/degC resistors do you use? Do you have any previous experience with them? > > > Like noise , jumps could have its own thread. I think I have seen 5-10uV jumps on most of my 10V refs based on buried zeners. On some it has been years between and some very often. For some I have only seen it during temperature scans over a span of a couple of °C. Sometimes at low temperature and sometimes at higher temperatures. Of course the LM399 is special with its fixed heater. On one DMM based on LM399 I can have minutes or sometimes hours with 1.5ppm shift in reading but when it goes back for a long time. This have been ongoing for 20 years for that particular DMM. > > Lars > > > > > > > Från: Ian Johnston > Skickat: ‎onsdag‎ den ‎25‎ ‎november‎ ‎2015 ‎23‎:‎29 > Till: Discussion of precise voltage measurement > > > > > > Hi all, > > Thought you guys might be interested in a small PCB I made in order I > could test, select & check pre-age (if there is such a thing!) my LM399's. > > 10 channel. > Each channel has it's own chopper op-amp (self bias the LM399), > 2ppm/degC tempco resistors throughout, > 2x2 sockets for the LM399's (same footprint). > Test points for each channel. > Jumper link on the power rail to disable each channel individually > (op-amp and heater power). > Holes around each LM399 footprint to help isolate any thermal stress. > > Here:- > http://www.ianjohnston.com/temp/lm399select.jpg > > So far it's working out great.......i've already discovered one of my > LM399AH's (sitting in ch.4 in the photo) is somehow unstable. It's > output jumps around by about 10uV's. Need to investigate........ > > The idea is that I can test, select and pre-age LM399AH's before > soldering them into my target PCB's. I understand that soldering > can/will reset any aging.......but I figured it might help. > > I did a pre-test on those Ebay LM399's I bought and they seem to work > ok.......still to properly evaluate.....but I do notice that the new > ones's I bought from Digikey are all giving over 7Vdc, the Ebay ones all > well under 7Vdc. > > Cheers! > > Ian. > > > -- Ian Johnston www.ianjohnston.com https://www.youtube.com/user/IanScottJohnston /"Because it feels good to make stuff!"/ [Sent by MDaemon Mail Server at IanJohnston.com]