I have a batch of voltage references ICs on the shelf, several thousand of each type. They all have date codes of around 1983, ie 30 years old. It would be great to have some drift data on them. The ones listed below are the best of the bunch. - AS431, TO-92 and SOIC-8, 20ppm/C 120nV/sqrtHz (0.5% or 0.1%?) - LM385Z, TO-92, 80ppm/C, 600nV/SqrtHz, 60uV, 20ppm/SqrtKhr - LM329CZ, TO-92, 75nV/sqrtHz, 30ppm/C 7uV noise 20ppm (8ppm?) /SqrtKhr. - AS2951/2954, SOIC-8, 20ppm/C I am willing to send a handful (like at least 50 pcs) to anyone who can set up and monitor the drift of them for a few months and send me some data. Any takers? I will do similar monitoring myself, but would appreciate data from others, that's why I am doing this. I am also building some other references to compare with, based on LTZ1000 and other zeners. I am not asking anything in return, except that you return drift data, measured with some decent method, ie single digit ppm stability. If you happen to have some resistors rated like 10ppm/C or lower, surface or hole mount, I'd appreciate some of those in return, in values 1K-10K, but that is not a requirement. Jan

I have a 3456A.  Is that good enough? Bob On Sunday, March 9, 2014 4:41 PM, Jan Fredriksson <jan@41hz.com> wrote: I have a batch of voltage references ICs on the shelf, several thousand of each type. They all have date codes of around 1983, ie 30 years old. It would be great to have some drift data on them. The ones listed below are the best of the bunch. - AS431, TO-92 and SOIC-8, 20ppm/C 120nV/sqrtHz (0.5% or 0.1%?) - LM385Z, TO-92, 80ppm/C, 600nV/SqrtHz, 60uV, 20ppm/SqrtKhr - LM329CZ, TO-92, 75nV/sqrtHz, 30ppm/C 7uV noise 20ppm (8ppm?) /SqrtKhr. - AS2951/2954, SOIC-8, 20ppm/C I am willing to send a handful (like at least 50 pcs) to anyone who can set up and monitor the drift of them for a few months and send me some data. Any takers? I will do similar monitoring myself, but would appreciate data from others, that's why I am doing this. I am also building some other references to compare with, based on LTZ1000 and other zeners. I am not asking anything in return, except that you return drift data, measured with some decent method, ie single digit ppm stability. If you happen to have some resistors rated like 10ppm/C or lower, surface or hole mount, I'd appreciate some of those in return, in values 1K-10K, but that is not a requirement. Jan _______________________________________________ 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.

Hello Bob, the instrument will be your least worries. with 20-80ppm/K you will need a very well stabilized (<0.1K) temperature environment for the references. Otherwise you will not detect ageing but the temperature of your environment. A volt nut would only use the LM329 from the list (buried zener). The others are bandgap references with much larger ageing drift. But to detect ageing after some 100 hrs you will need a well stabilized current source (< 0.1% drift during setup) or a even better voltage source (<0.05% over temperature and ageing) and a rather stable resistor for each reference in the setup. If you use sockets for the references their drift will outperform the reference drift if the board is handled (e.g. for measuring) so you will need to have a (relay-) multiplexer for your instrument. If you solder the references to a board you will create stress to the device and this will give a large initial drift after soldering. with best regards Andreas Am 10.03.2014 01:35, schrieb Bob Albert: > I have a 3456A. Is that good enough? > > > Bob > > > > > On Sunday, March 9, 2014 4:41 PM, Jan Fredriksson <jan@41hz.com> wrote: > > I have a batch of voltage references ICs on the shelf, several thousand of > each type. They all have date codes of around 1983, ie 30 years old. It > would be great to have some drift data on them. > > The ones listed below are the best of the bunch. > > - AS431, TO-92 and SOIC-8, 20ppm/C 120nV/sqrtHz (0.5% or 0.1%?) > > - LM385Z, TO-92, 80ppm/C, 600nV/SqrtHz, 60uV, 20ppm/SqrtKhr > > - LM329CZ, TO-92, 75nV/sqrtHz, 30ppm/C 7uV noise 20ppm (8ppm?) /SqrtKhr. > - AS2951/2954, SOIC-8, 20ppm/C > > I am willing to send a handful (like at least 50 pcs) to anyone who can set > up and monitor the drift of them for a few months and send me some data. > > Any takers? > > I will do similar monitoring myself, but would appreciate data from others, > that's why I am doing this. > > I am also building some other references to compare with, based on LTZ1000 > and other zeners. > > I am not asking anything in return, except that you return drift data, > measured with some decent method, ie single digit ppm stability. > If you happen to have some resistors rated like 10ppm/C or lower, surface > or hole mount, I'd appreciate some of those in return, in values 1K-10K, > but that is not a requirement. > > Jan > _______________________________________________ > 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. > _______________________________________________ > 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, I have an 8.5 digit wavetek, and a bunch of Solartron and wavetek 7.5 digit meters, but it seems to me that doing justice to the 329 references is going to need stable wirewound or foil resistors, and a few months will be only just long enough. MK On 10/03/2014 07:12, Andreas Jahn wrote: > Hello Bob, > > the instrument will be your least worries. > with 20-80ppm/K you will need a very well stabilized (<0.1K) > temperature environment for the references. > Otherwise you will not detect ageing but the temperature of your > environment. > > A volt nut would only use the LM329 from the list (buried zener). > The others are bandgap references with much larger ageing drift. > But to detect ageing after some 100 hrs you will need a well > stabilized current source (< 0.1% drift during setup) > or a even better voltage source (<0.05% over temperature and ageing) > and a rather stable resistor for each reference in the setup. > > If you use sockets for the references their drift will outperform the > reference drift if the board is handled (e.g. for measuring) > so you will need to have a (relay-) multiplexer for your instrument. > If you solder the references to a board you will create stress to the > device and this will give a large initial drift after soldering. > > with best regards > > Andreas > > > Am 10.03.2014 01:35, schrieb Bob Albert: >> I have a 3456A. Is that good enough? >> >> >> Bob >> >> >> >> >> On Sunday, March 9, 2014 4:41 PM, Jan Fredriksson <jan@41hz.com> wrote: >> I have a batch of voltage references ICs on the shelf, several >> thousand of >> each type. They all have date codes of around 1983, ie 30 years old. It >> would be great to have some drift data on them. >> >> The ones listed below are the best of the bunch. >> >> - AS431, TO-92 and SOIC-8, 20ppm/C 120nV/sqrtHz (0.5% or 0.1%?) >> >> - LM385Z, TO-92, 80ppm/C, 600nV/SqrtHz, 60uV, 20ppm/SqrtKhr >> >> - LM329CZ, TO-92, 75nV/sqrtHz, 30ppm/C 7uV noise 20ppm (8ppm?) /SqrtKhr. >> - AS2951/2954, SOIC-8, 20ppm/C >> >> I am willing to send a handful (like at least 50 pcs) to anyone who >> can set >> up and monitor the drift of them for a few months and send me some data. >> >> Any takers? >> >> I will do similar monitoring myself, but would appreciate data from >> others, >> that's why I am doing this. >> >> I am also building some other references to compare with, based on >> LTZ1000 >> and other zeners. >> >> I am not asking anything in return, except that you return drift data, >> measured with some decent method, ie single digit ppm stability. >> If you happen to have some resistors rated like 10ppm/C or lower, >> surface >> or hole mount, I'd appreciate some of those in return, in values 1K-10K, >> but that is not a requirement. >> >> Jan >> _______________________________________________

On 9 Mar 2014 23:39, "Jan Fredriksson" <jan@41hz.com> wrote: > > I have a batch of voltage references ICs on the shelf, several thousand of > each type. They all have date codes of around 1983, ie 30 years old. It > would be great to have some drift data on them. Since you have so many, it would be interesting if you built up some references, measured their stability, then shipped a random sample around the world - perhaps to two or three time-nuts on different continents. Then see if the transport actually has any effect on them. The discussions about the Fluke being shipped powered up made me think of this one. Of course shipping them powered up could be an additional interesting experiment, but as I wrote before, there are issues with that, although perhaps if the current consumption was low enough, the size of the battery needed would be incredibly small, and one could probably chose a type for which there is no strict regulations about carrying them. Dave