> From: john@devereux.me.uk > To: volt-nuts@febo.com > Date: Fri, 31 May 2013 14:05:39 +0100 > Subject: Re: [volt-nuts] cadmium solder alloy for low thermal emf? > > "Andreas Jahn" <Andreas_-_Jahn@t-online.de> writes: > > > Hello > > > >>I do not imagine cadmium bearing solder being easy to acquire. The > >> Wikipedia entry for solder says Pb90Sn10 can be used as a replacement > >> for Cd70Sn30 in low thermal EMF applications: > >> > >> http://en.wikipedia.org/wiki/Solder > >> > >> On Thu, 30 May 2013 04:00:19 +0200, Volker Esper <ailer2@t-online.de> > >> wrote: > >> > >>> > >>>By the way: does anyone know, if Agilent uses special solder alloy? I've > >>>heard that a cadmium containing solder is used to get extremely low > >>>thermoelectric voltages (or voltage differences). > >>> > >>>Is that right? If so, which alloy has to be used? > >>> > >>>Thanks > >>> > >>>Volker > >>> > > > > Within LT AN86 Cd60Sn40 is recommended for a limited temperature range > > of 0 to around 40 degrees. > > http://cds.linear.com/docs/en/application-note/an86f.pdf > > > > But: the thermal EMF is only zero against copper. > > Most precision integrated (hermetical) cirquits use Kovar. (39uV/K > > against copper) > > Relay contacts will be either copper berillium or another material. > > So in most cases a optimized solder for copper/copper connections will > > not be useful. > > I have never understood why it matters anyway. The conductors being > soldered together end up in very good thermal contact. So there should > be no thermocouple generated by the solder-conductor interfaces (since > there is no temperature difference between the ends of the wires being > soldered). > > So if you have copper-solder-copper, say, then the copper-solder > junction is microns away from the solder-copper junction and is > surrounded by metal. So surely they will be at the same temperature > unless there is a huge heat flow. > > The referenced AN86 even suggests introducing balancing > copper-solder-copper junctions, by cutting tracks and bridging with > solder. > > [...] > > But then why does it make a difference to stop the air currents around sensitive/accurate circuits? MK > -- > > John Devereux > _______________________________________________ > 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.

m k <m1k3k1@hotmail.com> writes: >> From: john@devereux.me.uk >> To: volt-nuts@febo.com >> Date: Fri, 31 May 2013 14:05:39 +0100 >> Subject: Re: [volt-nuts] cadmium solder alloy for low thermal emf? >> >> "Andreas Jahn" <Andreas_-_Jahn@t-online.de> writes: >> [...] >> > Within LT AN86 Cd60Sn40 is recommended for a limited temperature range >> > of 0 to around 40 degrees. >> > http://cds.linear.com/docs/en/application-note/an86f.pdf >> > >> > But: the thermal EMF is only zero against copper. >> > Most precision integrated (hermetical) cirquits use Kovar. (39uV/K >> > against copper) >> > Relay contacts will be either copper berillium or another material. >> > So in most cases a optimized solder for copper/copper connections will >> > not be useful. >> >> I have never understood why it matters anyway. The conductors being >> soldered together end up in very good thermal contact. So there should >> be no thermocouple generated by the solder-conductor interfaces (since >> there is no temperature difference between the ends of the wires being >> soldered). >> >> So if you have copper-solder-copper, say, then the copper-solder >> junction is microns away from the solder-copper junction and is >> surrounded by metal. So surely they will be at the same temperature >> unless there is a huge heat flow. >> >> The referenced AN86 even suggests introducing balancing >> copper-solder-copper junctions, by cutting tracks and bridging with >> solder. >> >> [...] >> >> > > But then why does it make a difference to stop the air currents around > sensitive/accurate circuits? Hmm, well, it must be because the connectors or IC legs are of a different material than the wire/tracks. *Then*, each junction *is* a thermocouple, so that if it sees a temperature different to another part of the circuit it will generate an error. And if it sees an air current, that temperature will fluctuate and so will the error voltage. But adding some magic solder to the junction should not help :) For the "balancing junctions" of AN86 the suggested (copper track - solder bridge - copper track) junction consists of two opposing thermocouples that oppose each other. Such a junction will "measure" the temperature difference between the ends of the solder bridge. But this will be miniscule and highly dependent on the geometry, the shape of the tracks and solder blob. -- John Devereux