Does anyone have any input on a Transmille 8081 >From what I can see it looks as good or better than a 9458A -- *John Phillips*

I'm a little bit suspicious of this one. Rumors have it that they use a V/F converter as opposed to the "normal" multi-slope technique. This may work just fine, and then again maybe not. I noticed that they write about "high linearity" in their brochure, but no figures for ADC linearity [either INL or DNL] are to be found anywhere on their data-sheet or in their manual. If Transmille is so proud of their DMM's linearity, then why not show it prominently in the data-sheet? The *average* HP/Agilent/Keysight 3458A has 0.02ppm [that's 20 parts per BILLION] INL--- better than *ANY* other 8.5-digit DMM out there. Why is this such a big deal? Well, for metrology [which is what you are likely doing with an 8.5-digit anything], a main technique for working with standards is to measure ratios. Every other function on an 8.5-digit DMM is marketing drivel-- and I have no expectation that their performance will be that good. The Datron 1281 and the Fluke 8508A are always used to show how good a DMM can be when measuring a resistor, and yet neither of them come even close to an old ESI 242D bridge or a direct current comparator [for lower resistances]. Not even *close*... People are always looking for a single tool to do everything, and just like with all-in-one stereo sets, you end up getting less than stellar results than you might if you purchased each component separately. If, [for some reason], you think you need an 8.5-digit DMM, the 3458A is *still* king of INL, after all of these years. The reason? It's analog ASIC. Matched JFET switches to ultra-high ratio stability highly interdigitated resistors. This, and balancing JFET switch transitions during an ADC cycle is the "secret sauce" that is at the heart of the 3458A's stellar INL spec. No one else has that. No one else ever will. *WAY* too expensive for even Keysight to redesign these days. Just is not going to happen...

Looking at the video measuring 1 G ohm I think they must have some very good (intelligent) filtering going on to get the display to be that stable. I do realize they are using a high voltage and coaxial cables to make the measurement but even a person breathing in the room would case more real fluctuation in the measurement. On Wed, Apr 22, 2015 at 7:36 AM, Ken Peek <ken.peek@diligentminds.com> wrote: > I'm a little bit suspicious of this one. Rumors have it that they use a > V/F converter as opposed to the "normal" multi-slope technique. This > may work just fine, and then again maybe not. I noticed that they write > about "high linearity" in their brochure, but no figures for ADC > linearity [either INL or DNL] are to be found anywhere on their > data-sheet or in their manual. If Transmille is so proud of their DMM's > linearity, then why not show it prominently in the data-sheet? > > The *average* HP/Agilent/Keysight 3458A has 0.02ppm [that's 20 parts per > BILLION] INL--- better than *ANY* other 8.5-digit DMM out there. Why is > this such a big deal? Well, for metrology [which is what you are likely > doing with an 8.5-digit anything], a main technique for working with > standards is to measure ratios. Every other function on an 8.5-digit > DMM is marketing drivel-- and I have no expectation that their > performance will be that good. > > The Datron 1281 and the Fluke 8508A are always used to show how good a > DMM can be when measuring a resistor, and yet neither of them come even > close to an old ESI 242D bridge or a direct current comparator [for > lower resistances]. Not even *close*... > > People are always looking for a single tool to do everything, and just > like with all-in-one stereo sets, you end up getting less than stellar > results than you might if you purchased each component separately. > > If, [for some reason], you think you need an 8.5-digit DMM, the 3458A is > *still* king of INL, after all of these years. The reason? It's analog > ASIC. Matched JFET switches to ultra-high ratio stability highly > interdigitated resistors. This, and balancing JFET switch transitions > during an ADC cycle is the "secret sauce" that is at the heart of the > 3458A's stellar INL spec. No one else has that. No one else ever will. > *WAY* too expensive for even Keysight to redesign these days. Just is > not going to happen... > > _______________________________________________ > 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. > -- *John Phillips*

-------- In message <1429713386.29372.25.camel@D2>, Ken Peek writes: >and balancing JFET switch transitions >during an ADC cycle is the "secret sauce" that is at the heart of the >3458A's stellar INL spec. No one else has that. No one else ever will. >*WAY* too expensive for even Keysight to redesign these days. Just is >not going to happen... Sorry to burst your bubble, but I'm pretty sure that's an integral feature of a number of ADC chips already. Probably something about a patent expiring... PS: There's a very instructive walk through the HP multislope in AOE3. Doesn't really go into the 3458 much, but they lay out how the subsequent versions in 34401/34972 do things. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence.