What are 6.5 Digit Multimeters Good For?
in [Design]
|
From: Dave Platt on 12 Mar 2010 00:36 In article <MPG.260350d8fa1b2c289896e7(a)209.197.12.12>, D from BC <myrealaddress(a)comic.com> wrote: >32 ADC >http://focus.ti.com/lit/ds/symlink/ads1282.pdf >$77 at Digikey >I think that's 10 digit. 2^32 = 4294967296 >Say full scale is 1V then 1st step is 232picovolts. >uhuh.. >I get 1mV fuzz just by shorting out my scope probe! *chuckle* Some years ago, a guy in one of the audio forums suggested that audio really ought to be recorded and delivered using 32-bit PCM, "just to make sure." I pointed out that if one were to set the playback amplitude so that a least-significant-bit signal produced audio at the listening position equal in power to the random thermal noise of air molecules hitting the eardrum, a full-scale 32-bit signal would vaporize the power lines, explode the building and kill the listener :-) -- Dave Platt <dplatt(a)radagast.org> AE6EO Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior I do _not_ wish to receive unsolicited commercial email, and I will boycott any company which has the gall to send me such ads!
From: D from BC on 12 Mar 2010 02:39 In article <hncf6d$q8d$1(a)news.eternal-september.org>, Jon_Slaughter(a)Hotmail.com says... > > D from BC wrote: > > In article <hnc95i$82b$1(a)news.eternal-september.org>, > > Jon_Slaughter(a)Hotmail.com says... > >> > >> John Larkin wrote: > >>> On Thu, 11 Mar 2010 17:56:35 -0800, D from BC > >>> <myrealaddress(a)comic.com> wrote: > >>> > >>>> 6.5 digit multimeters sell around $1000.00. > >>>> For electronics development, are these $1000 multimeters really > >>>> necessary? > >>>> What are they good for? > >>> > >>> Measuring to PPM accuracy, and measuring microvolt-level voltages. > >>> And as a traceable standard for calibrating products. > >>> > >>> The Fluke 8845A is excellent. > >>> > >>> John > >> > >> What's the big deal? Can't they just switch to 24-bit ADC's on the > >> cheap and get the accuracy? Or is there some special techniques > >> required to get that resolution in practice? > > > > It's probably trouble to push the noise floor down with the front end > > electronics in the multimeter. > > I read that this is not difficult by paralleling ADC's or using averaging. > Since the noise generally is symmetric(generally gaussian) by averaging it > will cancel out. > > I guess it is more important that the ADC be stable but as long as any > fluxuations are symmetric(such as clock jitter) then averaging should take > care of most of the problems. > > > 32 ADC > > http://focus.ti.com/lit/ds/symlink/ads1282.pdf > > $77 at Digikey > > I think that's 10 digit. 2^32 = 4294967296 > > Say full scale is 1V then 1st step is 232picovolts. > > uhuh.. > > I get 1mV fuzz just by shorting out my scope probe! > > Yes, but maybe that is due to the scope not using that 32-bit ADC? ;) True.. My scope has 8 bit voltage resolution. The point I was trying to get across was if an 8 bit system shows it's noise floor then it's probably gets increasingly difficult to control noise as sytem resolution increases(16,24,32).
From: D from BC on 12 Mar 2010 02:42 In article <hncf6d$q8d$1(a)news.eternal-september.org>, Jon_Slaughter(a)Hotmail.com says... > > D from BC wrote: > > In article <hnc95i$82b$1(a)news.eternal-september.org>, > > Jon_Slaughter(a)Hotmail.com says... > >> > >> John Larkin wrote: > >>> On Thu, 11 Mar 2010 17:56:35 -0800, D from BC > >>> <myrealaddress(a)comic.com> wrote: > >>> > >>>> 6.5 digit multimeters sell around $1000.00. > >>>> For electronics development, are these $1000 multimeters really > >>>> necessary? > >>>> What are they good for? > >>> > >>> Measuring to PPM accuracy, and measuring microvolt-level voltages. > >>> And as a traceable standard for calibrating products. > >>> > >>> The Fluke 8845A is excellent. > >>> > >>> John > >> > >> What's the big deal? Can't they just switch to 24-bit ADC's on the > >> cheap and get the accuracy? Or is there some special techniques > >> required to get that resolution in practice? > > > > It's probably trouble to push the noise floor down with the front end > > electronics in the multimeter. > > I read that this is not difficult by paralleling ADC's or using averaging. > Since the noise generally is symmetric(generally gaussian) by averaging it > will cancel out. > > I guess it is more important that the ADC be stable but as long as any > fluxuations are symmetric(such as clock jitter) then averaging should take > care of most of the problems. That's seems like the averaging feature on oscilloscopes... Given time, the signal cleans up nice.
From: Archimedes' Lever on 12 Mar 2010 02:50 On Thu, 11 Mar 2010 19:11:54 -0800, D from BC <myrealaddress(a)comic.com> wrote: >I get 1mV fuzz just by shorting out my scope probe! But have no clue as to why.
From: Archimedes' Lever on 12 Mar 2010 02:55
On Thu, 11 Mar 2010 18:53:32 -0800, D from BC <myrealaddress(a)comic.com> wrote: >In article <a69jp59q5nvmomn6rqobm4q4a3m0a4088k(a)4ax.com>, >jjlarkin(a)highNOTlandTHIStechnologyPART.com says... >> >> On Thu, 11 Mar 2010 17:56:35 -0800, D from BC >> <myrealaddress(a)comic.com> wrote: >> >> >6.5 digit multimeters sell around $1000.00. >> >For electronics development, are these $1000 multimeters really >> >necessary? >> >What are they good for? >> >> Measuring to PPM accuracy, and measuring microvolt-level voltages. And >> as a traceable standard for calibrating products. >> >> The Fluke 8845A is excellent. >> >> John > >Ah.. the tool that sets the tools. > >Of course there's always something better.. :P >8.5 Digit multimeter >http://www.home.agilent.com/agilent/product.jspx?nid=- >536902435.536881781.00&cc=US&lc=eng Try Keithley http://www.keithley.com/products/dcac/dmm/highper http://www.keithley.com/products/dcac/sensitive/lowvoltage/?mn=2182A |