Empathy List Archives

EB

ed breya

Wed, Jul 11, 2012 8:49 PM

That which is more fundamental to the problem is the unavoidable (at
room temperature) noise from the resistors. Even a "perfect" resistor
with zero tempco has noise, so if you use resistors to measure
current with a high-precision voltmeter, eventually you reach a
resolution where the noise becomes dominant. If you reduce the
bandwidth by averaging, the precision should reach the basic
stability of the resistor - but the (in)stability itself may be
viewed as noise of very low frequency - always too low to filter out
unless you have a very long time.

The resistors chosen for DMM current ranges may just be rational
choices and compromises for the types of performance to be expected
under normal usage and conditions, and considering the noise limits
to resolution, versus the cost of extremely low tempco sample resistors.

Ed

That which is more fundamental to the problem is the unavoidable (at room temperature) noise from the resistors. Even a "perfect" resistor with zero tempco has noise, so if you use resistors to measure current with a high-precision voltmeter, eventually you reach a resolution where the noise becomes dominant. If you reduce the bandwidth by averaging, the precision should reach the basic stability of the resistor - but the (in)stability itself may be viewed as noise of very low frequency - always too low to filter out unless you have a very long time. The resistors chosen for DMM current ranges may just be rational choices and compromises for the types of performance to be expected under normal usage and conditions, and considering the noise limits to resolution, versus the cost of extremely low tempco sample resistors. Ed

BS

Bob Smither

Wed, Jul 11, 2012 10:50 PM

On 07/11/2012 03:49 PM, ed breya wrote:

That which is more fundamental to the problem is the unavoidable (at room
temperature) noise from the resistors. Even a "perfect" resistor with zero
tempco has noise, so if you use resistors to measure current with a
high-precision voltmeter, eventually you reach a resolution where the noise
becomes dominant. If you reduce the bandwidth by averaging, the precision should
reach the basic stability of the resistor - but the (in)stability itself may be
viewed as noise of very low frequency - always too low to filter out unless you
have a very long time.

The resistors chosen for DMM current ranges may just be rational choices and
compromises for the types of performance to be expected under normal usage and
conditions, and considering the noise limits to resolution, versus the cost of
extremely low tempco sample resistors.

I may be off here, but I doubt that thermal (Johnson) noise would limit the
precision of current readings.

As an example - consider using a one ohm resistor to measure 1 mA. The voltage
corresponding to 1 mA is 1 mV. Johnson noise is given as:

en = sqrt(4kTRB)

or about

en ~ 0.13 * sqrt(R) nv / sqrt(Hz) at 300 K.

A one ohm resistor in a 1 Hz bandwidth would be 0.13 nV or 0.13 ppm of the 1 mV
reading.

Thermals in the circuit could easily reach uv levels as others have suggested.
Another limitation might be resistor excess noise - a 1/f noise source that can
be much larger than the thermal noise - but I would think that engineers at HP
would pick a resistor technology (bulk metal foil) that minimizes that noise source.

Best regards,

Bob Smither, PhD Circuit Concepts, Inc.

---=======
Friendly relations with all nations, entangling alliances with none.
--Thomas Jefferson

---=======
Smither@C-C-I.Com http://www.C-C-I.Com 281-331-2744(office) -4616(fax)

On 07/11/2012 03:49 PM, ed breya wrote: > That which is more fundamental to the problem is the unavoidable (at room > temperature) noise from the resistors. Even a "perfect" resistor with zero > tempco has noise, so if you use resistors to measure current with a > high-precision voltmeter, eventually you reach a resolution where the noise > becomes dominant. If you reduce the bandwidth by averaging, the precision should > reach the basic stability of the resistor - but the (in)stability itself may be > viewed as noise of very low frequency - always too low to filter out unless you > have a very long time. > > The resistors chosen for DMM current ranges may just be rational choices and > compromises for the types of performance to be expected under normal usage and > conditions, and considering the noise limits to resolution, versus the cost of > extremely low tempco sample resistors. I may be off here, but I doubt that thermal (Johnson) noise would limit the precision of current readings. As an example - consider using a one ohm resistor to measure 1 mA. The voltage corresponding to 1 mA is 1 mV. Johnson noise is given as: en = sqrt(4*k*T*R*B) or about en ~ 0.13 * sqrt(R) nv / sqrt(Hz) at 300 K. A one ohm resistor in a 1 Hz bandwidth would be 0.13 nV or 0.13 ppm of the 1 mV reading. Thermals in the circuit could easily reach uv levels as others have suggested. Another limitation might be resistor excess noise - a 1/f noise source that can be much larger than the thermal noise - but I would think that engineers at HP would pick a resistor technology (bulk metal foil) that minimizes that noise source. Best regards, -- Bob Smither, PhD Circuit Concepts, Inc. ========================================================================= Friendly relations with all nations, entangling alliances with none. --Thomas Jefferson ========================================================================= Smither@C-C-I.Com http://www.C-C-I.Com 281-331-2744(office) -4616(fax)

MS

Mike S

Thu, Jul 12, 2012 2:01 AM

On 7/11/2012 5:15 AM, Frank Stellmach wrote:

This is the worst realized electrical unit, i.e. the 'mise en pratique'
is difficult to an error level of about 1e-7 only.

On 7/11/2012 6:50 PM, Bob Smither wrote:

I may be off here, but I doubt that thermal (Johnson) noise would limit the
precision of current readings.
A one ohm resistor in a 1 Hz bandwidth would be 0.13 nV or 0.13 ppm of the 1 mV
reading.

It seems you just showed that it does (limit the precision).

--
Mike

On 7/11/2012 5:15 AM, Frank Stellmach wrote: > This is the worst realized electrical unit, i.e. the 'mise en pratique' > is difficult to an error level of about 1e-7 only. On 7/11/2012 6:50 PM, Bob Smither wrote: > I may be off here, but I doubt that thermal (Johnson) noise would limit the > precision of current readings. > A one ohm resistor in a 1 Hz bandwidth would be 0.13 nV or 0.13 ppm of the 1 mV > reading. It seems you just showed that it does (limit the precision). -- Mike

BS

Bob Smither

Thu, Jul 12, 2012 2:42 AM

On 07/11/2012 09:01 PM, Mike S wrote:

On 7/11/2012 5:15 AM, Frank Stellmach wrote:

This is the worst realized electrical unit, i.e. the 'mise en pratique'
is difficult to an error level of about 1e-7 only.

On 7/11/2012 6:50 PM, Bob Smither wrote:

I may be off here, but I doubt that thermal (Johnson) noise would limit the
precision of current readings.
A one ohm resistor in a 1 Hz bandwidth would be 0.13 nV or 0.13 ppm of the 1 mV
reading.

It seems you just showed that it does (limit the precision).

Hmmm - I guess at 8.5 digits theoretically yes. But the OP noted:

"But the 24h 100uA DC current accuracy is quoted as 10ppm Reading + 6ppm Range,
giving a maximum uncertainty of 16ppm when measuring 100uA. Why the big
discrepancy?"

so a 0.13 ppm error is pretty small compared to the stated DCA accuracy of the
3458A. My point was it seems like whatever is degrading the reading to 16 ppm
is not the thermal noise.

--
Bob Smither, PhD Circuit Concepts, Inc.

---=======
If a person has integrity, nothing else matters.
If a person doesn't have integrity, nothing else matters.

---=======
Smither@C-C-I.Com http://www.C-C-I.Com 281-331-2744(office) -4616(fax)

On 07/11/2012 09:01 PM, Mike S wrote: > On 7/11/2012 5:15 AM, Frank Stellmach wrote: >> This is the worst realized electrical unit, i.e. the 'mise en pratique' >> is difficult to an error level of about 1e-7 only. > > On 7/11/2012 6:50 PM, Bob Smither wrote: >> I may be off here, but I doubt that thermal (Johnson) noise would limit the >> precision of current readings. >> A one ohm resistor in a 1 Hz bandwidth would be 0.13 nV or 0.13 ppm of the 1 mV >> reading. > > It seems you just showed that it does (limit the precision). Hmmm - I guess at 8.5 digits theoretically yes. But the OP noted: "But the 24h 100uA DC current accuracy is quoted as 10ppm Reading + 6ppm Range, giving a maximum uncertainty of 16ppm when measuring 100uA. Why the big discrepancy?" so a 0.13 ppm error is pretty small compared to the stated DCA accuracy of the 3458A. My point was it seems like whatever is degrading the reading to 16 ppm is not the thermal noise. -- Bob Smither, PhD Circuit Concepts, Inc. ========================================================================= If a person has integrity, nothing else matters. If a person doesn't have integrity, nothing else matters. ========================================================================= Smither@C-C-I.Com http://www.C-C-I.Com 281-331-2744(office) -4616(fax)

MS

Mike S

Thu, Jul 12, 2012 3:15 AM

Apropos: http://www.sciencedaily.com/releases/2012/07/120711101042.htm

With regard to the discussion, since in the SI, Amps are the base unit,
Volts are derived from Amps, and Ohms from Volts - remarkable that Amps
are the least realizable in practice? (I guess it's just the difference
between NIST and everyone else)

--
Mike

Apropos: http://www.sciencedaily.com/releases/2012/07/120711101042.htm With regard to the discussion, since in the SI, Amps are the base unit, Volts are derived from Amps, and Ohms from Volts - remarkable that Amps are the least realizable in practice? (I guess it's just the difference between NIST and everyone else) -- Mike

W

Will

Fri, Jul 13, 2012 7:30 AM

The 1990 Josephson constant 2e/h is just an approximation which means
that the "Josephson Volt and Ohm" are not exactly the same as the SI
Volt and Ohm.

No practical quantum Ampere so far?

2012/7/12, Mike S mikes@flatsurface.com:

Apropos: http://www.sciencedaily.com/releases/2012/07/120711101042.htm

With regard to the discussion, since in the SI, Amps are the base unit,
Volts are derived from Amps, and Ohms from Volts - remarkable that Amps
are the least realizable in practice? (I guess it's just the difference
between NIST and everyone else)

--
Mike

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.

The 1990 Josephson constant 2e/h is just an approximation which means that the "Josephson Volt and Ohm" are not exactly the same as the SI Volt and Ohm. No practical quantum Ampere so far? 2012/7/12, Mike S <mikes@flatsurface.com>: > Apropos: http://www.sciencedaily.com/releases/2012/07/120711101042.htm > > With regard to the discussion, since in the SI, Amps are the base unit, > Volts are derived from Amps, and Ohms from Volts - remarkable that Amps > are the least realizable in practice? (I guess it's just the difference > between NIST and everyone else) > > -- > Mike > > > > _______________________________________________ > 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. >

PK

Poul-Henning Kamp

Fri, Jul 13, 2012 7:40 AM

In message CAE6XXrjKES2=BPCNyzOcbXFM73+4ztOTAkUEcTTaqTsDuvrJGw@mail.gmail.com
, Will writes:

No practical quantum Ampere so far?

It has always surprised me that the Ampere was not defined in terms of
electron charges per second...

--
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.

In message <CAE6XXrjKES2=BPCNyzOcbXFM73+4ztOTAkUEcTTaqTsDuvrJGw@mail.gmail.com> , Will writes: >No practical quantum Ampere so far? It has always surprised me that the Ampere was not defined in terms of electron charges per second... -- 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.

A

Artekmedia

Fri, Jul 13, 2012 9:35 AM

Coulombs per fortnight to be sure :-)

On 7/13/2012 2:40 AM, Poul-Henning Kamp wrote:

In messageCAE6XXrjKES2=BPCNyzOcbXFM73+4ztOTAkUEcTTaqTsDuvrJGw@mail.gmail.com
, Will writes:

No practical quantum Ampere so far?

It has always surprised me that the Ampere was not defined in terms of
electron charges per second...

--
Dave& Lynn Henderson
Manuals@ArtekMedia.com
www.Artekmedia.com
PO Box 175
Welch,MN 55089

Coulombs per fortnight to be sure :-) On 7/13/2012 2:40 AM, Poul-Henning Kamp wrote: > In message<CAE6XXrjKES2=BPCNyzOcbXFM73+4ztOTAkUEcTTaqTsDuvrJGw@mail.gmail.com> > , Will writes: > >> No practical quantum Ampere so far? > It has always surprised me that the Ampere was not defined in terms of > electron charges per second... > -- Dave& Lynn Henderson Manuals@ArtekMedia.com www.Artekmedia.com PO Box 175 Welch,MN 55089

PK

Poul-Henning Kamp

Fri, Jul 13, 2012 12:22 PM

In message 4FFFEBEC.8030802@embarqmail.com, Artekmedia writes:

Yes, I'd expect the last non-metric country in the world to go for that :-)

Coulombs per fortnight to be sure :-)

On 7/13/2012 2:40 AM, Poul-Henning Kamp wrote:

In messageCAE6XXrjKES2=BPCNyzOcbXFM73+4ztOTAkUEcTTaqTsDuvrJGw@mail.gmail.com
, Will writes:

No practical quantum Ampere so far?

It has always surprised me that the Ampere was not defined in terms of
electron charges per second...

--
Dave& Lynn Henderson
Manuals@ArtekMedia.com
www.Artekmedia.com
PO Box 175
Welch,MN 55089

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.

--
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.

In message <4FFFEBEC.8030802@embarqmail.com>, Artekmedia writes: Yes, I'd expect the last non-metric country in the world to go for that :-) >Coulombs per fortnight to be sure :-) > >On 7/13/2012 2:40 AM, Poul-Henning Kamp wrote: >> In message<CAE6XXrjKES2=BPCNyzOcbXFM73+4ztOTAkUEcTTaqTsDuvrJGw@mail.gmail.com> >> , Will writes: >> >>> No practical quantum Ampere so far? >> It has always surprised me that the Ampere was not defined in terms of >> electron charges per second... >> > >-- >Dave& Lynn Henderson >Manuals@ArtekMedia.com >www.Artekmedia.com >PO Box 175 >Welch,MN 55089 > > >_______________________________________________ >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. > -- 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.

S

Steve

Fri, Jul 13, 2012 1:09 PM

But what if a leap second occurs during the fortnight the standard is being set? ;)

Steve

On Jul 13, 2012, at 7:22 AM, "Poul-Henning Kamp" phk@phk.freebsd.dk wrote:

In message 4FFFEBEC.8030802@embarqmail.com, Artekmedia writes:

Yes, I'd expect the last non-metric country in the world to go for that :-)

Coulombs per fortnight to be sure :-)

On 7/13/2012 2:40 AM, Poul-Henning Kamp wrote:

In messageCAE6XXrjKES2=BPCNyzOcbXFM73+4ztOTAkUEcTTaqTsDuvrJGw@mail.gmail.com
, Will writes:

No practical quantum Ampere so far?

It has always surprised me that the Ampere was not defined in terms of
electron charges per second...

--
Dave& Lynn Henderson
Manuals@ArtekMedia.com
www.Artekmedia.com
PO Box 175
Welch,MN 55089

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.

--
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.

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.

But what if a leap second occurs during the fortnight the standard is being set? ;) Steve On Jul 13, 2012, at 7:22 AM, "Poul-Henning Kamp" <phk@phk.freebsd.dk> wrote: > In message <4FFFEBEC.8030802@embarqmail.com>, Artekmedia writes: > > Yes, I'd expect the last non-metric country in the world to go for that :-) > >> Coulombs per fortnight to be sure :-) >> >> On 7/13/2012 2:40 AM, Poul-Henning Kamp wrote: >>> In message<CAE6XXrjKES2=BPCNyzOcbXFM73+4ztOTAkUEcTTaqTsDuvrJGw@mail.gmail.com> >>> , Will writes: >>> >>>> No practical quantum Ampere so far? >>> It has always surprised me that the Ampere was not defined in terms of >>> electron charges per second... >>> >> >> -- >> Dave& Lynn Henderson >> Manuals@ArtekMedia.com >> www.Artekmedia.com >> PO Box 175 >> Welch,MN 55089 >> >> >> _______________________________________________ >> 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. >> > > -- > 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. > > _______________________________________________ > 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@lists.febo.com

Re: [volt-nuts] HP 3458A DC current accuracy

Best regards,