(no subject)
I have the 3458A's set to ACSYNC (but there is not much difference with normal mode readings). With the 2703 set to 1.00000V at 1KHz, two of the three 3458A's read within +/- 20 uV. Same general error percentage at 10V and 100 V. I haven't done a lot of testing at odd voltages. I may try to get Swerleins's algorithm up and running on my GPIB computer. The other 3458A needs to be hardware calibrated... I'm waiting for the room temp to get a little warmer before dragging out the 732B and SR104.
Things start getting weird at 100 KHz. I found a one combination of lead/ground connections on one of the 3458A's where the reading does not change much between 1KHz and 100 KHz. Most combinations and the other 2 3458A's have around a 400 uV change between 1KHz and 100 KHz. The error at 10 KHz is not much different than 1 KHz/100 Hz.
There would be a problem fully calibrating the 2703 with the 3458A... the 3458A only goes to 700 VAC. The 2703 has a 1000V calibration step. I could fire up the Fluke 540B, but that thing is a royal PITA to use.
0.005% at what voltages and frequencies? (Of course 3458A ACSYNC specs are
worse than 0.005% at mid-band audio, although better than the 2701B's
specs.) You're setting your 3458A's to ACSYNC mode, right? ...and running
ACAL ALL first?
Well, 20ppm aggreement at 1kHz is indeed spectacular! IIRC, the 3458A ACSYNC
spec is approx. 92ppm at 1kHz 1V. In actual practice I've observed many
3458A's to be within 7ppm at 1kHz 1V. However I've seen some measure in the
range of 17 or 19ppm error. A bigger issue occurs at higher frequencies. At
20kHz some 3458A's look pretty good, while others might be off as much as
130ppm, depending on settling time after you first switch to the 1VAC range.
The 3458A's claim-to-fame for AC mostly relates to frequencies below 20kHz.
It degrades as we approach 100kHz. It's not surprising you see strange
differences of dependencies of combinations of lead/ground configurations at
100kHz according to different serial numbers of 3458A (same goes for
combinations of different serial numbers of AC calibrators and 3458A's).
Anyway, I can only hope my 2703B will perform as nicely as your unit. (But
I'll expect something worse, so as not to be too dissapointed!)
Greg
----- Original Message -----
From: "Mark Sims" holrum@hotmail.com
To: volt-nuts@febo.com
Sent: Friday, April 01, 2011 3:45 PM
Subject: [volt-nuts] (no subject)
I have the 3458A's set to ACSYNC (but there is not much difference with
normal mode readings). With the 2703 set to 1.00000V at 1KHz, two of the
three 3458A's read within +/- 20 uV. Same general error percentage at 10V
and 100 V. I haven't done a lot of testing at odd voltages. I may try to get
Swerleins's algorithm up and running on my GPIB computer. The other 3458A
needs to be hardware calibrated... I'm waiting for the room temp to get a
little warmer before dragging out the 732B and SR104.
Things start getting weird at 100 KHz. I found a one combination of
lead/ground connections on one of the 3458A's where the reading does not
change much between 1KHz and 100 KHz. Most combinations and the other 2
3458A's have around a 400 uV change between 1KHz and 100 KHz. The error at
10 KHz is not much different than 1 KHz/100 Hz.
There would be a problem fully calibrating the 2703 with the 3458A... the
3458A only goes to 700 VAC. The 2703 has a 1000V calibration step. I could
fire up the Fluke 540B, but that thing is a royal PITA to use.
0.005% at what voltages and frequencies? (Of course 3458A ACSYNC specs are
worse than 0.005% at mid-band audio, although better than the 2701B's
specs.) You're setting your 3458A's to ACSYNC mode, right? ...and running
ACAL ALL first?
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.