Do you really have a .01 PPM voltage reference? A personal JJ? That's taking this hobby to a new level. > > Daniel, > They are made like that... Problem is with drift. > When you cal a 3458A the first step is to short the inputs and wait for the > thermals to die. Then you do a Cal 0 and the incitement stores all the 0 > offsets for that set of terminals. they you switch to the other set and do it > again. > Next step is apply 10 votes to the terminals or in my case it is 9.9999411 and > enter cal 9.9999411 front and back set of terminals. remove the voltage and > plug in a 10k standard resistor and in my case enter cal 9999.884 for front and > back. In most cases AC does not have to be done. The meter is comparing its > measure values with the values entered and calculates the correction factor > to be used each time a value is displayed. > Most good meters now days do have a null function so you can look at drift > or compare 2 values.

no, this is work and it is not that accurate. that is just our best guess (about 1 ppm) of an old fluke 732B. We compare with our other 732Bs. Any available 3458As are monitoring the Flukes. Each hour a program picks up the Mean Max Min standard deviation from each meter average them together and plot the Mean for all the meters... Some times I have had 10 meters but sales have been up so I am down to 1 this week. I have about 5000 hours of data. Most 3458As come in within a few ppm of this standard. We send 3458As out to Agilent for repair and calibration then check our lab standards. Old standards and meters seem to be a lot more stable than new ones. On Wed, Aug 14, 2013 at 9:45 PM, Demian Martin <demianm_1@yahoo.com> wrote: > Do you really have a .01 PPM voltage reference? A personal JJ? That's > taking > this hobby to a new level. > > > > > Daniel, > > They are made like that... Problem is with drift. > > When you cal a 3458A the first step is to short the inputs and wait for > the > > thermals to die. Then you do a Cal 0 and the incitement stores all the 0 > > offsets for that set of terminals. they you switch to the other set and > do > it > > again. > > Next step is apply 10 votes to the terminals or in my case it is > 9.9999411 > and > > enter cal 9.9999411 front and back set of terminals. remove the voltage > and > > plug in a 10k standard resistor and in my case enter cal 9999.884 for > front and > > back. In most cases AC does not have to be done. The meter is comparing > its > > measure values with the values entered and calculates the correction > factor > > to be used each time a value is displayed. > > Most good meters now days do have a null function so you can look at > drift > > or compare 2 values. > > _______________________________________________ > 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

When I was looking to purchase a 3458A, I spoke to several people including folks at the Loveland Cal Lab, about whether to purchase 'used' or 'new'. The message was, as you say, 'old meters seem to be more stable than new' as explained by the years of 'aging' and letting everything 'stabilize'. Therefore, I purchased 'used'. Joe -----Original Message----- From: volt-nuts-bounces@febo.com [mailto:volt-nuts-bounces@febo.com] On Behalf Of John Phillips Sent: Thursday, August 15, 2013 1:19 AM To: Discussion of precise voltage measurement Subject: Re: [volt-nuts] Agilent calibration I have about 5000 hours of data. Most 3458As come in within a few ppm of this standard. We send 3458As out to Agilent for repair and calibration then check our lab standards. Old standards and meters seem to be a lot more stable than new ones.