[volt-nuts] What's the probability of a random used 3458A passing a Keysight calibration?

[Dr. David Kirkby]

On 16 January 2018 at 05:01, John Phillips <john.phillips0 at gmail.com> wrote:

> my experience is that most of the eBay meters that do not give errors are
> very close to spec. These old meters do not drift as much as a new meter.
> If you have a good 10 volts and 10k resistor calibration is a snap...
> verifying cal in not as easy. The high-frequency AC cal is more difficult.
>

Clearly in the case of the eBay item, it was the high-frequency AC volts
that was out of spec. I would imagine a number of labs that may have good
enough DC and resistance standards, may well not have good enough high
frequency AC standards for this. So maybe that meter would pass at some
other labs, who have higher uncertainties than Keysight.

Anyway, not that I can afford a 3458A, but I added up the cost of the
meter, plus the repair cost, and found it was was not much below the cost
of a new meter. I can appreciate your point about a new meter drifting
more, but I can also imagine that some of the caps in those old meters
might start to show problems.  Maybe an Agilent meter might be a sweet spot
- not as old as an HP, so but less stable than a newer Keysight.

Dave