HP 3458a calibration (UK)

[IanJ]

Hi all,

I bought my 3458a from outside the UK just under a year ago and as such I have never used calibration services inside the UK.

Can anyone give me an idea as to the procedure and maybe costs involved with Keysight. I am in Scotland, and the nearest centre is Telford, England I think.

Before I go ahead and build batch no.3 of my Precision Voltage Source's I'd like to get my 3458a re-calibrated, specifically DCV calibrated only in order to keep cal costs down. Am hoping this is something that Keysight can accomodate.

Thanks,

Ian.

[Alex Nikitin]

Can anyone give me an idea as to the procedure and maybe costs involved with Keysight. I am in Scotland, and the nearest centre is Telford, England I think.

Before I go ahead and build batch no.3 of my Precision Voltage Source's I'd like to get my 3458a re-calibrated, specifically DCV calibrated only in order to keep cal costs down. Am hoping this is something that Keysight can accomodate.

Thanks,

Ian.

Hi Ian,

We've just calibrated our HP3458A Opt 002 at Keysight UK and also got a 3-year service/calibration agreement with them, the total cost was ~£1.5K + VAT for 3 years (including 3 yearly calibrations with uncertainties)*. The actual calibration is done at Wokingham, Berkshire but we've dropped the unit in person at Keysight UK Telford location (as a part of our company is located in Telford) and they use a regular shuttle service to deliver the equipment from Telford to Wokingham and back. Email to the UK Support (contactcentre_uk@keysight.com) and ask for a quotation. I've found it easy to deal with Keysight support.

Cheers

Alex

* - you can do it for one year only, it is a bit more expensive per year.

[IanJ]

Hi Ian,

We've just calibrated our HP3458A Opt 002 at Keysight UK and also got a 3-year service/calibration agreement with them, the total cost was ~£1.5K + VAT for 3 years (including 3 yearly calibrations with uncertainties)*. The actual calibration is done at Wokingham, Berkshire but we've dropped the unit in person at Keysight UK Telford location (as a part of our company is located in Telford) and they use a regular shuttle service to deliver the equipment from Telford to Wokingham and back. Email to the UK Support (contactcentre_uk@keysight.com) and ask for a quotation. I've found it easy to deal with Keysight support.

Cheers

Alex

* - you can do it for one year only, it is a bit more expensive per year.

Thanks......email sent. I'll update with how I get on, specifically costs due to my asking DCV only.

Ian.

[plesa]

Hi Ian,

do you have quote for DCV calibration? Maybe I'm wrong but the lab which can calibrate 3458A is only in Germany and in US.

[IanJ]

Hi Ian,

do you have quote for DCV calibration? Maybe I'm wrong but the lab which can calibrate 3458A is only in Germany and in US.

No reply yet! Email was sent on 11th.

Ian.

[alanambrose]

This is for business or personal? Don't know if it helps any, but I have a 732A I'm planning to get calibrated at NPL - I think their next session for DCV is March. If we could find a UK person with a SR104 with similar intentions we should be able to cal a lot of 3458As.

Alan

[ap]

When you have those two standards, what you can do only is adjustment, and limited validation (calibration), at these two values. All other validations are missing. While there is some guidance in the US (MIL use) that adjustment is sufficient every second time an item with artifact calibration needs adjustment/calibration, this is not necessarily good practice. So in the case of the initiator of this post, at least all DCV ranges should be verified.
Secondly, many cal labs use inadequate gear to do so, i.e. the uncertainties are too high. You cannot use a Fluke 5500 or 5700 to do that, as some labs do(!). Even a 5720/5730 is challenged. If one wants to do it properly, with low uncertainties, a lot of manual work is needed, and gear with low cal uncertainties. That is inconvenient of course, most labs these days being used to do some connections and run a software. E.g. validating the DCV range is best done with a 4910 DC standard (or similar; with low uncertainties not only for the 10V but also for 1V) for the 1V/10V range, and a very stable source for the other voltages, with 1kV being the least critical due to the high 3458A uncertainties in that range. So for 0.1V and 100V, e.g. a Fluke 752 and its low uncertainty, in combination with a stable source, may be used. Or a good low uncertainty DC calibrator. For DC current e.g., it is way more precise to use a set of precision shunt resistors and measure the shunt voltage (after referencing the meter to a DC standard) than just using a DCI calibrator. This is how we do it, but all this takes time. Still there are cal labs out there (many, I would say) that charge a lot of money, but do it the cheap way, and you may find out that you have a problem later. So when you have your 3458A calibrated, ask the right questions before you place the order.

[VintageNut]

When you have those two standards, what you can do only is adjustment, and limited validation (calibration), at these two values. All other validations are missing. While there is some guidance in the US (MIL use) that adjustment is sufficient every second time an item with artifact calibration needs adjustment/calibration, this is not necessarily good practice. So in the case of the initiator of this post, at least all DCV ranges should be verified.
Secondly, many cal labs use inadequate gear to do so, i.e. the uncertainties are too high. You cannot use a Fluke 5500 or 5700 to do that, as some labs do(!). Even a 5720/5730 is challenged. If one wants to do it properly, with low uncertainties, a lot of manual work is needed, and gear with low cal uncertainties. That is inconvenient of course, most labs these days being used to do some connections and run a software. E.g. validating the DCV range is best done with a 4910 DC standard (or similar; with low uncertainties not only for the 10V but also for 1V) for the 1V/10V range, and a very stable source for the other voltages, with 1kV being the least critical due to the high 3458A uncertainties in that range. So for 0.1V and 100V, e.g. a Fluke 752 and its low uncertainty, in combination with a stable source, may be used. Or a good low uncertainty DC calibrator. For DC current e.g., it is way more precise to use a set of precision shunt resistors and measure the shunt voltage (after referencing the meter to a DC standard) than just using a DCI calibrator. This is how we do it, but all this takes time. Still there are cal labs out there (many, I would say) that charge a lot of money, but do it the cheap way, and you may find out that you have a problem later. So when you have your 3458A calibrated, ask the right questions before you place the order.

Questions

1. What resistors do you use for this?
2. How are the resistors calibrated?
3. How often are the resistors calibrated?

I visited a cal lab recently. The resistors used are very old. A Measurements International bridge is used to calibrate the resistors. This is the first time I have seen one of these bridges in person.

[ap]

An SR104, a Leeds Northrop 4210 and hermetic metal foil resistors (temperature controlled/heated to better 0.1K) are used as references. Calibration is partly derived in a ladder calibration, using SR1010s. Calibration is once a year as a minimum.