Hi, The first question is, what do you need to power that's 2KW? And what ever that is, do you need that much accuracy at 2KW! :) Having a couple of 1.2KW power supplies on the bench as I type this, I tend to agree that that level of accuracy probably isn't needed. If this were my supply, a simple check to the limit of my DMM would be all I would do. You should be able to extrapolate 2 or 3 times to a more reasonable current level. Short of that, look for a good current shunt. Yes, load testing a supply to 2KW isn't trivial. My recommendation is a whole bunch of incandescent light bulbs. A120V bulb if you can find them, work reasonably well at 70V. We get them here for pretty cheap, and you can solder solid copper wire onto them to make a big array... Good luck on the supply! Dan > I have an HP 6674A power supply with option J06, which means that the PSU > is 70 V @ 30 A rather than the standard 60 V @ 35 A . > > I've replaced a couple of bits in this and will need to replace some more, > so it would be prudent to get this calibrated. > > The service manual calls for a 8.5 digit 3458A, which seems a bit over the > top given the displays for voltage and current are only 4 digits. I think > my 6.5 digit 3457A will be good enough. This is a 2.1 kW PSU, not a > precision measuring instrument. But I don't have any ammeter that can read > 30 A, so I can not just use an external ammeter if I want to know the > current. > > The service manual also calls for a 0.010 Ohm 0.04 % 100 W resistor, with a > recommended resistor of a Guildline 9230/100. I am wondering how practical > it is to make such a resistor and verify its performance on the 3457A. > Some versions of this PSU have a lower output voltage (56 V) but higher > current (42 A). With the 30 A PSU I have, the maximum power that could be > dissipated in a 0.01 Ohm resistor is obviously 9 W. > > Any suggestions about what I can use that will not cost a fortune? > > Looking on eBay, Guildline resistors are several hundred USD each. I can't > justify that given the cost of a Keysight calibration of the PSU is $199 in > the USA. > > I much admit that I have never really much attention to calibration of a > PSU. It seems a bit of a waste of money when you can use a decent > multimeter if you really want to know the voltage or current. But with a > PSU of this size, I don't have an ammeter good enough. > > The other tricky bit about calibration of this PSU is the need for loading > it to full power (2.1 kW) then dropping the load to 50% and measuring the > recovery time. It should recover to within 100 mV in 900 us. Likewise it > should recover the same if switched from 50% of load to 100 %. The manual > calls for an electronic load, but I suspect a FET switch and some big > resistors in water will do. I don't see any need for such resistors needing > to be very high precision, but obviously something decent is needed to > calibrate the ammeter. > > Dave.