Keithley 2430 Sourcemeter dead

[JohnPi]

I have a Keithley 2430 Sourcemeter that appears dead -- it doesn't turn on -- no fan or any other activity.

I checked the AC input with a variac -- fuse is OK, and when the AC voltage is over about 80 V, it begins to draw current (e.g. bridge rectifier ?). There are no obvious burn marks on the board


I can't find any documents on the power supply for this (or any) Sourcemeter, so I don't know where to begin probing.

Does anyone have a block diagram > component locator ? test point measurements etc. for these Sourcemeters ?

[TiN]

Probably Vicor DC/DC converter is dead. It's that thingy under the big heatsink.
I've repaired 2425 that had exactly that problem. $270 dollars later and it's all working happy again.

[JohnPi]

Thanks -- do you have a part number or source for the Vicor ?

[TiN]

Draft article is here. I don't know if your 2430 is using same module or not, so I'd suggest checking that before you buy replacement. Check input and output voltages first, maybe something else is bad in your unit and module is fine.

[JohnPi]

Thanks. I removed and tested the Vicor. It doesn't output 24 V, and it appears the input is shorted.

In the xDevs video, they supplied an external 24 V to the power supply board -- it also appears shorted. A closer look showed damage at the inverter FET - IRFI540N -- exactly as in their video too. Removing it removed the short, and then applying 24 V makes the system 'beep' and only consume ~ 400 mA @ 24 V, although the display doesn't come on.

Have some new IRFI540N on order. If the replacement works, I'll get a Vicor VI-263-EU/F3 module: 300V@250 W -> 24V@200W.

[TiN]

Check if VFD glass is ok. Maybe it's also physically damaged. Easy to see by color of getter.

[JohnPi]

That seems OK now. Once I get the replacement FET, I'll be able to power up the whole system from an external 24 V supply, then confirm functionality before replacing the Vicor.

[JohnPi]

I was able to replace the FET; it turns out the Vicor wasn't dead. Now the Source meter powers up and the display and buttons work. However when I enable the output, it doesn't turn on, but indicates "OVP". Any ideas ?

[sean0118]

However when I enable the output, it doesn't turn on, but indicates "OVP". Any ideas ?

That might be normal if it's set to source a constant current and there's no load connected? The output will float up until it hits the programmed output over voltage protection level.

[JohnPi]

Even in V mode, it generates OVP.

While I don't think it's the issue, I'm, trying to upgrade the firmware from C20 to C34 -- after some trouble finding the Keithley Flash Wizard app (and it won't work with a new USB comm port  -- hint -- use a USB-RS232 with a different chipset), it repeatedly gives a S-Record checksum error (even on different FW revs downloaded from Tek.com). It is possible that I can't directly upgrade from C20 to C34 (or C32 also on the site), but I can't find any in-between versions.

[Kleinstein]

The overvoltage protection limit may have to be set. There may be a corruption in the default / last settings and a wrong set OVP limit may trigger early.
I had such a problem with a lab supply. Corrupted settings may happen when turned off just at the wrong moment.

Otherwise one could test reading back the voltage with the ouput not enabled. Chances are the path for reading the votlage would initially be the same as for the OVP check.

The other parts to check are the output power FETs / transisstors, as natural candidates from failure.

[TiN]

That might be normal if it's set to source a constant current and there's no load connected? The output will float up until it hits the programmed output over voltage protection level.

That's not how these SMUs work, there should not be limiters active in default configuration. If 24** hits compliance max limit it just blinks that on 2nd line display with setting.