Electronics > Repair

Repairing a very old Saftronics Inverter


This is kind of a very long shot in the dark. I have an old Saftronics PC3 inverter module that I was asked to try and repair. It's around 20 years old (maybe older) being part of another old machine, and some parts are practically non-existent. I can't find a good reference for it. It may be a rebranded product, but it kind of looks like what is shown in the user manual for the Yaskawa brand (if this helps at all). Our module has no digital display that seems to appear on similar modules when you Google it, only LED indicators to show its status. Sorry, I can't really provide more info about it.

According to what people have said about the module, it first stopped working with the LEDs indicating an overheating issue. There was a small DC brushless fan attached to it to cool down the large heat sink attached to it, but the fan stopped working. The fan was repaired and was working again, but the module still showed an error. I don't know how long the period was from when the fan was broken to when it was repaired (this was months ago so details are sparse), but we hypothesize that within that time frame, the excess heat may have caused another component to break. One component that came to mind was what the heatsink was attached to: a 7MBR30NF060 IGBT Module from Fuji Electric. This module was attached to the heatsink via epoxy (along with a rectangular piece of metal). Doing some research on it, the IGBT module is obsolete. Again, this is a long shot, but are there any other options that are recommended to try/replace before spending lots of money on one part that may or may not work? I'm just at a loss on how to proceed, other than replacing it.

If the dc bus terminals are brought out, start by doing a diode check on the input bridge (input terminals to dc bus) and the output IGBT’s (dc bus to output terminals)

Shorted power devices are by far the most common problem followed by bus capacitors then caps in the gate drivers failing. Once the gate drivers get soft, the IGBT’s will fail (eventually).

Thanks for your reply. When you say "brought out", do you mean burnt out?

Yeah, that was my oversight, assuming that the IGBT module is bad before even testing it. It's quite difficult to unsolder completely, but doing what I could, I used a diode check on the IGBT module given its schematic. They all seem to work properly given the diode orientation and the pin placement. I did the same with the relay that was on the board, checking the coil resistance and making sure it performs at its rated voltage. That passed too. The screw terminals don't look burnt at all.

Since the module and relay showcased expected readings when tested, I guess the next step is to check the transistors and the polarized caps. Before knowing more about what happened to the device, my first thought was to swap out the transistors and capacitors, at least the ones I can tell which part it is. Without schematics, this is all a long shot. This isn't something that can be easily tested on the fly after all.

If I may ask, do you have experience with this model of inverter? Since this has a digital display variant and a simple LED indicator variant, I was wondering if it would be possible/advisable to transfer over the digital display from one model and put it on this broken one to get a better indicator of what kind of error this thing is facing? LED indicators indicate four possible faults while the digital display can narrow it down a bit more.

According to the LED indicators, it's either an overheating issue or an overload issue with the motor/inverter, or overtorque has been detected. I wasn't given much information on what the circumstances were (though I do wonder....since the transistors seem to be fine and capacitors seem to be fine from what I've tested, is it possible that the fault indicator was never resetted; though, since the power was most likely turned off to replace the fan, it most likely was reset in that manner)?

~~~Actually, upon further inspection, I think I have found some possibly bad diodes external to the IGBT module. Diode tester shows that they're bad (at least from what I can pick up; all the coating on the board is making it really hard to test). Perhaps it's like you said; I didn't consider it because it wasn't connected to the module. If they're cheap, it'll be easy to replace. The only issue is trying to find the right diode. Markings don't indicate anything, but at least I got something close to the package size.


[0] Message Index

There was an error while thanking
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod