Altivar 71 Flashed at Power On

Three possibilities come to mind, incorrectly wired or incorrectly matched voltage or defective VSD....

I would rule out the wiring since the VSD came installed in the panel and the input was pre-wired. The only connections made to the VSD on site were the output phase wiring which was verified using phase sequence meter and continuity checking.

Regarding the incorrectly matched voltage. could you please elaborate?

I was speaking generally as I am not privy to the details of the installation...Was the voltage checked before install? Was the voltage being monitored? Is there surge protection upstream? If this is three phase, is there a delta leg, was it positioned correctly? Things like this and others that may effect the voltage input should be checked as part of the troubleshooting list if no other fault is found...

The voltage was not checked before switching on the supply to the machine. And there is no surge suppression installed on the VSD ( will re confirm this and revert). Moreover this brings on another point of discussion. The voltage input to the rectifier bridge is 220VAC phase to neutral, the rectifier is a 3 phase full bridge rectifier, following the formula of Vdc= 1.6554*Vpeak this turns out to be 364 volts. Now the capacitor in the filter circuit ( the one which flashed) is rated at 400V. A surge in the voltage could have brought about the flashing of the capacitor due to over voltage. Quite neck to neck design here considering there is no surge suppression at the rectifier in stream.

If you are feeding it 3 phase, your rectifier is working line-to-line, not line-to-neutral, so the L-L voltage would be 380V, the DC bus voltage would then be 535VDC. The caps are sometimes connected in series-parallel circuits, so the cap voltage on an individual unit may or may not be relevant.

If you are feeding it with 230V single phase, then your bus voltage would be 324VDC.

The ATV-71 drive has built-in MOV protection on the AC input side, as most modern drives do now. There is also a current limiting resistor on the DC bus that switches out after initial power up. It also likely has an EMC filter on the C side as well, which would have had some limiting effect to line spikes. I seriously doubt you just happened to connect it to the line at an exact moment that there was a severe enough spike to make it through all of that to cause the diode to blow out. Besides if that were the case, other devices connected to the same circuit would have blown as well.

Capacitor failure mode is often violent and destructive, a diode blowout is easily a result of the overall circuit damage.

Yes you should perform a reforming procedure on the other unpowered VFD. Do a web search on "VFD capacitor reforming" and you will get hits on many papers and articles explaining it. But basically, you must slowly increase the AC voltage over a period of hours. The best way is with a variac if you have one, otherwise you can do stages of lower voltage by using dropping resistors or transformers.

Two points, a 400 volt cap is far too low for 230VAC, it should be at least 1000VAC. A should all high voltage components, switching on may have caused a spike of well over 400 volts.

Also, did the lectrician use phase to phase instead of phase to neutral? Check carefully.

Just Had the same problem today. ALtivar 71, 600V, 3 phases. Flashed at power on. Huge spark at top of the Drive. I just openned it to look inside. All 3 lugs ( L1, L2, L3) are dark and burned. It has not been used for 2-3 months, in a shelter outside. Maybe it go too hot ( summer ). I suppose temperature went close to 50 degres celcius on a hot day...Kind of weird, use to work perfectly this spring

I have also similar issue ,6 altivar 71 VFDs are burned out from last april to this august 24th . regard less running or standby. VFD rating 380V to 480V ,system voltage 480V ,60HZ , no any input surge suppressor and contactor ,out put filter[choke] available . drive to motor distance 150 Meter to 350 Mtr ,the same time other VFDs in the same MCC tripped by mains over voltage .

Cherro,

You'll have to install protections on the altivar VFDs yourself since over-voltage protection on input is not installed as a standard option. I have had similar over-voltage incidents on other VFDs (ABB, AB) and they were safe due to the OV protection installed on them.

thanks for your replay

If the Drive has been sitting in the warehouse for some time, it is possible the Caps. may have been too dry.

I didnt understand what you wanted to say. Could you please elaborate a bit.

It's sat on the shelf for too long, as simple as….

The device has been in the warehouse for about 7 months. is that enough time to induce faults like settling of electrolyte in the capacitors? Moreover the VSD has a Semikron SKiiP 11NAB065V1 rectifier bridge and inverter bridge (IGBT based), now the module is a surface mount PCB on a heat sink and the contacts to the module are pads on the PCB which touch a connecting plane on cover type plastic cap. It is possible that due to transportation the connecting plane moved from its position causing loose connectivity which brought about the flashing of the diode. What do you think?

Hard to visualize your description of the assembly, but that seems unlikely. The VFD is going to be designed to be transported and even to be mounted to mobile equipment, I would seriously doubt a simplistic design flaw like that. Maybe an assembly flaw though, that happens. Usually that is discovered in testing. Whomever assembled this (you said it came pre-wired) may not have tested it.

To elaborate on the shelf life issue; capacitors with electrolytes can dry out if left un-powered for extended periods of time. If that happens, they fail rapidly, in a manner that pretty much matches what you described here. The length of time is generally considered 1 year, but that is not absolute. Plus, you know how long it was sitting on the shelf at your facility, but not how long Schneider sat on it before shipping it. You need to find the date code and determine when it was made to know how long it has been sitting unpowered. Just for future reference, there is a procedure called "reforming" of the capacitors that should be performed before powering up at full voltage to prevent this.

Thanks for the feedback JRaef. As you pointed out the device has been in our warehouse since September and the normal shipment time ex works excluded is around 3 months which makes the total idle time around an year. This means i should get the electrolyte reformed for the other VSD as well. Theoretically speaking the Flashing of the capacitor should cause the diode on the rectifier bridge to flash and not the other way around. Am i right?