"Reforming" VFD Capacitors

I'v only had one electrolitic go bang. I look after the PA system in our church. Nearly 30 years ago we installed a 100 Amp Philips unit, and half way through the first church service with our nice new amplifier going very well there was a muffled bang from the room were the equipment was kept, and then silence, went out to have a look, and the room was full of white smoke. Took the amplifier home and found a 2000uf capacitor had blown, and when I took it out found it had been installed back to front, no wonder it went bang. New capacitor, and the system worked well, and still would except that someone decided we needed to up date all the gear, at least it hasn't blown up---yet.

Ian S C

So what is the way to reform a set of VFD caps? Simply hook up to a variac and slowly up the volts to mains plus over an hour or so?

I must admit I am a bit askance at having to reform the caps after a year. My electrolytic caps are fine, even after a few years in storage, are VFD caps some special sort of aluminium oxide caps, as they appear to be somewhat delicate?

Andrew.

This is a nice reminder to recondition the capacitors in my old home-made hifi amp, they haven't been used for nearly 35 years...

A diode might be advisable

Hello Bazyle,

I was simply thinking of connecting the variac to the input of the VFD. That way you have the VFD rectifiers in circuit before the caps.

Andrew.

I arrived at a breakdown one day to find an 11kW inverter had tripped its supply breaker, I turned it back on to be be greeted by an almighty bang, I must have jumped 4ft backwards and a cloud of dust and smoke came out of the panel, the bang was ear splittingly loud. I managed to avoid a change of laundry but resolved that it would have to be the apprentice who turned any tripped drives back on in the future. Luckily I have a good sense of humour and am able to laugh when this story is recalled by those who witnessed it and just about everyone else as it did get around the shop.

Mike

SEW recommend just powering up on the mains at a maximum of 2 years, if this is not done then they recommend the procedure I posted above. Bear in mind that these caps are sitting on over 700V DC on the DC link of a 415V input inverter and this can reach 900V when braking a load. Any weakness in the caps is going to be exposed with a bang.

Mike

The best way to keep capacitors in good order is simply to use them regularly. The need to reform capacitors occurs only after they've not been used for years. Beware if you're the kind of chap who buys bargains. That second-hand VFD may not have been powered up in decades.

The insulation inside an unused electrolytic capacitor tends to weaken with time. Periodic application of power reforms the insulating layer. When a supplier recommends regular power-ups he is ensuring that the insulation is kept thick. It's not that his capacitors only last a few years, it's that maintenance much reduces the chance of a failure later on.

Assuming you've been negligent or have inherited suspect capacitors one way of reforming them is to power-up the host equipment via a Variac. The procedure is to apply a low voltage for about 20 minutes, then increase it in 3 or 4 stages to full voltage. If the capacitor warms up, replace it. I wouldn't recommend it on a VFD - semiconductor circuits dislike being fed low-voltages in this way and you might damage something else.

Better to remove the capacitors and wire them up as shown in the circuit below.

The 40W (or smaller) bulb limits the amount of current that can flow especially if the capacitor fails. When switched on the bulb should flash briefly, then dim rapidly down and go out. At that point leave the capacitor powered up for half an hour. If the capacitor warms up or the bulb glows continually, stop immediately and buy a new capacitor. Of course if you've gone the trouble of removing old capacitors in the first place you might just as well replace them without playing electrical Русская рулетка!

Reforming capacitors in old valve radios is often done in-situ with either the variac or bulb method. It's a special case. The intent is to restore the radio if possible using it's original components for heritage reasons. Be aware that old-fashioned valve radios are more resiliant to this particular electrical mistreatment than modern stuff.

Usual warnings about the circuit and how it's used. Think electrocution, burns and a significant chance the capacitor might explode. Mains electricity can kill. Please don't try it unless you understand what you're doing.

Dave

Here we have the info kindly supplied by Teco after my recent enquiry.

I still await a reply from Fuji

Bill

What happens when a capacitor vents and then ....

Admittedly a somewhat larger capacitor than any likely to be used in a home-workshop VFD – have a look at the debris spread Fig. 3 on page 13 of the report.

https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/691440/R052018_180320_Guildford.pdf

Edited By Dave Martin on 22/03/2018 22:25:38

Thanks for publishing that link Dave. An interesting read with my breakfast coffee.

Previously I thought that film capacitors either self-healed or failed quietly after overheating . Never occurred to me that the plastic film might decompose into inflammable gases that can go bang after mixing with air inside a sealed box.

Read the report and weep if you're the type who thinks Health and Safety has gone mad. Despite elaborate precautions:

• During manufacture an employee made an unauthorised shop-floor change to a winding machine resulting in a faulty batch of capacitors.
• A salesman persuaded the customer's engineers that a pressure switch (commonly fitted to this type of capacitor) was unnecessary.
• Not discussed in the report, but after a number of failures under test, someone decided that the faulty batch could still be used provided individual units passed more rigorous soak testing. The extra test was inadequate.
• The capacitor's risk assessment calculated necessary mitigations based on the extremely low failure rate of correctly manufactured capacitors. The responsible committee did not reconsider when it became known that the faulty batch of capacitors were failing rapidly.

Looks like two junior employees have been thrown to the lions for this. Is it just a coincidence that the engineering and management teams came out of this squeaky clean? As Railway quality 11mF capacitors must be expensive to write-off, who decided to put them into service despite a sequence of failures during testing? We will never know; whoever it was hasn't accounted for the error.

There's no CE mark on the Capacitor Label.

Dave

Edited By SillyOldDuffer on 23/03/2018 10:54:33