Philips EL3542A trips RCD

[Stripyscarf]

Hello,

I'm pretty new to the world of vintage electronics and have an issue with a reel to reel tape recorder. I've been asked to try and fix a Philips EL3542A reel to reel that I was told last time it was used "it tripped the fuse".

I opened up the front and back to try and see if anything was obviously wrong, and I couldn't see any loose wires, dodgy looking capacitors, anything that had cleary got too hot or any other such damage. I checked the mechanics and everything appeared to work as it should (I'm told the belts had been replaced relatively recently). I then reassembled it and plugged it in via an RCD and a series lamp limiter.

Nothing tripped and the motor started up, so I played a tape. The tape played fine and sounded as good as can be expected from a 60 year old machine. After 15 minutes or so, suddenly everything stopped. It had tripped both the RCD I had plugged it in to and the RCD on the main house fuse box. However none of the MCBs had tripped and the 3A fuse in the plug didn't blow.

My guess is that something is heating up slowly over the 15 minutes and suddenly failing and causing a fault to earth. Also, I assume this is what happened last time "it tripped the fuse" so it isn't a component that completely broke last time. I know valves have heaters in, so could it be one of the valves that is faulty?

Any suggestions of what the fault could be would be greatly appreciated. Also, any ways of testing the faulty component without plugging it in again would be useful as my housemate were not impressed when I cut off electricity to the entire house!

Many thanks.

[paulsherwin]

This will not be a valve fault.

The RCD is tripping because something is causing mains live to leak to the earth connection. The most likely explanation is a faulty mains transformer. You are unlikely to be able to fix this without the appropriate skills and experience.

This is a safety critical fault, so take great care when power is applied.

[Stripyscarf]

Ah thanks for the info, a faulty transformer makes sense. What would cause the delay before the RCD trips, is it just the wires or core getting warm?
It does look like a difficult/expensive job to fix. so maybe it isn't worth it!

[ricard]

It is possible to get the mains transformer rewound, there are people on this forum who excel at this, whether it is worth it for such a machine is another question, probably cheaper to get another machine for spares and extract the mains transformer from that one, as it's unlikely to have the same fault.

[Lucien Nunes]

As Paul says the most likely suspect is the transformer, although it could also be the motor, or just possibly the mains wiring or power switch that physically made contact with the chassis or something electrically connected to the chassis as it warmed up. It might be readily localised using an electrical insulation resistance tester a.k.a. Megger, which applies a high voltage to subject the insulation to electrical stress, and measures the resistance while doing so. A similar insulation test function is available on a portable appliance test set. If it is a gross failure rather than just low insulation, it will show up on a simple resistance test on a multimeter from line and neutral to earth.

If the machine is to be warmed up to provoke the problem for investigation, it might be wise to use an isolating transformer so that it does not trip the RCD. As mentioned, being a fault that affectes user safety, it must be conclusively located, corrected, and then properly tested, again ideally with a high voltage insulation test, before considering it safe to return to service.

[Electronpusher0]

One possible reason that the RCD does not trip immediately is that there is some leakage current at switch on but the initial leakage current in the faulty component is not enough to trip the RCD but as it warms the leakage increases until it goes over the threshold for the RCD and trips it.
Definately a safety issue.

Peter

[emeritus]

A neutral to earth short can also trip an ELCB. I dicovered this many years ago when replacing an earthed fluorescent light in the shed powered from the 13A circuit via a fused spur from which I had removed the fuse. My wife discovered this only last week when attempting to dislodge a jammed piece of toast from the toaster, thereby managing to disconnect half of our electric circuits (split load consumer unit). She had thought that simply switching it off at the wall socket rather than unplugging it, would be ok, but evidently managed to connect part of the exposed element to the earthed case with the knife she was using.

[Lucien Nunes]

N-E faults can lie low, only causing an RCD to trip when sufficient load is switched on elsewhere (not necessarily on the same RCD) to create sufficient voltage drop in the supply N conductor to divert enough current through the fault to cause the trip. Therefore it is possible (although unlikely) that the tape recorder has a permanent hard N-E fault, but the trip was actually triggered by a separate event rather than it warming up.

Time to stop speculating and get some insulation readings. Stripyscarf, what resistance do you see with a multimeter between L+N and E?

[Michael Maurice]

Can Stripyscarf confirm that the mains plug is wired correctly and whether it has a two core or 3 core mains cable?

According to my diagrams the unit has a two core cable.

[wd40addict]

I have the Cossor CR1601 clone of this machine and mine does have an earth lead so presumably this feature was phased in during the production run as it's not on the circuits.

Even more confusingly the manual refers to Red-Black-Green mains cable whereas the cable actually fitted is Blue-White-Green/Yellow!

[Stripyscarf]

Thank you all for your replies!
I can confirm that it's fitted with a 3 core earthed cable and the plug is correctly fitted.
I've measured the L-E and N-E resistances with a multimeter and they're both greater than 2Mohms (the highest my 'meter can measure).
So from what's been said, I presume this means the insulation is failing after the device has warmed up?

[Station X]

A DMM is useless for measuring INSULATION resistance. You need a Megger or PAT Tester.

[trsomian]

I disagree slightly with Station X. A DMM is certainly useless for proving that insulation is good, but not necessarily totally useless for confirming that it is bad

[stacman]

Neutral to earth faults, described by John Ward electrical, makes some useful videos, and presented in an easy manner,
Worth a watch,
https://www.youtube.com/watch?v=JTkuoMplER8

Regards, Alan.

[Station X]

Originally Posted by trsomian

I disagree slightly with Station X. A DMM is certainly useless for proving that insulation is good, but not necessarily totally useless for confirming that it is bad

I wouldn't disagree with that, but if I were to qualify every statement I made, with every possible, if, but and and, I'd end up writing a book. I prefer to keep things simple, especially where newbies are concerned.

[Stripyscarf]

Yes, I know the multimeter doesn't prove the insulation is good, but I think it would have detected if there was a constant low current fault to earth and some other event added to that triggered the RCD as Lucien Nunes decribed. Unfortunatly that wasn't the case. Would I be right in thinking the only 'realistic' chance of repair would be a new transformer, if I can find one cheap?

[Michael Maurice]

You need to determine if the transformer is at fault or as has been said before, the motor.

Disconnect the motor and run the tape recorder, if it trips the transformer or something on the primary of the transformer has failed, it it doesn't trip then its likely to be the motor.