Plug in, switch on, blow up
I had occasion to pull a little used piece of test kit off the shelf recently, and rather than the now expected reliability of a piece of well ‘burnt-in’ kit, all I got was a WHOOSH and a cloud of smoke after plugging it in and switching it on.
As background to this exciting moment, I should mention the purchase of a cheap infra-red hand-held thermometer from Lidl. Once again, this proved to be a mistake, it didn’t work, nor did a replacement. Both had the same fault, as the reading started out ok, but just steadily ramped down as long as the unit was being used. The falling reading obviously rendered it useless, and at about 1°C per minute, would eventually go negative even when presented with targets at room temperature.
Although the problem was obvious, I wanted to carry out some proper checks, so dug out a temperature calibration bath, which would allow me to create a target with a known and stable temperature to test the Lidl device against:
And that was when the fun started, as a cloud of smoke poured out of the bath when I applied power – made all the more exciting by the fan to force cool the interior.
Although I killed the power almost instantly, I noted that despite this cloud, everything appeared to be working, and no fuses blew.
After checking nothing obvious was damaged, and no wiring had been burnt I decided to briefly power it up again. Partly because I could see no damage or failure, and partly because it smelt terrible, but the fan cooling meant this smell could not be localised – EVERYTHING smelled of burnt electronics.
No drama and no more smoke, and it was working perfectly – a puzzle? I knew something had been burning and smoking, so deeper digging was needed.
The only candidate remaining was the temperature controller fitted to the bath, but it was working normally – so it had to come out and be dismantled. Initially, it looked fine, with no sign of anything having burnt or been smoking. However, when I examined the components on the controller PCBs, the problem began to give itself away as I noted that some of the capacitors had cracks in their casings, and this betrayed the fact that they had swollen.
All the same make and type, once removed and examined it was clear that they had failed when mains had been applied, as the insulation within them vaporised, expanding and bursting their casings as the gas made its way to the outside world. The pics below reveal the evidence:
Fortunately, this was not a major problem, just a nuisance.
Lacking a circuit diagram for the controller, a look at the PCB and the fact that these caps had mains applied to them suggested they were suppression or filter components, intended to deal with electrical interference and/or prevent it, rather than act as part of the active controller circuit.
A rummage around the junk box produced some reasonably similar cap of the right voltage, so this functionality could be preserved.
Test runs of 12 hours and more confirmed the bath operated normally.
As for the Lidl infra-red thermometer?
This little gem will star as a later entry in my ‘Lidl technical problems’ series.
No comments yet.