An arduino unrelated question regarding a recent rectifier failure

Hi all,

My treadmill developed a fault the other day.

The main PC board would have to be replaced at a cost of 300 euros!

Luckily i was able to pin point the fault on a bridge rectifier which was then replaced forjust a few euros.
Nothing else was blown which brings the question, why has this heavy duty 800V/40A rectifier failed especially considering the machine is less than a year old with minimal use?

The rectifier operates far below its ratings and attached on a large heat sink. After replacement I was able to measure the temperature on the heatsink at about 45 Deg C.

I know this is more of an academic kinda question but still I d love to hear the electronics' community views.

Voltage spikes on the mains perhaps - an inductive load being switched can in bad cases
put kV transients on mains wiring (old fluorescent lighting fittings and old fridges are
especially bad for this, and industrial motors etc)

Lightning strikes are also able to do this.

Opto-isolation for use with mains is normally rated in the 3kV to 5kV range to prevent flashover
during such spikes, so 800V rectifier is actually a weak point if its on the primary side of the
SMPS. Avalanche breakdown of a device connected to the mains will generate current spikes
up in the 100's of amps range, ie 10's of kW are dissipated, vaporizing any semiconductors
or bond-wires.

Perhaps there are protection devices like MOVs that have also failed on the board - I'd suggest
checking for such devices are also replace them (as it will be hard to check they are fully
intact).

Thanks MarkT. There was only only MOV but there is also another metal sealed "box" before the pcb which says 'filter " on it. Maybe EMI filter?

the machine is less than a year old

Then it should be covered by law and should not have cost you anything to repair. It would be in the UK.

Grumpy_Mike:
Then it should be covered by law and should not have cost you anything to repair. It would be in the UK.

Correct.
Truth is , it has less than a year of use while it was boight just over two yeqrs ago. It was sitting idle, unplugged for more than a year.

Could have been a spike or any of several things. I would not rule out a part with a defect at manufacture time which finally showed up.

Ron