Getting into hot water

[(*steve*)]

Arriving at the office today I saw the following:



And I noticed a circuit board had been pulled out.



Hmmm, something is going on... Some subtle clues on the top side:



And some evidence of poor design and Hodges to fix it:



It looks to me that there is/was a high resistance connection at one end of the fuseholder.

If I catch the person servicing it I'll have a few questions. Although I don't expect the on-site repair Droid will be the component level repair Droid.

At the very least, there's a reason why the hot water regularly fails -- poor design

 

[(*steve*)]

After discussion with a colleague, we think the problem is caused because the clip for the fuse is soldered on one side only leaving the plated through holes to carry the full current. This explains the pattern of heating. The fact that the bodge wires are barely thicker than the fuse wire means it is probably of only marginal benefit.

 

[(*steve*)]

Aaah, the service person turned up. The board was binned, so I have un-binned it

The bodge wires appear to be under the conformal coating, so they're a factory mod. It's suggestive of a known issue.

Apparently this board costs several times the annual maintenance fee for these units, so you think it would be something to get on top of if it happens frequently.

The service guy said these units are very unreliable but didn't know if this fault is common (he also didn't know this unit has failed multiple times in the last 12 months).

Shrug.

 

[Rapidrob]

I have seen this many,many times in my working carrier. 9/10 time the dip solder job at the factory is FUBAR.
I would re-solder the connections or do as the fellow did and use a circuit board jumper. After I did this the board never failed again.
The high current draw on the loose fuse surly did not help the circuit but I bet the board is salvageable.

 

[(*steve*)]

Did you note that the wire jumpers were a factory mod. They clearly didn't fix the problem. However I concur that the soldering was likely the primary fault in this case.

 

[Tha fios agaibh]

service guy said these units are very unreliable but didn't know if this fault is common (he also didn't know this unit has failed multiple times in the last 12 months.

I agree it's poorly designed, but I've seen overheating like this from a fuse clip being stretched out too. I always give them a little squeeze before fuse replacement so they fit snugly.
More importantly, I always put a date on a replacement board so the service guy coming after me has a clue what's going on.

 

[shrtrnd]

There you go. The operative word is that you believe the repair guy intended to 'fix' the problem.
The repair guy is there to extend the life of the product. 'Fixing' the problem is what you engineers are
supposed to do in the design phase.
If you want to salvage the board for reuse when the next failure occurs, back-engineer it, and correct the
cause of the failure. It may be the wiring carrying the current is under-rated, but I'd be looking at upping the
current carrying capacity of the components connected to those wires.
When a product is designed, the manufacturer tries to integrate the cheapest parts possible for cost savings in
mass production. What will handle the current for a few years (before 'planned obsolescence' kicks-in), is not
always what would have done the job for extended service life. How is the manufacturer going to stay in business
if their product lasts 20 years, instead of 4 years?

 

[(*steve*)]

Actually, the person was representing a company that is paid a relatively small sum to keep it working indefinitely. A real fix for them is not having to repair it again (and they are not in a position to refuse to fix it)

 

[Harald Kapp]

A real fix for them is not having to repair it again

But the repair is charged to another account within the company. So it's no longer the designer's problem nor is it the problem of the manufacturing department.
Maybe "Total Cost Of Operation" is not yet part of their management's vocabulary.

 

[(*steve*)]

Oh yeah. I see what you mean. The maintenance company isn't related to the manufacturer. It may be in the interests of the manufacturer to sell more of these boards with the potential fault.

What I'm saying is that the maintenance company (who is losing the money) should be more interested in finding a fix. But perhaps one loss making device type among the (presumably) many they look after is less expensive than an investment in a qualified person's time.