Premature pump death (Autopsy)
 

It (Hydor Selz S30) only managed about 2 months continuous service. Little piece of fscking sh*t :mad:
Of course, it waited for me to void the warranty by cutting off the molded plug before it decided to pack in. Luckily it didn’t take my PC with it: ) (suspect it would have done, had I left it on - despite shutdown/f, prolly would have locked up or something first :shrug: )

So after removing the pump from the loop, I decided to investigate further. To start with, it was not humming at all, so I suspected burnt out motor winding.
On close examination, I thought I could see some fine cracks on the back of pump body, so I decided to open it up, hoping to find wet corroded motor coil. I got a faint whiff of burnt "something" as well, maybe just imagining this though. A bit of sweating and cursing with a cutter, and all is revealed. Not.

Next test, multimeter….
3.8MOhms = nearly open circuit. Maybe water in there? Check if theres any electrolysis going on.
Volt meter reads 32mV , dropping slowly. Guess that’s that then.
I might do some more dissecting if I get a minute, to confirm my suspicions...

Conclusion: I think it's buggered because some water has managed to get inside, probably through the cable exit (even though its all sealed in resin). Possibly the water ate away at the connection between the supply lead and the winding, leading to an open circuit.
So that would mean I'm also lucky not to have died from electrocution the numerous times I've had my hand in the water at the same time…hehehe whoops. IP68 does mean it will run submerged, huh?

Anyway system is now running pumpless, bucket of water sitting nearly on top of case…at a nice underclocked 53°C (not bad!) until I buy a new pump. Which will _n o t_ be running submerged :rolleyes:

Just noticed a large crack in the resin, wonder how that got there :confused:

It didnt jump off a building when you werent looking, did it? That resin is hard stuff, I dont see how you could of cracked it short of a S&W magnum 500 shot at close range...

Pretty crazy stuff man. See if you can drill out the motor :D

bummer :(

Perhaps you can make the best of the situation at hand by further dissecting the pump and seeing what else is in there.

yeah bummer. But I got me new one, and I'm going to be nice to it:) And it better be nice to me, or it *will* get shot :D
I think the crack may have been caused by thermal shock (???) - it did get dunked from warm into cold water a few times. Even so....
dunno.

So yes, dissection with a bench grinder or something.

how cold was the water you dunked it into? and how warm was the pump? if it was only like, a 20 degree diff or anything, thats probally not the case.

Did some chiselling…
In the first pic, you can easily make out a large fault line, the resin is cracked right through. I didn’t do that!
A nice sharp blow with a chisel, and this piece came right off, revealing the motor coil.
After lots more hacking, nibbling and chiselling, I got the motor coil out completely. That resin is damn hard stuff, and it totally encloses the coil.
I can see how it may have cracked though - it seems as though the motor coil "expanded" more than the resin and just pushed it o u t wards. (hot pump dumped into cold water - is 20°C diff enough to do this?)
The last pic shows some rust on the core so water obviously did get in!

After cutting the windings open, I found that the varnish insulation has completely "gone" in places, so this coil must have got very hot - for some undetermined reason… :confused:
Hence decision to measure new pump winding resistance to see how hot that one gets (copper resistance varies by 0.4% per °C)

edit: o u t ward is rude???