Corsair CV650 output coil

[andyc]

Hello!
Does anyone know where I could find this output coil for a Corsair CV650 PSU? I can't find a schematic and there is nothing written on it.
Thanks!

[прямо]

Why? Is the coil burnt to death?

[rogfanther]

You will probably find it only in another dead psu of that model.

One possible way is to unwind the wires, taking not of the number of turns, and recreate this coil with another core. This one seems well cooked, and will have lost its magnetic properties, or at least changed them.

[andyc]

Why? Is the coil burnt to death?

Yes, it's supposed to have a nice copper color.

rogfanther
Thanks! I didn't know that the core also sustains damage due to heat, I was planning on reusing it. Are these coils also dipped in laquer after winding? This one is pretty stiff but I am not sure if it's laquered or the melted enamel on the wires is keeping it together.

[rogfanther]

Some are covered in lacquer, others are not. In the situation it is now, hard to say if that is the melted enamel or burned lacquer.
Did the psu fan failed ?

[andyc]

rogfanther No, it was mounted upside down in one of those cases that pull air from the bottom of the PC and was sat on a very shaggy carpet. The rest of the PSU survived.

[momaka]

I can see the ruler next to the coil, but can't easily tell its diameter. Looks like possibly it's about 33 mm (without factoring in the copper coils.) If so, that's a T130 -size core. As for the core type, I can't tell with 100% confidence, but I think I can see that it's yellow on the bottom. Could you unwind a turn or two and post pictures of both sides of the core?
If it's a yellow core with white paint on one side, then it should be a Micrometals T130-26 type core. If it's a yellow core with red paint on one side, then it's more likely to be a T130-8 type core... though I double this would be the case here, as Micrometals type -8 cores are not common in PSUs. Another possibility is yellow with gray paint on one side, making it a T130-35 type core... but again, these are not really seen in PSUs. Meanwhile, Micrometals -26 type core is quite popular in PSUs, so I think that's most likely your part here.

That being said, an alternative drop-in replacement for -26 type material is often -52 material. These cores are light green with one side painted blue. So if you can find a T130-52 core, that should work too.

Now, looking at the number of wire strands on the core, I don't think this one is for the 3.3V rail. Looks more like it's the main output toroid, as there appears to be two sets of different windings from the thick wire strands (probably for the 5V and 12V rails) and a thin winding for the -12V rail.
Before going out to buy a new core and rewinding it, it's often possible to steal one from a cheapo PSU, just so you can verify that everything else works. Generally speaking when replacing -26 type cores, you can often drop down the size to a 28 mm diameter core... i.e. T106-26... at least for testing purposes.

[stj]

it's dipped in glue to stop it vibrating,
if you dont see any shorts between the windings it's probably good.
a ring-tester would help but most people these days dont even know what that is!

[momaka]

it's dipped in glue to stop it vibrating,
if you dont see any shorts between the windings it's probably good.
a ring-tester would help but most people these days dont even know what that is!

Nah, I've seen a few PSUs with baked output toroids now, and this is definitely *not* the color from any glue/varnish job. Dead giveaway - notice how the wires on the bottom of the toroid are a nice red or copper color. That's because they didn't get baked (as much) by the toroid core as the wires on the actual core. Once they start winding on the toroid core, they turn black/brown - a sign of complete thermal runaway of the core. And if you look carefully, you can even see in the above picture where some of the varnish has started to peel. That's exactly how all three of the failed ones I got turned out. Barely touching the wire made the varnish peel away and reveal the bare copper.

That said, rewinding the core is pointless, as it has degraded from the high temperature and lost its magnetic properties... so the new windings will overheat again, or the PSU just won't work right (lower-than-normal output inductance, resulting in possible resonance on the output and/or over-stressing the O/P filter caps.)