Rails dropping with Sparkle 530

[Gautam]

It's been going strong for two months now, and still is, in fact. I don't have any stability problems, but my +12v is a lot lower than it used to be. It used to be at around 12.1v under a full load, tested with a DMM. Now, it's at around 11.86v. I know that's not low enough to panic at all, but what concerns me is the huge drop, which I can't account for in any way. It also very slightly fluctuates; it's at 11.92 when idling. Before, I don't recall it really varying at all between idle and load.

I haven't made any changes since then. The temperature has gotten a little hotter, but this wouldn't explain why the rails are low even the second that the system starts up. Could it simply be dust? I tried shooting an air duster through the fan grills, but would it be worthwhile to open it up and thoroughly dust it? Would there be any reason that my pots may need adjusting, even though I've never touched them before? Should I even be concerned?

 

[Oklahoma Wolf]

Not sure there's much there to worry about, this can happen over time, but you could tweak the pots to bring it back up again. Maybe check the capacitors while you're at it and see if any are bulging at the top. You can also lightly brush off any dust, which will be a little more effective than shooting air through the vents.

I'm starting to have the opposite problem with my CWT - the 5v has gone up over time to 5.15v, and after two years of constant use it does need some capacitors replaced. I'm going to use that as an excuse to grab one of the Fortron 530s for my own and then fix the CWT for emergency use

 

[docinthebox]

11.86V translates to 1.2% load fluctuation, which is just the way it is with a $75 PSU I guess. Better load regulation requires a more complex circuit with more components which in turn is more expensive to design and manufacture, like the pc power and cooling 510 deluxe which guarantees 1% load regulation in its spec but costs more than 3 times the price of the fsp530. The high price is not without reason. To expect Fortron to deliver 1% load regulation with a $75 PSU is simply not very reasonable IMHO.

 

[Gautam]

: It seems I've stripped one of these damn screws too much to even open it, not that I even had the strength to... Half-tempted to just take my dremel and cut it off...

I know that 1.2% regulation is spectacular, it's just that it once used to be better. If it had come out of the box running at 11.86-11.92, I wouldn't have minded, buti t's the fact that it changed that sort of concerns me. Don't think I really have a choice now, though. As long as it remains stable, I'll ignore the rails, to an extent.

 

[Gautam]

Erm, ok, so, 30 minutes of basically doing nothing other than attempting and failing to open my power supply, and I'm at 12.21v idle and 12.15v load...much more like it...but I didn't actually do anything...sometimes it's best not to ask questions...

 

[tfr2087]

Yeah, a psu is like dog, leave it alone and it will do its thing.

 

[Gautam]

Thank you tfr

 

[greenman100]

could be the ATX power supply connector's pins are loose, or looser than they used to be, due to thermal cycling/removing-replacing the connectir

 

[wannaoc]

11.86V translates to 1.2% load fluctuation, which is just the way it is with a $75 PSU I guess. Better load regulation requires a more complex circuit with more components which in turn is more expensive to design and manufacture, like the pc power and cooling 510 deluxe which guarantees 1% load regulation in its spec but costs more than 3 times the price of the fsp530. The high price is not without reason. To expect Fortron to deliver 1% load regulation with a $75 PSU is simply not very reasonable IMHO.

You must not know that much about Fortron/Sparkle power supplies to say that. Actually they are well known as one of the best overclocking power supplies and more than often deliver 1% regulation. The only thing pc p&c power supplies have over them is price.

Chances are it was just the ATX plug getting loose or something simple like that since it "fixed" itself. If all else fails next time just turn the rails up inside the psu. Thats why they put those pots in there. Just make sure to test them while the psu is hooked up so it shows the true voltages.