• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

sparkle 250w dieing under power...

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

RCtruckguy

Member
Joined
Oct 11, 2002
Location
columbus,ohio
well i got 2 250watt sparkles to convert to always on psu's and both die under a load, and it takes a 1amp load on the 3.3 to start both also. i've cliped the wires so i really can't rma :( thanks for your help~RCTG

edit: these are the ones with the "noise killer" from newegg
 
What load? Light load? Too much load?

I had a PSU quit on me without much load because a transistor or diode wasn't fastened all the way down against its heatsink. I doubt that Fortron/Sparkle would make that mistake, but you may want to check the temperatures. CAUTION: one of the heatsinks will likely be attached to high voltage, 170V DC, so you do NOT want to touch it while the power cord is plugged in!!!

Why did you buy two 250W Sparkles for $60 when Directron had 300W Fortrons for $27 apiece?
 
Last edited:
oh didn't know that, dosen't really matter anyway just needed 13amps on the 12v rail.

it dies under any load at all be it 900ma or 6amps (they are power supplies for r/c chargers) it can run a charger but once you try to charge a battery (6 amps) it cuts out. thanks for your help~RCTG
 
Sparkles have short circuit protection on them, that's why they instantly cut-out, and require a "hard reset" to turn them back on.
Both of mine will run fine with just the jumper hooked to the green ATX power wire and one of the grounds...is that how yours are wired? I did it all the time to run my pump relay for cleanouts, without the CLR being used as coolant :p

Are any powered devices wired right? Does it work if they are removed?

Are they wired together to power the same device?, or are they seperate?
 
well they won't start up unless there is a load on the 3.3v line and once any load is placed on the other lines it dies. they are wired right.

i have the green and a black shorted to turn it on thanks for your help~RCTG
 
You may want to ask the experts in sci.electronics.repair. I had a non-switching PSU do that because the output transistor was bad and the small transistor connected to it had to do all the work, but that wouldn't apply to a switching PSU. You have to find out whether it's dying because it can't put out enough power or because something is shutting it off, like the protection circuit. If there's an optical coupler (tiny chip with just four pins, soemtimes three of these chips), it may have gone bad. I would try resoldering everything first.
 
" CAUTION: one of the heatsinks will likely be attached to high voltage, 170V DC, so you do NOT want to touch it while the power cord is plugged in!!! "
Not DC, high-frequency AC.
Not to mention that the flyback effect can cause the peak voltage to go well over 500v(almost 700v in a dell supply)!!
And since it is high frequency(91,621Hz in that dell supply), you can get a RF burn(happened to me before).
 
RCtruckguy said:
well they won't start up unless there is a load on the 3.3v line and once any load is placed on the other lines it dies. they are wired right.

i have the green and a black shorted to turn it on thanks for your help~RCTG

Huh. When I place the jumper in my ATX plug, it's unplugged from the motherboard, so there's nothing loading the 3.3 volt lines. There are loads only on the 5's and the 12's from the harddrives and CD's being plugged in at the molex connectors.
 
" CAUTION: one of the heatsinks will likely be attached to high voltage, 170V DC, so you do NOT want to touch it while the power cord is plugged in!!! "
Not DC, high-frequency AC.
Not to mention that the flyback effect can cause the peak voltage to go well over 500v(almost 700v in a dell supply)!!
And since it is high frequency(91,621Hz in that dell supply), you can get a RF burn(happened to me before).

I definitely measured 170V DC because when I accidentally set the multimeter to AC volts, it displayed about 0V. I'm sure that this was 120 Hz pulsating DC because it was from one of the two large 200V filter capacitors on the primary side.
 
There may be one or two low value (< 10 ohms) resistors on the high voltage side to measure the current, and it's possible that one overheated and cracked internally. That would cause the protection circuit to think that too much current was being drawn.

Sometimes an optical coupler burns out a diode or stays fully on, causing the regulation to go funny. They're usually those small chips with four pins sticking out and maybe straddling a slot cut into the circuit board. I believe that most four-pin couplers are about the same, so maybe you can take one from another PSU, or swap parts between your two Sparkles.

Are you loading the +5V lines? Maybe they need a load because I find it hard to believe that both PSUs would fail in the same way, especially something as good as Sparkle.
 
no, i have to load the 3.3 to start them, not the 5v, i'll find more bulbs (my load) and try the other lines at the same time~RCTG
 
I"VE FOUND IT!!!!!! IT WAS A FrRIGGEN BROWN WIRE!

my dad ordered anothe one and we saw the brown sensing wire in the atx connecter. on the other 2 we have we just connected a 3.3 and the brown and it works fine!, if u need to use ps's alone,as little wire as possaable and it dosen;t work look in your atx connecter and see if it had that sensing wire!

well thanks for trying to help me guys, i appreaciate it. good luck to all you PSU modders!~RCTG
 
Aren't they supposed to design the PSU so that if that brown voltage sense wire isn't connected the PSU will instead read the voltage from the PC board, through a < 100 ohm resistor? I read that the resistor is needed in case the voltage sense wire breaks and causes the voltage reading to drop to 0V, which could make the PSU overcompensate and make the voltage go too high.

I have an Antec SmartPower that was designed for remote voltage sensing on the +5V, but the sense wire was left out. I think the voltage sense resistor is 20 ohms. After I installed a remote sense wire, the +5V at the mobo rose almost 0.2V (digital meter reading).
 
i dont know if they are supposed to be designed that way or what but i'm glad i got it worked out! Well then again it wouldn't have worked b/c all three of them never and will never touch a mobo. these are for r/c battery chargers and are gonna me make me some money!~RCTG
 
Back