What could cause this PSU to output 12.5V on the 12V?

[Pentium4]

Re: What could cause this PSU to output 12.5V on the 12V?

Yeah the heatsinks aren't the greatest. But the fan is designed to get 12V all the time so it moves lots of air

[momaka]

Re: What could cause this PSU to output 12.5V on the 12V?

Edit: Added updated pictures

^ Those dummy load resistors on the secondary are giving me the creeps. Check their value. Knowing Deer/L&C, they probably put 100 Ohms on the 12V rail, and 10 Ohms or less on the 3.3V rail - a good way to keep those caps nice and toasty.

But the fan is designed to get 12V all the time so it moves lots of air

If the fan is sleeve bearing, then this isn't very good for it. Either switch to dual-ball bearing if you want to run at full 12V or add some type of controller or reduce the speed on the existing fan to prevent it from wearing too much.

[Pentium4]

Re: What could cause this PSU to output 12.5V on the 12V?

I will check tomorrow! It's an NMB ball bearing fan so it's a nice one

[Pentium4]

Re: What could cause this PSU to output 12.5V on the 12V?

Measured those resistors

[mariushm]

Re: What could cause this PSU to output 12.5V on the 12V?

3.3v - 15 ohm - 0.756 watts - 0.229 A
5.0v - 26 ohm - 0.961 watts - 0.192 A
12v -115 ohm - 1.252 watts - 0.104 A

Yeah, 1 watt is a bit of dissipated power... they will be hot..

[Pentium4]

Re: What could cause this PSU to output 12.5V on the 12V?

Thank you for the info, should I replace them with ones of lower value?

[mariushm]

Re: What could cause this PSU to output 12.5V on the 12V?

I don't think you have much choice.

You have to have some load on the rails to prevent them from oscillating and going all over the place. Even the 5v stand by output may go ugly if you remove them.

They are sensible values, you can see they're less than 1% of what the rails are designed to output. But it's still around 1 watt of generated heat for each resistor.

You can use some wires to move the resistors away from the pcb and the capacitors, for example get a small thin piece of wood (it's insulator and you don't have to worry about shorting something out and it handles heat) and strap the resistors to it and then glue it to the metal case.

Unlike the resistors on the primary side, there's no issue with the length of the leads of the resistors here, so you can just move them further from the capacitors.

[Pentium4]

Re: What could cause this PSU to output 12.5V on the 12V?

Oh okay, cool. Good idea! I thought having long leads on resistors on the secondary caused issues? At least they're high quality caps

[momaka]

Re: What could cause this PSU to output 12.5V on the 12V?

Thank you for the info, should I replace them with ones of lower value?

Lower value - no. You want higher value (less heat dissipated). Around 1/4 to 1/2W per rail should be enough to keep the PSU happy.
With L&C power supplies, though, it's easy because the resistors they put in there work perfectly when you shuffle them around a bit . I've done this to a number of L&C PSUs now, and none of them oscillated afterwards - with and without a load.
The way I do it is I move the 100 Ohm resistor (the one you noted 110 Ohms) from the 12V rail to the 5V rail. The 25 Ohm resistor you can move to the 3.3V rail and the 15 Ohm from the 3.3V rail can be removed altogether (or you can put it in series with the fan so it runs at 10-11V rather than 12V).
As for the 12V rail, you can put pretty much whatever resistor you want. 270 to 300 Ohms (1W rated minimum) would be nice, but you can go as high as 1000 Ohms (1/2W rated) if you don't have any other resistors on hand. And if you intend this power supply to power a computer that draws CPU power from the 12V P4 connector, then you can even leave the 12V rail without any resistors and it will be stable (though I recommend you put something -anything- on there to discharge the caps when the PSU turns off).

Even the 5v stand by output may go ugly if you remove them.

The 5VSB is regulated by its own 431 shunt so the output rails of the main supply cannot affect it.

[Pentium4]

Re: What could cause this PSU to output 12.5V on the 12V?

Wow, very interesting! Thanks for the reply. I think I will use this guy with a 12V based CPU. I measured the efficiency of this thing earlier today and it was only around 70%

[Pentium4]

Re: What could cause this PSU to output 12.5V on the 12V?

Started working on this thing again the input caps were indeed actual 470uF but I have some 820uF Elite caps I will put in there, it's overkill but I have a lot of them...I also added some MOV's to it.

Also going to put a real 35 transformer in there from an Allied, the pins line up perfectly. Upgrading the transformer will increase the efficiency at higher loads? And...would it be worth replacing the D304X's with 13009's? Sorry for all the questions

[ben7]

Re: What could cause this PSU to output 12.5V on the 12V?

Yes there could be an increase in efficiency, but don't be fooled! The circuit was designed for the original transformer, so a bigger one (with different specs) might wreak havoc.

[Pentium4]

Re: What could cause this PSU to output 12.5V on the 12V?

Even though it's from another Deer from the same topology?

[Th3_uN1Qu3]

Re: What could cause this PSU to output 12.5V on the 12V?

Drop it in and it oughta work. You might have to adjust the primary snubber, but you could just transplant it from the PSU it was originally in. The snubber is a power resistor in series with a cap, across the transformer primary.

[Pentium4]

Re: What could cause this PSU to output 12.5V on the 12V?

Thanks for the reply, is it that red cap next to the switchers? There's one in the same spot in the Allied the bigger transformer came out of

[Th3_uN1Qu3]

Re: What could cause this PSU to output 12.5V on the 12V?

No, the red cap is the coupling cap for the transformer. If it's of a bigger value in the donor PSU swap that too, if it's the same leave it alone.

[Pentium4]

Re: What could cause this PSU to output 12.5V on the 12V?

No, the red cap is the coupling cap for the transformer. If it's of a bigger value in the donor PSU swap that too, if it's the same leave it alone.

Cool the donor one is actually a little smaller value (i think) the Allied says 1.0K and the L&C one says 105K. sorry for my lack of knowledge but what would the snubber look like or where would it be located?

[momaka]

Re: What could cause this PSU to output 12.5V on the 12V?

Cool the donor one is actually a little smaller value (i think) the Allied says 1.0K and the L&C one says 105K.

105 = 1 uF. The 1.0 signifies it's a 1uF as well. So they are the same. K means 10% tolerance.

sorry for my lack of knowledge but what would the snubber look like or where would it be located?

As Th3_uN1Qu3 mentioned, the snubber is a large resistor and a ceramic (or sometimes polypropylene) cap in series with that resistor. From the picture you provided above, that would be large 47 Ohm resistor and that ceramic cap behind it (C4). Check me on that, though. They should be in series and together connected across the primary side of the main transformer.

May also want to transplant the snubber diodes that are across the BJTs from the donor PSU if they are of a bigger current rating and have at least the same or higher voltage rating.

[c_hegge]

Re: What could cause this PSU to output 12.5V on the 12V?

And...would it be worth replacing the D304X's with 13009's? Sorry for all the questions

No, unless the 13009s are TO-3P or TO-247 while the D304Xs are TO-220. Both of them are rated for 12A, so assuming they are the same size, neither will handle more than the other.

[Wester547]

Re: What could cause this PSU to output 12.5V on the 12V?

^ D304Xs are rated at 10A (@25*C), at least that's what their datasheet says.... 13009s are rated at 12A in terms of collector current, though at 25*C. So a pair of 13009s (especially in TO-3PN or TO-247 form) may not be an ill-fitting replacement (especially considering BJTs are current driven).