Switch mode PSUs question

Status
Not open for further replies.

cornishlad

New Member
I'm reasonable ofay with analogue/traditional PSUs but I'm afraid my knowledge of modern SMPUs is zilch - its an age thing..I'm putting together an amp for our band on the cheap and have bought a 100watt/2n3055/MJ2955 amp module off eBay.. That was cheap but the power supply bits are not.
But I already have four power bricks from Xbox360's - 12v @16amp...from charity shops...they are sooo cheap
the question then is , is it possible to connect the four of them up appropriately, in series, to get +24v - 0 -24v or would smoke be the result ?
 
If you got them for cheap from a charity shop then you've got little to lose by trying them all in series. Have a fire extinguisher to hand, though .
If they survive repeated switch ons and switch offs when series-connected to a dummy load then you should be good to go.
 
I'm assuming that it all comes down to whether there is complete (or sufficient) isolation between the output and the mains input. I can not measure any continuity between same on the Xbox bricks.
Nigel..thanks for the link...I found similar on eBay and emailed the Chinese supplier to ask if two of them could be connected to give +24v-o- -24 and they replied rather briefly " they can not be cascaded".. I would like to have been told why...I can hardly ask Microsoft the same question hence the query here..But it is certainy a cheap option which is in budget !!
The Xbox bricks are of course completely sealed so can't look inside. I'm quite prepared to connect two in series and test on dummy load, but don't want to risk all four at the same time. Questiohn is, if two were successful are there oher issues which might arise if two pairs were connected in series to provide the +&- supplies..
BTW one attraction of the Xbox psu is that there is enough ooomph to supply two amp modules....which would be good..
 
Last edited:
Like most PC PSU's the ground connection is common AND attached to chassis ground.

Back in the 80's, I had a custom torroidal transformer made 4x35Vx3A which cost $130 USD. I added 4 surplus 9600 uF capacitors at $3.50 ea and a bridge rectifier.
Earlier i had a 20 A constant voltage transformer, but the hum was unbearable. The amp sounded great, though.
Later, I added a 500 W sine wave voltage regulator. $1000 USD value for $125.00 and it helped the torroidal supply immensely. 3A is too small for 100 W. I did get measure $125 W with one channel driven.

For a capacitor filer IDC = 0.62 * I[ac]

(3^2) * 8 ohms = 72 W; the 3 is definately too small.

If I did it again, I might use 6 ohms for design and make I[dc] equal to the peak AC current into 6 ohms.

With one channel driven, 2 windings are effectively not used. There's still some losses in the transformer wiring, but the extra capacity is still there.
 
Last edited:
I know that st least some of the Xbox bricks have a 2 wire mains power connection - a standard IEC connector, but without a ground pin. All of the different versions may not have this weird 2-pin IEC connector, but it suggests that the output is not connected to mains ground.

Since you measure no continuity between any combination of mains and output wires, connecting the supplies is series will be fine.

There is a trick to turning on the output. I'll refer you to a thread I posted on Digital-DIY for details:

Xbox Power Supply - 12v at 14amps
 
Last edited:
Thank you all....I am gaining confidence that this scheme will work and be safe...So far I have only ever powered up one of them up after cutting off the output plug...but I have just dug out all the others - I found another 5 ! and they are all the same model rated at 203 watts each...And, more importnatly, they all DO NOT have an earth pin in the IEC mains connector. JonSea, the link above does not work for me...but, if I remember correctly, I did find by experiment that a small switch between ground neg and a wire carrying 5v sparks it up ok..perhaps that's what the link would have said....Roger
 
Hmmm. Thw link works ok for me. But at any rate, here's the important info:

 
Hi,

Here are some thoughts.

First, when you connect two power supplies in series that's not the whole circuit in most cases. There is also usually a low impedance (low resistance) load. So you have two power supplies in series also in series with a low value resistance.

Now turn one of those supplies on, only one, and you see that the second one ends up with about -12v across its input where it should be +12v. That's probably no good unless it has a protection diode in anti parallel to the output, and then there will be maybe -1v or a little less. That's not good either, but it's a lot better than -12v.
Why consider this case? Because it is almost impossible to get both supplies up at the same exact time.
So adding protection diodes if they are not there already might be a good idea, even if it appears to work without them at first.

Next, we can change the circuit at our own option to one with JUST two power supplies in series and NO LOAD (yet). That makes this a little better because now there is no path for the second power supply above to get -12v. If it has 0v just before it turns on, it still has 0v after the first one turns on all the way.
The solution then is of course to use a relay or other switching device on the output of the power tupply and turn on the two supplies while the relay is off, then after a second or two turn the relay on. Also have the relay turn off automatically if the power line goes away for any reason (lightning storm, etc.).

These ideas might help to get two supplies that are the same but dont work well in series working in series. It's still a guess though what will happen if one goes into current limit before the other, but it may survive by way of the protection diodes.
These diodes should be able ot handle more than the rating of the power supply, 2x the current rating might be good.
 
thanks for the link...I found similar on eBay and emailed the Chinese supplier to ask if two of them could be connected to give +24v-o- -24 and they replied rather briefly " they can not be cascaded".. I would like to have been told why.

Because it's not true, there's no reason you can't use two of them as a spilt supply - I suspect they didn't understand what you were asking.
 
Now turn one of those supplies on, only one, and you see that the second one ends up with about -12v across its input
How (assuming you meant output rather than input)? Can you sketch the circuit arrangement you have in mind that would do this, or say what assumptions you are making about the supply internals?
 
If you power up two supplies and check for a voltage between the positive output of one and the negative output of the other. Any voltage suggests the isolation isn't good enough.

Mike.
 
Because it's not true, there's no reason you can't use two of them as a spilt supply - I suspect they didn't understand what you were asking.
That's from post #10...I wondered the same...And that link worked next time I tried..A very useful site that I'd never heard of - I've ordered some well priced lathe tools off them already...
MrAl..After some thought I get that - similar issue as having a 2 cell torch with one flat battery I think...I will act on that when I experiment.
Pommie (Mike) "If you power up two supplies and check for a voltage between the positive output of one and the negative output of the other. Any voltage suggests the isolation isn't good enough." I don't follow that though....with two connected in series for 24v the + of one will be connected to the - of the other with a piece of wire...how could there be a voltage difference?..I don't doubt you are making a valid point but it will need a better explanation..
 
What Mike is saying.....

Plug two supplies into mains power with no connections between their outputs. Then measure between one wire one the first supply to either wire of the second. Ideally you won't measure any voltage between the two supplies. However, I don't think this is going to be a valid test. With a high impedance digital meter, I think you'll be fooled by insignificant leakage voltage.
 

Yes, it's a pointless test, it proves nothing - you'll get a voltage reading if the supplies are isolated or not (the ones I linked to are isolated, as are pretty well any proper supply you're going to come across).
 
Ok, when checking the voltage connect a lowish value resistor (100Ω) in parallel with the meter. No significant current should flow (so no significant voltage across resistor) if they are isolated.

Mike.
 
Last edited:
How (assuming you meant output rather than input)? Can you sketch the circuit arrangement you have in mind that would do this, or say what assumptions you are making about the supply internals?

Hi,

The circuit is simple, just two power supplies in series and those two also in series with a small "load" resistance, that is meant to draw current at 24v which here is the total series aiding voltage. In other words, connect two 12v supplies in series to make a single 24v supply, then connect a low value resistor across that 24v supply. Normally the resistor gets 24v.

However, turn only ONE of those supplies on first and leave the second one off, and we could see -12v across the second supply when it should be +12v. That is because there is now a low impedance path between the positive terminal of the upper supply and the negative terminal of the lower supply.

The circuit:
A o---12v---+---12v---+---RL---o B

and then connect A and B together to form a complete series circuit.
Next, turn only one on (lets say the top one) then we have:
A o---12v---+---open---+---RL---o B

and now assuming RL is small we see A connected to RL and RL connected to the (-) output of the second supply, meaning we end up with:
A o---(12v)---+---(-12v)---+---RL---o B

(again with A and B connected together).

The situation with a SPLIT supply (ground between the two to form plus and minus supplies) is not that much different unless BOTH loads are connected to ground and there is NO load that connects from the positive of one supply to the negative of the other supply (24v situation again).

For power supplies with an anti parallel diode, the reverse voltage may only be -0.7 to maybe -1.2v which probably isnt too bad although with some supplies it will blow out the diode and this defeat the purpose. A high current diode would keep it safe i think and should allow normal operation.
 

Very scholarly I'm sure, but of little relevance to the thread.

We're not discussing two separate power supplies, but rather two power supplies combined to make one split power supply, so they shouldn't be an option to switch only one of them on.

Obviously if you used two entirely separate supplies with separate mains leads then you're asking for problems, but that would be a very foolish thing to do.
 
Re.#9..I assumed MrAl was talking about a momentary situation that might arise if the PSUs did not spark up at the same instant...Of course they would all be powered by a common mains switch in any scheme I would try...
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…