XBMC4Xbox

I just replaced the PSU on my 2004 v1.6 because it was randomly turning off after anything from 1s to 90mins - progressively deteriorating as the days went by until I got fed up with it last week. That worked.

So now I have a Tuscany PSU which I would really like to fault-find - for fun/kicks/knowledge building. Visually it looks undamaged, no cracked or loose solder joints, burnt components etc...
Anyone got any experience they can share, pointers to resources, or tips...?

do you get switching of the hi voltage fet or voltage on the secondary of the isolating transformer when it plays up ? at least you can dioginose if it is hv or lv related.

Now we're getting somewhere, errr, can you point...
http://i.imgur.com/DJK2s.jpg

Are you in Europe? Cause I have an extra PSU for a 1.6v Xbox

if you dont know what the fet is or the mains isolation trassformer i would leave it alone and buy a new powersupply.

but the smps fet looks like it is the one next to the transformer between the two capicators it has a heat sink on it.

you need to check the gate signal frequency with a multi meter or scope it shuld be somewhere around 2khz with a positive bias from the dc output of the diode bridge.

on the low voltage side of the transformer you should see the same frequency in an ac wave form you need a true rms multi meter to measure it or a scope. voltage on these components are 230v and it sounds like you should not mess with them.

professor_jonny wrote:if you dont know what the fet is or the mains isolation trassformer i would leave it alone and buy a new powersupply.

I know what they are, a schematic is what I'm missing.

professor_jonny wrote:but the smps fet looks like it is the one next to the transformer between the two capicators it has a heat sink on it.

Excellent, ta.

professor_jonny wrote:you need to check the gate signal frequency with a multi meter or scope it shuld be somewhere around 2khz with a positive bias from the dc output of the diode bridge.

on the low voltage side of the transformer you should see the same frequency in an ac wave form you need a true rms multi meter to measure it or a scope. voltage on these components are 230v and it sounds like you should not mess with them.

I haven't got a scope, but I have a nice Fluke multimeter. Ta for the tips.

**EDIT**
Looking much more closely, it looks like there is some scorching around IC1 - 5 pin device connected to a small heatsink - which I'm assuming is the switched mode FET with something else built into the package.
Whether that scorching is indicative of a failure I'm less clear, it may just be normal after years of use.
I'm a bit reluctant to plug it into mains and probe it live. And I can't see the part number printed on it - the capacitors obscure the view .

**EDIT 2**
So I took one of the big caps off to get a view - it's a Fairchild KA1M0680-E45. Shall I take a punt and see if I can get a replacement? Any tips?

xboxprince wrote:Are you in Europe? Cause I have an extra PSU for a 1.6v Xbox

Southern England.

I just needed to know because of the whole voltage difference with the US an over seas I'll sell you mine.
I'll pm you

xboxprince wrote:I just needed to know because of the whole voltage difference with the US an over seas I'll sell you mine.

Thanks Prince, but this thread is all about how one might start fixing a broken Tuscany rather than buying a new one - but thanks for the thought.

...I hope PJ or someone else is able to help with the next step before I start free-styling and it becomes irredemable.

What you really need to do is download the SMPS control IC documentation and datasheet in there it will have sample schematics of how to implement the chip it a circuit.

From that you could draw up your own schematic and compair things and from there figure out what is wrong.

That is the best method of finding out what is wrong as it will give you information on the feedback signal and scaling frequency and modulation how it works and you will know the signals and references on the pins of the IC and then you can test the sorrounding components for faults.

Heating around a fet is normal that will discolour the copper and silkscreen layer and soldermask.

Just looking how many pins it has on it it is probally a fet controled transistor like a darlington pair arrangment, they do that as it requires little gate current to turn on (advantage of the fet) but can switch a large current and high voltage cheaply (advantage of a transistor).

It is unlikely it has failed they usually blow up but they often contain thermal managment and shut down on overload.

I don't know how far you've got but if you still need help and you have a meter set it to DC volts and while it is connected to the mother board check and see if you have 5vDC.
Use this pinout as a guide so you know where to stick your probe.
http://i1038.photobucket.com/albums/a46 ... inouts.jpg
You want to check if you have 5vDC after turning the Xbox on and if not if you have 3.3vDC. I'm suspecting you will have nothing at all but this is needed to be checked. This test will determine whether you have a faulty power supply or if it is actually the on board 5vDC to 3.3vDC regulator. If you have no power at all then I would be checking the power supply's fuse that by the way looks nothing like a fuse but is a fuse no the less. If the fuse checks out fine, (if you respond to this post and still need help I point exactly where the fuse is), then your problem will most likely be the varistor, again see how things check out first. These two parts are the only parts I have ever had problems with on these power supplies and lightning was the reason they failed. The varistor is designed to go high resistance when to much voltage is drawn through them, 275volts if your machine is a PAL 220-240 volt machine and this high resistance is designed to pull more amps and thus blows the fuse, before it blows the Christ outta the machine and at worst case, start a fire. Both of these parts are easy to get hold of and a cheap to buy so don't be discouraged. The fuse can be replaced with a normal fuse, it just won't sit flat because they are bigger and depending on how long the over voltage condition existed, it may not have damaged the $2 varistor.

Xman, thanks. The failure mode is that the box switches off shortly after switch-on, so it's not going to be the fuse. I'm curious, though, what the fuse is - as you say there's nothing on the board which looks like a traditional fuse. Is it the black piece of wire arcing rainbow-like across the width of the board?

Separately:
- which part is the varistor?
- and how do I determine whether it's busticated?

(BTW, I know the main board is fine since it works with a substitute PSU.)

it is unlikeley the fuse or varisistor is gone it will be something in the ps good latch circuit on the power supply or motherboard.

i have a 1.6 psu and it buzzes if you plug in a dvd drive it is because one of the diodes in the smps circuit is blown.

spicemuseum wrote: - which part is the varistor?
- and how do I determine whether it's busticated?

(BTW, I know the main board is fine since it works with a substitute PSU.)

Well that proves it isn't the fuse or the varistor at fault.
Check how good your P.S. OK signal is. It should be a good strong 3.3vDC. If it is down like 3.1vDC, this could be a result of the 5vDC going to the mother board being to low to start with. You would measure this 3.3vDC in the white wire and the other probe to any black wire. If that is going low for any reason, like the HDD coming on line etc, it could be going below the threshold of that power supply and turning it off. It would be real handy to check all the voltages just so they can be taken out of possible causes. By all means do as P.J. suggests because he is looking more at PS problems where as mine are more mother board related. There is absolutely no reason for me to repeat P.J.s excellent advise so I'm not, just looking at other possible causes.

professor_jonny wrote:...it will be something in the ps good latch circuit on the power supply or motherboard.

Sounds good. Can you point me in the right direction...?

I would check all the voltages on the plug as xman suggested first if you had a multi channel logic tracer you could connect it up to the 5v, 12v, ps ok, ps on, signal and see what happens during power up and forced shutdown and see what actually happens and what might be the cause.

a audrino makes a cheap logic tracer or if you have access to one a labjack or labview card in a pc would be ideal.
i have 2 labjack's at work and they are a very cool bit of kit and cheap enough for the home handy man to buy:
http://labjack.com/products?gclid=CNvs- ... vAodwhMAYg

i have the u12 models and i can say they aree great for this sort of thing.