Xbox 360 power supply faulty

Thread Starter

eyesee

Joined Oct 19, 2013
78
I am trying to repair a 16.5A Xbox 360 power supply.

After managing to get to the PCB, I have found that the input 5A ceramic fuse had blown.

I have checked the switching transistors, rectifiers and transient suppressors and there doesn't appear to be a short across any of them.

What else would cause the input fuse to blow?
 

AlbertHall

Joined Jun 4, 2014
12,345
Capacitor(s)?
Can you actually measure a short?
Have you replaced the fuse and tried the PSU? Very occasionally fuses do blow for no obvious reason.
 

dendad

Joined Feb 20, 2016
4,452
Wire up a 40W mains lamp in series with the power supply. It will help limit smoke when working on them. I have a short extension lead with a box set up that has a lamp socket on it. That way a larger lamp can be put in for working on bigger supplies.
Usually a switch mode power supply will start up, as long as it has no load, with the lamp in series. The lamp will flash as the caps charge and then may glow dim. But if it stays bright you know there is a problem as too much current is being drawn.
Using a lamp as a current limiter will save you blowing fuses and other bits while you track down the culprit.
And the main failure in switch mode supplies are the capacitors. When they dry out and loose capacitance, a lot of transients can cause spike that pop other parts. A capacitance meter and ESR meter is a must. Do an Ebay search for "M328 LCD LCR-T4 Transistor". You can get one for under $10.
 
Last edited:

ian field

Joined Oct 27, 2012
6,536
Wire up a 40W mains lamp in series with the power supply. It will help limit smoke when working on them. I have a short extension lead with a box set up that has a lamp socket on it. That way a larger lamp can be put in for working on bigger supplies.
Usually a switch mode power supply will start up, as long as it has no load, with the lamp in series. The lamp will flash as the caps charge and then may glow dim. But if it stays bright you know there is a problem as too much current is being drawn.
Using a lamp as a current limiter will save you blowing fuses and other bits while you track down the culprit.
And the main failure in switch mode supplies are the capacitors. When they dry out and loose capacitance, a lot of transients can cause spike that pop other parts. A capacitance meter and ESR meter is a must. Do an Ebay search for "M328 LCD LCR-T4 Transistor". You can get one for under $10.
For the first line test; I'd isolate the switching transistor(s) - make sure the input filter/rectifier/reservoir are OK and deal with the more complex stuff later. You can repeat the test with the switching devices back in but B/E or G/S shorted, static voltage won't be as useful test as actual operating but the test has saved me a few £ once or twice.

At the next stage; the bulb in place of the fuse trick is OK as long as there's an NTC thermistor to limit turn on surge (I usually install one if there wasn't). Never use the bulb trick if you've just rebuilt the PSU with shiny new components - you might have missed the part that caused it all to start with.
 

dendad

Joined Feb 20, 2016
4,452
Actually I would say always use the bulb trick! How can that be more dangerous for your new components than just straight on the mains?
It will limit the turn on surge to a safe level. And yes, it is a PTC effect but that is ok. The smaller limited turn on surge is a good test anyway.
There are very few switch mode power supply parts that will be damaged by the limited current allowed through a lamp.
Many years ago a work mate told me this trick. He fixed TVs and found it saved him much smoke.
 

ian field

Joined Oct 27, 2012
6,536
Actually I would say always use the bulb trick! How can that be more dangerous for your new components than just straight on the mains?
It will limit the turn on surge to a safe level. And yes, it is a PTC effect but that is ok. The smaller limited turn on surge is a good test anyway.
There are very few switch mode power supply parts that will be damaged by the limited current allowed through a lamp.
Many years ago a work mate told me this trick. He fixed TVs and found it saved him much smoke.
I use a suitable power resistor - *MUCH* safer than a bulb.

I learned the hard way and I've no time for people who berate me for *LEARNING*.
 

dendad

Joined Feb 20, 2016
4,452
Never use the bulb trick if you've just rebuilt the PSU with shiny new components - you might have missed the part that caused it all to start with.
Well yes, if you have a good power resistor by all means use it, but your answer above indicated to me that you advise using nothing. A lamp is much more readily available than a large power resistor, and very much cheaper.
Here in Oz, the lamps will have an inherently higher cold resistance than in the USA as we use 240V. A typical 60W lamp will limit the cold surge to about 4Amps and having a lamp socket will allow easily exchanging a range of lamps to tailor the current limit to suit the power supply under test. For instance, I would no way use a 100W lamp on a 5W plug pack.
All said, some sort of current limit for power supply testing is a must.
 

Thread Starter

eyesee

Joined Oct 19, 2013
78
Thanks for all the replies so far, there are a few things that I haven't checked yet such as the input capacitors. There is no short directly across the supply post the fuse and the filter components and rectifiers measure OK.

They don't make it easy to service this PSU! The input capacitors, switching transistors and most of the circuitry is under two large metal heatsinks. It seems the easiest way to remove the heatsink is to desolder the components and lift the heatsink off with them. It doesn't seem possible to solder the components back and attach to the heatsink after it is in place as the fastening nuts are difficult to reach.

I'm curious why it is safer to use a power resistor? Is this because the filament may rupture violently due to a large surge current causing the bulb to shatter? It's certainly seems like a good low-cost way to easily identify the fault.

I use a suitable power resistor - *MUCH* safer than a bulb.

I learned the hard way and I've no time for people who berate me for *LEARNING*.
 

dendad

Joined Feb 20, 2016
4,452
A power resistor has a set resistance so you can limit the max current by selecting the value. A lamp in the other hand, has a positive temperature coefficient and has about a 15:1 range of resistance, starting low when cold and increasing as it heats up. That is why I use them. If the power supply does not draw high current, the lamp stays cold and does not drop much of the input voltage, but when the current is high, the lamp heats up and the resistance increases and greatly limits the current to a smokeless less value, mostly :)
 

ian field

Joined Oct 27, 2012
6,536
Well yes, if you have a good power resistor by all means use it, but your answer above indicated to me that you advise using nothing. A lamp is much more readily available than a large power resistor, and very much cheaper.
Here in Oz, the lamps will have an inherently higher cold resistance than in the USA as we use 240V. A typical 60W lamp will limit the cold surge to about 4Amps and having a lamp socket will allow easily exchanging a range of lamps to tailor the current limit to suit the power supply under test. For instance, I would no way use a 100W lamp on a 5W plug pack.
All said, some sort of current limit for power supply testing is a must.
I frequently use a bulb because ICBA finding a suitable resistor - but I frequently pay a price for doing that.

The phrase; "your circus - your monkeys" springs to mind - I don't have to pay for the stuff you blow up.
 

dendad

Joined Feb 20, 2016
4,452
What value resistor and max current do you recommend?
It will have to be able to stand the full mains voltage across it and still allow enough current for the power supply to operate without a load.
I'd be interested to know what you use. Here in Australia for instance, having 240V, the resistor will need to be pretty big power wise.
 

ian field

Joined Oct 27, 2012
6,536
What value resistor and max current do you recommend?
It will have to be able to stand the full mains voltage across it and still allow enough current for the power supply to operate without a load.
I'd be interested to know what you use. Here in Australia for instance, having 240V, the resistor will need to be pretty big power wise.
I don't really understand why you can't figure this stuff out without keep asking me to do it for you.

You have to make a judgement call for each equipment - or find a hobby where you can't hurt yourself.
 

dendad

Joined Feb 20, 2016
4,452
Ok :)
I'll just stick to light globes. I've not had a failure with them at all.
A couple of lamps here for example on 240V...
25W = 182R = 1.3Amps max
60W = 71R = 3.4Amps max
Note that this will be different for 110V. The max current will be higher.
 
Last edited:

ian field

Joined Oct 27, 2012
6,536
Ok :)
I'll just stick to light globes. I've not had a failure with them at all.
A couple of lamps here for example on 240V...
25W = 182R = 1.3Amps max
60W = 71R = 3.4Amps max
Note that this will be different for 110V. The max current will be higher.
Some manufacturers go to the effort and expense if fitting a NTC inrush limiting thermistor - cancel that out by putting a PTC filament in series with it at your own risk.
 

dendad

Joined Feb 20, 2016
4,452
I was not going to reply but could not help but. You have this fixation on light bulbs being evil. I do not understand. Many people have used light bulbs in this way successfully for years.
You are pushing this anti light bulb wagon on other posts too. Then if someone disagrees with you, you get really nasty. That is not what this forum is about. We are all just trying to help each other solve problems, and there are often many ways to do that.

A PTC and NTC will not cancel out.
The NTC, starting with higher resistance, will still limit the inrush current as the PTC is cold and low resistance. The lack of an inrush current will not heat up the PTC, assuming the PTC rating is ok.
The when the NTC has heated and its resistance drops, the PTC will operate if the current goes too high, thereby limiting the max current.
So, having an NTC and a PTC together could actually be the ideal design.
 

ian field

Joined Oct 27, 2012
6,536
I was not going to reply but could not help but. You have this fixation on light bulbs being evil.
.

"fixated"?!!!!! - I pointed out that old trick isn't as wonderful as people think, and you run off on a terrible tantrum.

Is Darwin on vacation ATM?
 

Thread Starter

eyesee

Joined Oct 19, 2013
78
Why don't you just chop the plug off and wire it to an old Atx psu,...



View attachment 137358
This is more of a learning exercise for me to find out what caused the SMPS to stop working. Hopefully if I do get it working, it will be a useful combined high current 12V and 5V power power supply for general use.

I'm still not quite clear why it is not a good idea to use an incandescent light bulb for testing?
 

AlbertHall

Joined Jun 4, 2014
12,345
I'm still not quite clear why it is not a good idea to use an incandescent light bulb for testing?
Lots of people do that - including me when I was mending SMPS professionally. I, and the people I worked with, found it very useful and smoke reducing.
 
Top