From: pinnerite@gmail.com
On Fri, 18 Jul 2025 12:47:35 -0400
Paul wrote:
> On Fri, 7/18/2025 9:16 AM, Felix wrote:
> > Edmund wrote:
> >> On 7/16/25 16:50, pinnerite wrote:
> >>> I connected a brand new Asus Tuf 850W PSU to an ASUS motherboard with
> >>> just the processor and heatsink installed, in a MIDI Tower case.
> >>>
> >>> The PSU was not in the case.
> >>>
> >>> Switched on and...
> >>>
> >>> Nothing happened.
> >>>
> >>> I detached the PSU and attached a small tester to it.
> >>> Powered on ...
> >>>
> >>> No sign of life.
> >>>
> >>> What can cause a PSU to die under those conditions?
> >>> Surely it has inbuilt protection against overload.
> >>>
> >>>
> >>>
> >>>
> >> What comes to mind is, it didn't die but doesn't switch on.
> >> There might be a electrical check that isn't happy.
> >> Are you sure you connected all plugs and cables to the mainboard?
> >>
> >> As usual, if everything else fails, read the manual.
> >>
> >>
> >
> > also did he connect the CPU 12v power, and the case switch cable to
motherboard power switch pins?
>
> He has built systems before.
>
> He is looking for some encouragement.
>
> 1) He had a working system.
> 2) The power supply in the working system, stopped working.
> 3) A replacement supply is also not working and is not
> responding to PS_ON# testing. He tests the supplies before
> inserting them into a computer case. The tester is likely
> a window comparator type, which tells you whether all rails
> are within acceptable voltage.
>
> These are autoswitching supplies, the OP is in the UK and
> the supply would be assuming a 240V input (50Hz).
>
> As is standard for UK people, he would have checked the fuse on the cord
> fitted to the supply. The fuse has to be able to withstand
> the inrush of the supply, when the supply is switched on at
> the back of the unit. The internal safety Slo Blow fuse,
> easily withstands the inrush current.
>
> A Kill-A-Watt meter, can be used to detect power draw, as
> a "sign of life" from the unit.
>
> Overloading the +5VSB, could shut off that output, but it
> might be self-restoring rather than being a latch-off circuit.
> Some of the +5VSB supplies use switchers now, and the
> little switcher is not power factor corrected, so the
> reactive component seems rather high when you measure it.
>
> There is no particular advantage to connecting the PSU to
> the motherboard, until the question of whether the supply
> actually works is resolved. If the supply is good, it should
> be able to pass a test, outside of the computer case. All
> it takes, for example, is a PSU tester that operated
> PS_ON# properly and has the 24 pin pinout of an ATX when
> it measures the voltages.
>
> Paul
It seems to me after all this that there must be a short.
ChatGPT told me that wihin the box I should find a glass fuse.
I do have a stock so I went open the case of one not in warranty, only
to find I needed two torx screwdrivers. The bit sets that I have
include sizes that are two big.
Down to B&Q and bought the two smallest sizes. 19 & 12.
Got home to find the 10 was OK but I also needed an 8.
It is coming on a slow boat from China.
Th engineer that I hired to take the strain couldn't do it.
He recommended that I buy another motherboard, with LED diagnostic
lamps. I bought a Gigabyte B650 gaming dubree.
Anyway I have tested a Corsair 850Watt and it came up OK.
I plan to glue fibre washers onto the underside of the motherboard
before I fit it in place.
I hope to assemble the whole thing tomorrow afternoon.
Thanks for trying to help everyone.
Alan
--- SoupGate-Win32 v1.05
* Origin: you cannot sedate... all the things you hate (1:229/2)
|
