need help my A4000 won't start with new A4000 ATX PSU adapter

[klx300r]

So wanting to replace my old and trusty original A4000 PSU with A4000 ATX PSU power adapter I purchased a micor-ATX psu that fits nicely into the A4000 PSU chasis BUT....

1-tried connecting micro-ATX and nothing..PSU not starting at all nor A4000 so tried another dull size ATX PSU and same thing
2- contacted Amigakit and they told me it was some 'sense' thing with newer PSU's but I had my suspicions
3- tried starting the PSU without connecting to A4000 and voila PSU started up no problem with following voltage readings:

Grey (what you call purple)-+5.21v
Blue--11.32v
Yellow-+ 12.14v
Red-+5.23v
4- let Amigakit know and they said these readings were fine and adapter was working ok but offered no help as to why it wasn't powering on my A4000

So long story over I turn this mystery over to my fellow amigans and hopefully I'm not the only one in the world this has happened too?&^! :-D :-P ..how come this kinda luck never happens when I buy a lottery ticket ;-) :roll:

[AC/DC HACKER!]

Because luck does not exit. Did you fasten the Power firmly to the Amiga? Did it *click* into place? I don't know much about the readings you provided so I can so far only offer generic help.

I have been considering this same thing for my A4000's. Since I have an original PSU that if the wires aren't correct (close enough) causes problems in display because of current disruptions. I'd be glad to buy a newer ATX PSU if this does indeed function well.

Also read this, could be very helpful. Looks helpful. Big HMMMM!!!

http://eab.abime.net/showthread.php?t=25070

[klx300r]

Originally Posted by AC/DC HACKER!

Because luck does not exit. Did you fasten the Power firmly to the Amiga? Did it *click* into place? I don't know much about the readings you provided so I can so far only offer generic help.

I have been considering this same thing for my A4000's. Since I have an original PSU that if the wires aren't correct (close enough) causes problems in display because of current disruptions. I'd be glad to buy a newer ATX PSU if this does indeed function well.

Also read this, could be very helpful. Looks helpful. Big HMMMM!!!

http://eab.abime.net/showthread.php?t=25070

yes all 'clicked in' firmly and yes I'm aware of Steadman's and other options but was can figure out why this adapter doing this with 2 new(er) ATX PSU's ??

[dalek]

You should look up the datasheet for the PSU and check the minimum current loadings for each rail (esp. 12V, 5V, 3.3V). If there is a minimum specified you need to ensure you have a load that will draw sufficient current for the supply, or it may not turn on (they are all different).

Often adding an old mechanical! HDD is enough to load the 12V line allowing the PSU to turn on. The Amiga will usually draw enough on the 5V line. If there is a minimum on the 3.3V rail you will need a dummy load on it. Most ATX PSU's expect a PC to draw a lot on the 3.3V line, while 3.3V is unused on the Amiga.

Other 'proper' solutions for dummy loads are power resistors (calculate values based on minimum load) across each power rail, and maybe some way to dissipate that heat if necessary.

The above doesn't really explain why your PSU starts completely unloaded (though you may find it will shutdown after a little while) but it's an area I would be investigating...

[Daedalus]

Grey and purple are both separate signals on a standard ATX supply, and both will typically be 5V when the PSU is operating normally. Purple will be 5V even when the PSU is off, and grey will be 5V when it is on and stable (or just when it is on on cheap PSUs).

Most PSUs these days expect the main load on the 12V rail, with the 3.3V rail derived from the 5V rail and thus sharing its load requirements. The sense wire thing is still a potential issue on any rail. These outputs in particular may need a dummy load, and may also need to be attached together with other outputs of the same voltage, i.e. all 3.3V outputs connected together, all 5V outputs connected together, and all 12V outputs connected together.

Conversely, if the PSU has split outputs (common enough in modern high-end PSUs), it might not like them being connected together as they're designed to operate separately. The outputs will be marked something like 12V1 and 12V2 in the PSU specs in this case. This might explain why it works when disconnected, but not when connected.

Finally, a good PSU won't turn on if it detects a short on any rail. If the A4000 is working fine, then that's not the case here, but some issue with the connectors perhaps could result in a short on one rail.

[solarmon]

Typically in ATX Power supplies you need to connect the Green wire (PWR_ON) to ground (one of the Black wires) to get the PSU to power on.

https://www.electronics-tutorials.ws...ch-supply.html

EDIT: Sorry, I see now that the A4000 ATX PSU power adapter adapter has a green and black line connected with a spades to connect to the existing power switch. I'm assuming that this has been done, so please ignore my comment!

[klx300r]

Originally Posted by Daedalus

..Finally, a good PSU won't turn on if it detects a short on any rail. If the A4000 is working fine, then that's not the case here, but some issue with the connectors perhaps could result in a short on one rail.

well I'm seriously thinking a detected 'short' is probably the issue but system working great with orignal PSU so how can that be??? also I just tried a working full sized ATX PSU tonight and readings as follows when not connected to anything again:


Grey -+5.11v
Blue--11.52v
Yellow-+ 12.21v
Red-+5.11v


and yet again like the micro-ATX PSU my A4000 won't boot up at all. Tried attaching an old 'real' hard drive to see if some extra load draw would kick it up but same thing.

[klx300r]

so I'm positive that the issue is with the adapter since checked voltages when adapter connected and running in my A4000 and barely getting a reading...tried pushing down on the pins to get better connection but same issue??? maybe have to trim down plastic on adapter??
could it be some type of short on the MB??? if so then why does my original PSU work fine..mind you I was getting a screeching sound out of Paula with original PSU and though that original PSU was acting up causing it but maybe it was something else??


and yes system still boots up fine with original PSU

[Stedy]

Hi,

Does your micro-ATX PSU have smaller orange wires in parallel with the larger ones?
Some ATX supplies use remote sense on the +3.3V motherboard connections, if they get no feedback, the PSU shuts off or won't power up. You could try connecting all the +3.3V wires together.

I have a 600W PSU that powers up fine with no load, though I do connect all +5V, +12V and +3.3V wires together via an adaptor.

[klx300r]

Originally Posted by Stedy

Hi,

Does your micro-ATX PSU have smaller orange wires in parallel with the larger ones?
Some ATX supplies use remote sense on the +3.3V motherboard connections, if they get no feedback, the PSU shuts off or won't power up. You could try connecting all the +3.3V wires together.

I have a 600W PSU that powers up fine with no load, though I do connect all +5V, +12V and +3.3V wires together via an adaptor.

no smaller orange wires on both PSU's I tried and btw that doesn't seem to be the issue since the PSU powers up when not connected to computer/ anything

[Kin Hell]

System Tick Signal.... maybe lack of?

Or is that covered with Voltages as already discussed?

[klx300r]

Originally Posted by Kin Hell

System Tick Signal.... maybe lack of?

Or is that covered with Voltages as already discussed?

I believe that only applies to the A3000 ?

[Jope]

The A1000, A2000 and A3000 take an external tick. The A2000 and A3000 have it switchable between vsync and external.

[solarmon]

Provide some pictures of how you have your ATX power supply connected up to your A4000 - including all the wires and adapter. That might help identify any potential physical issues.

[demolition]

Does anyone know that the duty cycle is of the tick signal in the original PSU?

[solarmon]

Originally Posted by demolition

Does anyone know that the duty cycle is of the tick signal in the original PSU?

http://www.ianstedman.co.uk/Amiga/am..._supplies.html

This suggests that only A1000, A1500, A2000 and A3000 "requires a 'tick' circuit which provides a 50/60Hz +/-10% reference signal"

[demolition]

Yes I have seen that, but it doesn't say anything about the duty cycle. My original A2000 PSU doesn't seem to be outputting any tick signal although the remaining voltages are fine.

[Stedy]

@klx300r

What make and model of ATX power supply are you using?
I can't think of any logical reason for it not to work.

@solarmon

The A1000 PSU spec specified that the input frequency of the Tick signal, derived from the AC mains should be +/-10% and that it only works on the positive half of the cycle. The european and UK mains frequency is 50Hz +/- 0.5Hz.

My tick circuit used in the adaptor provides a 50Hz/60Hz with an accuracy of +/-0.1Hz at 50/50 duty cycle and works without issue.

[klx300r]

well guys the 3rd PSU I tried from a working PC in my storage wasn't lucky A loud quick chirp with the dreaded metallic smell from the PSU only...didn't see anythign differetn or hot on MB and no small...and alas now even my original PSU can't power up my A4000 anymore praying it only damaged the MB and nothing else...now time to send off the MB to Acil in the states to have a look and go from there....unless there's someone in Canada that can have a look and troubleshoot A4000 MB's please let me know

[shaf]

I repair Amigas, you can bring it to me, i'll take a look.

Shaf