I have a IIsi that doesn't work - mostly. After it's been turned off for several days, it'll fire up with its little chime, but only work for that first boot. Any more attempts to start it from then on will result in just a *click* and a small spin of the fan, then everything goes completely dead (no ctrl-command-reset works, no buttons on the back work etc) until power is removed and plugged back in, whereby trying to start it gives the same *click* and small spin of the fan.
What I'm curious about is whether the PSU from a IIci/IIcx/Q700 etc will work as a temporary plugin to check if the IIsi starts up OK. I realise the power supplies are very different shapes, but the connector onto the motherboard is very similar (perhaps even identical) at first glance. I'm hoping someone might know one way or another whether these are electrically compatible before I go hunting down the info on the connectors & power on each pin. Im heading off to sleep now, and plan to look up the info in the AM - but sending out feelers in a forum post overnight never hurts, and I might wake up to find someone's done just that before... 
Ta for any info,
dana
Could it rather be a logic board issue? I had the same problems with ram, but this was in
a PowerMac G3 Beige.
Don't think it'll help, but it's a possibility. If not ram, definitely Power Supply.
yeah the powersupply from the iicx/quadra 700 etc.. all will work in the iici, i haven tested this fact before
It could be a logic board issue - but I don't have a spare logic board to test that out - only possible spare power supplies :).
macg4: It's not a IIci, but a IIsi - I know the Q700/Centris power supplies will fit in a IIci - but the IIsi is another form entirely.
dana
sorry my mistake, i retract my statement
I checked out this issue, and the exact same thing happens in my 5300, chime and all.
Turns out that it has a weak power supply.
They are electrically compatible. I've used a IIsi PS as a test bench supply when working on IIci motherboards and vice versa. Obviously, they won't fit in each other's cases, but with everything out on the bench and some way to balance the PS upright, it works.
It is possible you have leaky capacitors on your motherboard. There are a few caps required to keep the PS turned on on the motherboard and with age they leak corrosive, conductive goo onto the motherboard. Solution is to clean, inspect and repair the motherboard and replace the caps. Use tantalum SM caps instead of electrolytic SM caps to avoid a repeat of the problem.
The leaked goo looks like a slight cola stain (sort of brown with hint of yellow) on the board. You may have to hold the board at an oblique angle to a light source to see it. Alcohol works well to clean it off. The stuff can eat through solder and any exposed traces on the board.
Thanks for the answer, I used a Q700 power supply on it, which did the same as the original IIsi power supply - the click then stop - so I just popped the IIsi together again ready to place it into storage, planning to come across a 2nd motherboard for it later sometime - and before packing it away tried it again. Booted almost regularly after that
It's still not perfect though. There's something funny going on with the 50MHz oscillator (it was overclocked to 25MHz before I got hold of it) which seems to be really sensitive to movement.
If I prod the oscillator at the front pushing towards the back, the machine will lock up if it's running, or if it's in a non-starting mood it'll suddenly chime and start. I resoldered the socket for the oscillator, and it's still doing the same. I'd have put it down to a small electronic anomaly if it weren't for it needing to be pushed in just the right direction, so I suspect the oscillator may be suffering from a dry joint or equivalent on the inside, and moving the correct leg fixes/breaks it.
dana
(and yes, every cap on the board was leaking. http://www.danamania.com/temp/leaky.jpg )
If you're going to store it for a while, I recommend that you clean the board first. The corrosive will continue to act while the board is stored.
On my first IIci, back around '96, capacitor leakage ate through a via (hole that connects traces from the back and front of the circuit board) or rather, it ate all the metal in the via so it didn't conduct any more. I had to follow the relevant traces back to exposed metal (component pins) and run a wire to bypass the damage. Not to difficult but rather tedious and that's what happens when you give the corrosive time to act.
Slightly OT, but I have seen loads of Lisa IO boards that have been corroded by leaky batteries. The fumes from the battery damage traces several inches away from the battery itself. Traces close to the battery may be completely destroyed (eg 5mm length of copper corroded and Lisa boards have big, thick traces) and corrosion is visible by eye elsewhere on the surface of traces and the case of components.
Damage caused by leaky caps should be more localized -- the fumes will be far less corrosive than those from a Lisa battery -- but make the effort to clean up all the same before storing.
Phil
It's cleaned - One of the treatments I give all the macs that come into my collection is a thorough disassembly, dusting, polishing and cleaning - the whole treatment.
Comparing http://www.danamania.com/temp/leakyclean.jpg it certainly looks a sight better than http://www.danamania.com/temp/leaky.jpg - and hopefully staves off damage from the leaked electrolyte.
dana
Hi I found a great article on how to replace blown or leaky caps on the motherboard
http://www.rjl.com.au/marketplace/Articles_show.asp?a=26
Out of curiosity, who made the power supply in this IIsi in question?
I've found that on an SE, a Sony power supply is much better than an Astec. On a IIcx or IIci (and probably also a Q700), the Astec again loses, this time to a GE.
Not sure who made IIsi power supplies or if there are multiple vendors of them.