Why won't this thing just stay dead?

Iyonixes are a pain. The last time I put a new PSU in mine it worked perfectly for a whole month and then went south so completely it wouldn’t even show anything on the screen, let alone boot. So I took out the hard disc to use with my Pi and kicked the dead case in a corner of my junk room. That was over a month ago.

Today I cleared out the room and came across the corpse. Just out of curiosity I plugged in a mains cable and the old monitor and mouse next to it and flipped the switch. It beeped. Hadn’t done anything like that last time. Then the screen lit up and before I knew it, the minimal desktop of a discless machine was staring me in the face.

Now I’m not sure what to do with it. Do I just bin the damned thing anyway or do I buy one of the cheap (EUR 10) 80GB IDE discs my local PC shop has on offer and try to resuscitate it? I know it’s going to be a disappointment, probably before the week is over…

I just don’t get it. It’s done something like this before, with the original PSU. Completely choked on it, had a replacement which croaked after five years and then accepted the original like it never had a problem with it in the first place. Only to go bad again eight months (!) later. Can anybody explain that? Is this just some capacitor going ballistic over time or is there another explanation?

More a rant than a problem with a ‘real’ solution I suppose, so Aldershot rather than General or Community Support.

I feel your pain. My Iyonix is still my primary machine, but sometimes it can be a real nightmare to get it to boot. And yet on other occasions it’s straight in as if nothing were ever wrong with it.

The strongest correlation I can see is with temperature; the warmer it is, the more likely it is to boot first time.

Another thing I do is power it up with my finger on the reset button, and keep it in reset for 60 seconds or so before letting go. (A latching toggle switch would make it easier.)

Once it’s up, it’s a pleasure to use, of course!

I too have always found too if the room is cold the Iyonix is less likely to start first time.
If I am using the works laptop nearby (you know, one of those Windows things) that throws out enough heat to assist !

Well, being the stubborn, stupid sod that I am, I of course bought a 80GB IDE HD to give it another try. The machine immediately showed what a mistake that had been by producing a Disc error 23, within seconds after !HForm started its job. That particular error has always been a clear sign of impending PSU related doom, so as far as I’m concerned, that’s it. It may still be kicking at times, but dead it is.

Is there anything particularly special about an Iyonix PSU, or can you just pull one out of the next scrap PC you find?

(the trend towards laptops and tablets means there are fewer secondhand desktops going around, though the supplies of scrap ones should hold up for a few years yet)

Is there anything particularly special about an Iyonix PSU, or can you just pull one out of the next scrap PC you find?

No idea about the ‘official’ view on that, but I wouldn’t count on it. In my experience PSUs that work with other machines fail unexpectedly in an Iyonix. The first replacement I bought years ago was transplanted to another machine when the Iyonix no longer wanted to know about it. Still works like a charm there.

Is there anything particularly special about an Iyonix PSU, or can you just pull one out of the next scrap PC you find?

One of the Intel support chips needs smother? power than most motherboards. Castle, others and our own tests found a significant proportion (about half?) of off the shelf PSUs don’t work reliably with an Iyonix PC.
HEC the maker of the PSU used in the classic case no longer makes that particular PSU and later models have a minimum power drain that causes big problems on an Iyonix.
We only have a very small stock of the suitable HEC PSUs left, but fortunately we have another make/model that Castle and ourselves have tested and have been supplying for some years. I have now stocked up on these so that we can continue to support Iyonix PCs for some years to come.

HEC the maker of the PSU used in the classic case no longer makes that particular PSU

Not that that matters, because that was the first one the Iyonix rejected…

a significant proportion (about half?) of off the shelf PSUs don’t work reliably with an Iyonix PC

Optimistic estimate as far as I’m concerned. My experience so far: 100% failure.

What annoys me is that it can take anywhere between a month and five years before the blasted thing decides it doesn’t like a PSU after all (and then manages to change its mind for eight months before spitting it out again). It’s not as if the PSUs themselves fail (which is what I thought at first, before I successfully used them in other hardware).

We only have a very small stock of the suitable HEC PSUs left, but fortunately we have another make/model that Castle and ourselves have tested and have been supplying for some years. I have now stocked up on these so that we can continue to support Iyonix PCs for some years to come.

I doubt I’d buy one of those, even if it were feasible to do so from where I live. To be honest, I just don’t believe even the ones you have would work in this thing here. It seems to be particularly nasty about its PSU.

If the problem is that the current drawn from thew PSU isn’t enough to keep the PSU happy: Why not simply add a suitable resistor across the PSU output to waste enough current to make it work?

Or isn’t that the problem?

Jim

If the problem is that the current drawn from thew PSU isn’t enough to keep the PSU happy: Why not simply add a suitable resistor across the PSU output to waste enough current to make it work?
Or isn’t that the problem?

The alternative PSU is I think a much better option currently. Though in five to ten years time we may have to do something like that.
n.b. IIRC At least three different voltages have minimum power requirements.

I never have had any PSU problems and every power supply I have tried has worked just fine for me. There has been lots of other strange symptoms but I think these have nothing to do with the PSU. More like the position of the moon and the stars in the sky. Original and the replacement PSUs all work the same as far as I can see. My current PSU was just grabbed from the shelf and works like the original, but is much more quiet in operation and is the only reason I changed it.

For example at my home both ethernet interfaces (mainboard and PCI card) work with DHCP. When at my work the interface on mainboard that works on home does not work with DHCP, only the external interface works there. System boots and works when case is open and sideways, but when you put the case back and lift it up then nothing happens when you push the power button. You can change position of PCI cards and then it will turn on again, or when you disconnect IDE-cable the system will turn on. Then when you put the cable back nothing happens.

If you get it working by swapping everything around and opening/closing case 100 times in row, then better not touch anything inside the case anymore and it will just work until the end of the world.

The alternative PSU is I think a much better option currently. Though in five to ten years time we may have to do something like that.
n.b. IIRC At least three different voltages have minimum power requirements.

That’s fine so long as they are cheap and avalable. But even three few-watt resistors don’t cost a lot. So the real cost would be fitting them I assume. However I’m not clear if this is the entire problem. Maybe something else like the voltages being a tad off the exact range a fussy Iyonix wants is part of this?

Jim

Maybe something else like the voltages being a tad off the exact range a fussy Iyonix wants is part of this?

I don’t think range is the problem, Castle said "One of the Intel support chips needs smother? power than some PSU’s give. And yes as I said “Though in five to ten years time we may have to do something like that.”

More like the position of the moon and the stars in the sky.

I did some computer work many years ago at a place near Farnham. The computer would frequently crash with a General Protection Failure (and the geeks here will know the vintage I am talking of). In order to work on the machine, I moved it to a desk with more space, poked around, it worked fine. Put it back together, it worked fine. Put it back on its desk, it worked for about five minutes before crashing. To cut a long and boring story short, after many many tests and swapping out most of the internal components except the motherboard itself, it worked everywhere in the office except on this table.
The person I was working with knew a guy that did EMC testing, so one evening he came over with a pile of kit and this freaky looking (and large) thing that looked like a TV aerial. Lots of poking and prodding and measuring the effluent spewed out by the fluorescent lights, but nothing untoward was detected. The electrics were on a ring main arrangement, but just in case this was checked out. Nothing. Measures with the computer in use in both locations. Nothing unexpected.

The old guy in the back room who had kept quiet up until now said “I heard somewhere that ley lines pass through here”. The EMC bloke said “we can’t measure those reliably yet”.

While I’m not one for pseudomystical babble, it was (sadly) about as good a theory as anything else.

In the end a bookshelf was repositioned and the desk moved about a metre along the wall. And the computer worked faultlessly until the harddisc died of old age, many years (!) later.