Amstrad PC 2286

[RodMerida]

Greetings!
I was having since a couple of weeks your same problem with my old Amstrad PC2086: I turnt it on, and it powered during half a second or even less, then it shut down. I thought like you, that it was a problem from the power supply, until I had the idea to disconnect it from the motherboard, plug it to power separately and turn it on, and it worked. Then I imagine it could be some overheated RAM chip. Do you know what It was?

The hard disk! Either the hard disk circuit board that it has attached under it or the molex cable was producing a shortcircuit. When I unplugged the molex cable everything worked. I got the circuit board from another malfunctioning hard disk (that malfunctions due to a problem in its engine mechanism), put it to my hard disk, replacing the problematic card, and it worked!! Now I have my good old Amstrad PC2086, the first computer in the family, working again. I can't describe my joy.

 

[ajacocks]

Welcome to the forum @RodMerida !

I'm really glad that you were able to save your hard drive. You might even be able to repair the damaged board, as you probably just have a shorted capacitor on it.

- Alex

 

[RodMerida]

That's my only problem. It seems like finding a pin in a straw loft.

Here I send some pictures of the "broken" circuit, and of the hard drive it belongs to. If you, guys, knew to provide me some hint of where to start it would be appreciated.

To the left, you have the non-working one. To the right, you have the good one, that I replace. And in the last two pictures you have the two only differences I found: one missing component in the non-working one (just over the number "C58"), that is present in the good one, and one little green cable in the good one, that doesn't exist in the broken one.

Another problem I have with this hard disk is that, even when I attach It a working circuit, it attempts to start reading but doesn't ever start, because its engine (that little black box in its corner) seems to be stuck. So it just makes "trrrrrrr", "trrrrrr", "trrrrrr", and sometimes it beeps three times.

 

[Caluser2000]

Welcome to the forum @RodMerida !

I'm really glad that you were able to save your hard drive. You might even be able to repair the damaged board, as you probably just have a shorted capacitor on it.

- Alex

Ditto

 

[GiGaBiTe]

That's my only problem. It seems like finding a pin in a straw loft.

Here I send some pictures of the "broken" circuit, and of the hard drive it belongs to. If you, guys, knew to provide me some hint of where to start it would be appreciated.

You're going to need to do some low level, bare metal debugging to figure the fault out, if it's not just a passive component that failed. Think logic analyzers and oscilloscopes.

To the left, you have the non-working one. To the right, you have the good one, that I replace. And in the last two pictures you have the two only differences I found: one missing component in the non-working one (just over the number "C58"), that is present in the good one, and one little green cable in the good one, that doesn't exist in the broken one.

You can't compare the two boards, because they're different revisions with different components. You have one with 1988 date codes and one with 1990 date codes, they were made at least two years apart. The design could have had dozens of changes in that time span. The green bodge wire is a correction in the design, leave it alone.

Another problem I have with this hard disk is that, even when I attach It a working circuit, it attempts to start reading but doesn't ever start, because its engine (that little black box in its corner) seems to be stuck. So it just makes "trrrrrrr", "trrrrrr", "trrrrrr", and sometimes it beeps three times.

That black box is a servo motor. If the motor isn't able to turn, either the servo motor is seized, or the heads that the servo motor is actuating are seized. If the servo motor is seized, you can try putting a few drops of machine oil on the rear exposed shaft to get into the rear bearing, then try to gently turn the shaft. If you can't turn it without excessive force, then stop because the heads are most likely seized to the platter.

If this is the case, you have a very dangerous situation. If the heads are seized, it also means the main platter motor is seized. When these motors are seized, they can pull excessive current and blow up the controller board. You'll need to open the lid of the drive to figure out what's going on inside.

But do not leave the drive plugged in for more than a few seconds with that fault. Also be sure to have your fingers on the various ICs on the controller board to check if they're getting hot. I suspect that the faulty board is faulty because it was damaged by the frozen motors.

 

[RodMerida]

But the card that got damaged was the one that was attached to the hard disk whose motor was working right. And the card that works is the one attached to the hard disk whose motor is seized. I noticed the motor that is seized has some kind of green, sticky slime coming from inside and covering a little of the axis. I suspect that is the real cause of that motor being seized. Isn't it possible to buy another new motor to replace the sticky one?

 

[GiGaBiTe]

Green sticky slime? If it's not grease, then it's a pile of corrosion, that motor got wet from something.

Replacing the motor may not be possible, because if the motor is aligned from the factory, even the shaft slipping slightly would end the drive. Your best course of action would be to try and free the stepper motor. You'll probably want to pull the lid off the drive to see if there's anything inside the drive causing it to get stuck.

 

[RodMerida]

It's like moss or lichen, that green thing.
It's something vegetal, definitely, lol. You could see it through a microscope and see a whole ecosystem there. It's even a little yellowish. It's seen when you remove the servo computer from its place, obviously.

The person who donated me that hard drive told me that happened to him with another servo motor, of another hard disk, that didn't work due to it. It was full of ground, he said. He opened it, cleaned it and the disk came back to life.
Maybe I should open this servo motor and immitate his steps. But still I will have a broken circuit card attached to it. Maybe by measuring continuity I could identify what capacitor is failing, couldn't I?

And after that, still I don't have a controller card for that hard disk, I would have to share the one I have for the other working one. Where could I find a suitable, compatible controller card without having to buy another whole hard disk?

Even if I repair it, I am seriously considering to sell this hard drive. Or else, to get some clonic, homebrew made compatible controller card for this RLL drive, like those white ones that are universal XT-IDE and are sold on eBay, or this black one (would they be valid?): https://www.ebay.es/itm/125643277275 . Before they sold some white ones made in the United Kingdom, so cheaper to send to my address, in Spain, specially before Brexit! Can I ask where you guys are from? I am courious to know if some of you, collectors of these kind of Amstrad PC computers are not either from UK or continental Europe, but from any other English-speaking country, like USA, Canada or Australia, since I don't imagine these Amstrad PC computers being sold overseas. I have always understood Amstrad personal computers like rather an European phenomenon, as well as Sinclair microcomputers. Am I wrong?

Alright, so after this historical divagation about regions regarding Amstrad retro products, here is a key question that I make myself and share with you, that could finally solve all my problems related with possessing just one controlling ISA card for these two hard disks: is there no way to add both hard drives into the same computer by setting one like master and another as salve? Such an advantage that would be!, multiplying the avalaible memory for storage per two, inside the same computer, without having to open It each time and change the hard disk to be able to access to different things.

And by the way, I am not sure whether opening the hard disk by inside will produce problems derives from micro-particles of poder entering inside its isolated, clean environment. Do you have any experience about doing that?
All this said, thanks for your time and advice, and have a good week!

 

[acgs]

I am courious to know if some of you, collectors of these kind of Amstrad PC computers are not either from UK or continental Europe, but from any other English-speaking country, like USA, Canada or Australia, since I don't imagine these Amstrad PC computers being sold overseas. I have always understood Amstrad personal computers like rather an European phenomenon, as well as Sinclair microcomputers. Am I wrong?

You are wrong about this, but not totally. Amstrad computers were sold in Canada, Australia, and the USA. But they were not as popular in these countries as they were in Europe.

I respectfully suggest that any more discussion about your 2086 or Amstrad's regional sales should be continued in a new thread or two. I'm suggesting this because this thread is about @VintageVic's 2286 restoration project, and your 2086 repair problems
1) won't be easy for other/future forum members to find under this thread's title
2) are for a system in the same line, but with many internal differences and therefore might potentially confuse readers.

 

[VintageVic]

I'm glad the Amstrad 2000 series rises discussions.

Not so glad about my 2286. Recently, it developed a 'faulty RAM error' duging POST. Alternatively, it can be
a 'Parity error'.

I have a bunch of old simms, but changes do not seem to help. It's stuck unfortunately.

Can someone check the Link settings - in case you have DIP ram's on the motherboard and not the simms?
I have the original ram ic's, but forgot how the LK settings were with them. Here are the link settings for simms,
but it does not have settings for the orignal ram:

Older AT PCs

 

[VintageVic]

Oh boy, these old amstrads are so much fun.

I was getting mad about what happened to my memory. I nearly went on thinking I have to replace
ths simm sockets. Until finally a solution. Put back jumper to LK 32 ( =enable on board floppy controller).

I was going to check how the system would work with external I/O card, thus I need to disable onboard floppy controller.
However, that is not possible. If I remove that jumper -> faulty ram at boot. When jumper is put back, no longer ram faults!

I double checked this by just removing the jumper while nothing was connected and yes indeed this is what caused my ram errors.
I do not know, if it is only an issue in my particular unit, but if anyone out there have Amstrad 2286 with ram fault, do check you have
jumper at LK32.

Now it's back at working state. Still, I have the issue that floppies work only at DD size, 720k. Now I'm not sure I want to touch it,
at least it is working as it it.

 

[bcg1976]

I'm also starting work on a dead 2286. Can you guys confirm that the power supply will run without anything attached to it? I know with a modern power supply you need to short some pina to get it going. I want to start with the first commandment and check line voltages. The power supply will run the fan for a second and then stop regardless of what is attached to it.

 

[VintageVic]

Welcome to the forum bcg1976. Great year by the way (I was born then)

No, the Amstrad pc2286 psu will not run without any load. I think I even needed
to connect it to the motherboard (and not only to a hard drive for example), but I have to confirm this to be sure.

 

[Scruit]

I'm also starting work on a dead 2286. Can you guys confirm that the power supply will run without anything attached to it? I know with a modern power supply you need to short some pina to get it going. I want to start with the first commandment and check line voltages. The power supply will run the fan for a second and then stop regardless of what is attached to it.

I started typing up a response and I think that myself and VintageVic are answering two slightly different questions.

Those two questions are:

1) Does the power supply have a PS_ON pin that needs to be shorted to tell the power supply to turn on, like a modern atx power power supply?
Answer: On my PC2086 there is no "psu on", "power good", "+5v standby". It has a physical power switch, and it's either on or off.

2) Will the power supply provide stable voltages when unloaded?
Answer: ...

So, here's my observation with the power supply I testing in a PC2086 I just got hold of - but I suspect it is not working correctly, so I must test it more.

The outputs I noted from the PSU with no load on it:

Main connector:
- The two +5v pins read as 5v
- The +12v pin reads as +8.5v

Smaller connector:
- The negative 5v pin reads as 0v.
- The negative 12v pin reads as 0v.


My thoughts:
I know that the unloaded voltage on many power supplies can't always be trusted, so I will have to test +12v with a dummy load connected. I am also wondering if the -5v and -12v are generated downstream of the +12v, so they MIGHT come good when the +12v is under load. I'll have to check the PSU schematic.

 

[ilpan]

Hi to the forum,
I own an Amstrad PC2286 computer, it was working fine last time i powered on but before a week when i tried to power it up the power supply refused to start.
It starts for some seconds and then nothing. I opened it and saw a lot of corrosion caused by leakage ot the 2 big capacitors at the secondary of the psu (6800μf / 16V).
I cleaned it up and replaced all capacitors, not only the 2 faulty, but the psu is not working. The fan is still spinning for some seconds and then stops.
Seems I have the same psu mailfunction as VintageVic has.
Unfortunally I do not have any chematic diagram of the psu, so I have to try repair it without it.

 

[VintageVic]

>1) Does the power supply have a PS_ON pin that needs to be shorted to tell the power supply to turn on, like a modern atx power power supply?
>Answer: On my PC2086 there is no "psu on", "power good", "+5v standby". It has a physical power switch, and it's either on or off.
>
>2) Will the power supply provide stable voltages when unloaded?
>Answer: ...

Thanks for pointing this out.
If there is no load connected, the psu will not provide voltages even if power switch is turned ON.


>So, here's my observation with the power supply I testing in a PC2086 I just got hold of - but I suspect it is not working correctly, so I must test it
>more.


It's been a while since I fiddled with the 2286 psu, but I think I noted that
the 2286 psu and the 2086 psu are not the same - not compatible. There were differences,
when I read the 2086 psu schematics from actual 2286 psu that I have.

Links to monitor, 2086 pcboard and psu are posted early in this thread (with low res though).
The 2286 schematics are much harder to find.