• Please review our updated Terms and Rules here

Amstrad PC 2286

VintageVic

Experienced Member
Joined
Dec 9, 2019
Messages
340
Location
Finland
Hi,

I scored this Amstrad PC 2286 at local auction site for eur 200 + 8 eur shipping.
The unit is defective (blown psu fuse when power is attempted), but I took it anyway,
because it happens to be the exact same unit that was my first ms-dos based pc back in the -91.
So I have very warm memories of this unit and they are really difficult to get these days.

It's going to take at least for a week for me to get this as the seller is going to ship it only next monday.

In the mean time, I want to start planning where to start with fault repairs.
I guess I need to strip everything from the main board (except ram) and try it again with the power supply and new fuse.
But even before that, need to check for possible defective filter caps in the psu itself. And obviously visual inspection
on all electrolytic caps on both psu and the motherboard.

Any other suggestions or experiences with these units? I think these are quite rare and thus expensive. Luckily
the unit is whole, including mouse, keyboard and the monitor as well. Otherwise I would not have paid asking price.

Unfortunately, the unit has been saved for who knows how long outside, in a "dry" storage. However, since temperatures
range from -30...+30 celsius here, cold outside storage is not really ever dry. The seller also admitted there is corrosion
visible somewhere.
 

Attachments

  • amstrad.jpg
    amstrad.jpg
    54.7 KB · Views: 41
I always test all boards for shorts before I try to power anything up.

If it's a similar design to the PC2086 then you'll have 3 isa slots under a plastic cover at the back, just under the monitor. I'd remove any cards from there, test them for short between the power and ground pins on the isa connector.

Unplug the power supply, figure out the wires (gnd, +5v, +12v, maybe -5 and -12 also) and then test for shorts between each of the power pins and ground. Test the HD too - middle pins on the power connector are ground, outer are 5v and 12v. Make sure no shorts there.

Once you have the PSU ready to test then test all the outputs. 5v, 12v etc. It may need a load on it to start up (some do, some don't) so an old HD may do it (sometimes does, sometimes doesn't) but don't hook up to mobo until you're sure it's giving good power rails.
 
hi all !
On the PC2086 there are four ISA slots. The forth is not accessible through the trap. Maybe it's the same on the PC2286.
The PSU should have +12, +5, -12 and -5V. BTW on my 2086 without a load the PSU shows less than 12 and 5V but the PC runs quite happily.
It reminds me I still have to write an update to my PC2086 PC adventures.
Have fun with your Amstrad. They are good machines.
 
They are good machines indeed, much better than the crappy plastic case would suggest. And the 4 AA batteries on the top under the monitor means there should be no battery damage within at least.
 
hi all !
On the PC2086 there are four ISA slots. The forth is not accessible through the trap. Maybe it's the same on the PC2286.
The PSU should have +12, +5, -12 and -5V. BTW on my 2086 without a load the PSU shows less than 12 and 5V but the PC runs quite happily.
It reminds me I still have to write an update to my PC2086 PC adventures.
Have fun with your Amstrad. They are good machines.
Huh. I had one in the very late 80's and I don't recall the 4th slot - but I also didn't take the case off so I must have just not seen it.
 
Thanks for all the comments!

The colour monitor arrived today, the computer itself is still on the way.
Tested the monitor with other pc. There is some life in it and something has also failed.
At least, it might be repairable, because dark red image shows when brightness is set to maximum.

Firstly, when turning power on to the monitor, there is green bar of sort on the left. It vanishes while
the power has been on for a minute or so. I'd guess green and blue colors are not working.

I wonder if there are schematics available for this monitor on the internet?
Exact model is PC14HRCD Colour monitor. Well - I have not done my due diligence about this yet either,
I will start with google.
 

Attachments

  • monitor.jpg
    monitor.jpg
    75.6 KB · Views: 31
  • green.jpg
    green.jpg
    119.3 KB · Views: 28
  • pins.jpg
    pins.jpg
    89.5 KB · Views: 27
I now have the full unit.

Computer looks quite corroded inside. I pulled out the psu, opened it and cleaned most of the
dust away. Also washed and dried psu metal cover box, it had lots of oxidy in it.

There is an issue with the psu I believe. Contrary to what the seller told me, it has not blown its
fuse. But turning power on, the fan briefly turns and then stops. This is without any load. Tried to
give it load firstly the floppy drive only and it flashes floppy led also once and psu then stops turning
the fan the same way as without load. Adding HD to the load changes nothing.

I could not find schematics for the monitor. There has been repair manual that has the schematics, but it
does not look like easily accessible on line. I fear it is the same case with psu schematics.

Psu pcb has fair amount of corrosion. I will try to look for bulged caps, but briefly looking at it revealed
nothing suspicious yet.
 

Attachments

  • interior.jpg
    interior.jpg
    124 KB · Views: 36
  • psu.jpg
    psu.jpg
    120.6 KB · Views: 35
  • psu2.jpg
    psu2.jpg
    55.6 KB · Views: 37
Well well. I tried to measure resistance of R658 and after applyin light pressure to
it with multimeter probes it just snapped right off. Perhaps this was one issue (open resistor)
if it was open already before my measuring attempt.

I have to fix this before continuing. I probably won't get good read with multimeter on the value of
this resistor. Colours appear to be brown - black - red. Not 100% sure, but perhaps there is beige or gold at the end after red.

On line resistor value calculator suggests this is 1kohm resistor. I think I should have some at stock...

edit:
ahha, it also appears that the trace on the other side has also been severed by corrosion. So even if
the resistor had been fine it would not have had effect.

Schematics would really help, but since there aren't any available, I just have to try to look carefully
where it should go. Traces on this pcb is a bit hard to see and corrosion makes it worse. What is
your favourite poison to clean this green corrosion away?

I happen to have some old white wine from -93 that I was going to throw away... perhaps it would
be suitable for the job? I mean, applying it to the pcb, not absorbing it orally, ha!
 

Attachments

  • Resistor.jpg
    Resistor.jpg
    193.1 KB · Views: 24
  • trace.jpg
    trace.jpg
    227.5 KB · Views: 25
Last edited:
I happen to have some old white wine from -93 that I was going to throw away... perhaps it would
be suitable for the job? I mean, applying it to the pcb, not absorbing it orally, ha!
You're going to waste 30 year old wine on a PCB?
 
I have no other use for it. But to those who are terrified... that was a low end bottle and it has
been stored in warm instead of more suitable cool place. I believe it is closer to vinegar than wine today
and as such I thought maybe good for the purpose.

I have not dared to use anything just yet on the pcb. Some suggest baking soda. I worry that I make
it worse. If I cannot see even remnants of the traces it is going to be impossible to make repair to it.

This unit has really suffered strong dampness or maybe these were just low in quality. Worst part
seems to be in middle edge. I might pull one cap close to there to see better if it has leaked on the pcb.
 
I ended up washing some of the pcb surface just with water/dish was liquid and toothbrush.
it helped a bit and I was now pretty confident about one broken trace which I then corrected with
a botch wire. Also installed workin 1k resistor. That big cap right next to resistor is a bit of a suspect,
but I'm going to let the pcb dry and then see if this had any change in its behaviour.
 

Attachments

  • pcb.jpg
    pcb.jpg
    164.6 KB · Views: 16
  • pcb2.jpg
    pcb2.jpg
    76.4 KB · Views: 16
Examining a working 2286, I have confirmed that resistor 658 was originally a carbon film 1K ohm. (1/4 watt, of course)

I think I might be able to help you further.
 
Wow, I'd appreciate help much with this!

I have two suggestions. As you can see from this picture, the lower and lower/left part of the pcb is fairly
corroded and there are color changes which make it difficult to follow the traces. I could really use high (enough) resolution
image of this side of the pcb. That would help me check each trace of the corroded area. Image sizes are really limited
here so I give my email if you're willing to email good quality image of the pcb to me: mikamaaranen at hotmail dot com.

I have also considered a backup plan if this fix fails. Perhaps I could replace original psu with a small AT or atx power supply.
To do that, I would need voltage pinouts for the motherboard and maybe for the floppy if it is Amstrad specific floppy drive. I'm
guessing HD power pinout is standard. I tried to search these pinouts from internet, but they too are hard to find.

Any help will be much appreciated - thanks in advance!

edit:
Oh yes, progress check up. There was no change in the psu after reinstalling the resistor and trace fix.
I pulled the 330uF cap next to it. Legs look like corroded, but not really sure has this leaked or is it
just because the whole motherboard has been in such a damp storage. ESR value is about 0,35 which
I believe is ok for this cap.
 

Attachments

  • pcb3.jpg
    pcb3.jpg
    178.7 KB · Views: 15
  • cap.jpg
    cap.jpg
    186.9 KB · Views: 14
Last edited:
I wonder if there are schematics available for this monitor on the internet?
Exact model is PC14HRCD Colour monitor. Well - I have not done my due diligence about this yet either,
I could not find schematics for the monitor. There has been repair manual that has the schematics, but it
does not look like easily accessible on line. I fear it is the same case with psu schematics.

The monitor's service manual is accessible, but hidden, at archive.org:
https://archive.org/details/manual_PC2086_SM_AMSTRAD_EN
OR​
https://archive.org/details/manual_PC2086S_SM_AMSTRAD_EN

Amstrad used to put all the service information for a single computer system in one manual. The PC14HRCD must have first been bundled as a monitor for the 2086.

The 2 copies of the manual are scans from photocopies, so you might need to get both, one being clearer in some places to the other and vice versa.

Also, assuming you get the machine working, do you have any of the software?

Have a look at this page for some help with that:
https://web.archive.org/web/20090221220118/http://web.ukonline.co.uk/cliff.lawson/files.htm

And here's some info about the hardware, for after you've got the machine working:
https://web.archive.org/web/20101216023809/http://web.ukonline.co.uk/cliff.lawson/atpcs.htm

This info is really old (11 years), but it might still be valid for finding the other service manual:
https://web.archive.org/web/20110214135050/http://web.ukonline.co.uk/cliff.lawson/manualss.htm
 
Wow, I'd appreciate help much with this!

I have two suggestions. As you can see from this picture, the lower and lower/left part of the pcb is fairly
corroded and there are color changes which make it difficult to follow the traces. I could really use high (enough) resolution
image of this side of the pcb. That would help me check each trace of the corroded area. Image sizes are really limited
here so I give my email if you're willing to email good quality image of the pcb

I'll see what I can do. My above post was delayed since I hadn't passed my "probation" yet. But, this is message 10(!), so that shouldn't be a problem any more. So, I'm assuming it posted too late for you to see it before you made your last post.

The 2086 schematics might be helpful since the machines look similar. But, of course, an 8086 based machine and a 286 based one will be vastly different. I haven't been able to check this yet, but if we're lucky, maybe they used the same PSU.

I have also considered a backup plan if this fix fails. Perhaps I could replace original psu with a small AT or atx power supply.
To do that, I would need voltage pinouts for the motherboard and maybe for the floppy if it is Amstrad specific floppy drive. I'm
guessing HD power pinout is standard. I tried to search these pinouts from internet, but they too are hard to find.

This might be a great idea anyway, since these amstrad bar-shaped, wind tunnel style PSU can really be noisy. Even if you get the original power supply working, a standard supply adapter would make a great backup.

As far as the floppy and hard drives go, they should both be standard. The only non-standard drive interface should be the one on the right-hand side of the system for the external floppy. It looks like a centronics printer connector (data) and a 4-pin din connector (power). Again, it's possible that the 2086 schematics are the same for this board and you might be able to use it for any necessary repairs.

I'll also see what I can find out about the capacitor. I'm still looking for the 2286's service manual. Which reminds me: when searching for Amstrad info, don't forget that Schneider was a reseller of these machines.
 
Thank you very much Acgs !

I have downloaded the 2086 service manual, very grateful for this!
I also got needed (bios) setup.exe and windows driver for paradise vga. A bit early for this, but
now they are saved for later use if I'm successful getting this thing back to life.

I have not yet studied the 2086 service manual, going to check it out next.

Admin may need to turn a switch to end your 'probation', but that'll probably happen soon
now that you have posted 10 times.
 
I started by enhancing the 2086 psu schematics. Some areas are really white on the scan
and it's a bit of guess work. Next I need to identify does the 2286 have exact same psu.

There seems to be several overload protection circuitry here. Now I'm guessing, but perhaps
one (or more) shorted components are causing overload and thus shutting down the psu
almost immediately after the fan spins a few rounds. And guessing even further, maybe the possible
short is not in +5V rail since the fan does spin briefly.

I'll post the slightly enhanced scan here, if someone is later struggling with similar psu issue.
 

Attachments

  • 2086_psu_schematics.gif
    2086_psu_schematics.gif
    102 KB · Views: 20
Okay,

I have examined the psu schematics, but especially the parts list of the manual.
It appears that the psu has been redesigned for the 2286 and thus is actually not the same as amstrad 2086 psu.
For example, the R658 that we know is 1kohm on this unit is listed 560ohm at the 2086 parts list. Secondly, this
psu had IC's 605 and 606 for example that the 2086 part list has not got at all. Other parts do not seem to match
exactly either... so I need a new battle plan with this.

And here it is: I'm going to list all the IC's with this psu unit. There are voltage regulators and diodes (bridges) that
have heat sinks attached to them. IC601 (STK7458) is very big IC with largest heat sink attached to it. There is nothing wrong
externarly on them but somehow I have gut feeling that some of these IC's have not survived extreme temperature changes
of Finnish weather. After making parts list of this unit I will try to find on-line source for replacements and order them.
And finally replace components that I have got.

Orders can take week(s) to arrive here, during which time I could continue to check the pcb visually for broken traces or
other hints of failed components. Perhaps not the most cost effective approach, but it will take really long if I order one component
at a time that I consider worth changing.
 
So, here is my parts list so far for the 2286 psu:

location | part
D616 | C25P046 (tai C25P048) Shottky barrier diod (TO-3P)
D614 | F10P20F Fast recovery diod (TO220)
IC606 | AN7905F 3-pin fixed negative output type monolithic regulator (TO220)
IC607 | AN7912T regulator (TO220)
IC605 | BA6993 monolithic dual comparator, 8pin MF package
IC604 | BA707 3 terminal regulator (TO92)
IC603 | LA6324 High performance quad operational AMP (DIP 14)
IC602 | MB3761 Voltage detector (DIP 08)
IC601 | STK7458 (hybrid amplifier?)
Q616,Q606 | C1815 NPN transistor (TO92)
Q601 | 1346 transistor
(?) | B1135 Transistor (TO220)

This is of course far from complete list, but a list of possible replacement candidates that I'm now looking for.
Cleaning the pcb did help a bit and I was able to check many traces last night on the most corroded areas. I could not find
more broken traces. I also checked most of the single diods on board and could not find problems with them. The
shottky barrier diod (D616) seemed most suspicious though as the resistance between its legs were under 100 ohm.
I will probably need to pull that out of the pcb and measure its resistance again after that.
 
Back
Top