peters Milan 040 issues

[peters]

The RAM sockets on my Milan are a bit dodgy, as I said earlier. Any disturbance inside the case and I can guarantee having memory issues. I've just done it again.

I have 4x 128MB Simms but I've only ever got one working and I'm fairly sure it's the sockets to blame. I would like 512MB.
.

Do you think it would benefit from a set of new sockets ? It's not something I could do.

[dml]

Do you think it would benefit from a set of new sockets ? It's not something I could do.

Before going there I would clean the SIMM pins with a good contact cleaner & firm brush (or paper card - something that will not remove gold plating) and also clean the socket contacts with contact cleaner & brush. Then confirm the solder joints are all good on the back (reflow them with fresh solder/flux).

If this isn't enough, try to figure out which socket(s) are a problem. If it is more than one, probably change them all.

There are a lot of pins on those things so removing them needs a lot of care and a last resort. It is better to save the board and destroy the sockets, if that makes it easier to remove them - which is another reason its better to diagnose whats wrong first and avoid swapping unless really necessary.

[peters]

There are 2 cards in.

If anything touches either of them then they will cause problems. Either not detected or ram slot errors.

Then I just have to keep pressing the cards and rebooting repeatedly until they work again.

These problems aren't new. It's been like this for about 20 years. I must have bought the 128MB Simms in about 2008. I spent weeks trying to get them to work without success.

What implement and chemical should they be cleaned with ?

I'm not sure I should try to resolder them.

[dml]

There are 2 cards in.

If anything touches either of them then they will cause problems. Either not detected or ram slot errors.

So definitely a mechanical problem somewhere.

- poor pressure from the SIMM bracket spring (is it a metal spring-clip or plastic clip holding the SIMM in position?)
- dirty/oxidised contacts on the memory or the socket (are the SIMMs and socket both gold plated, or is one or other tinned?)
- dry joints on the SIMM socket pins.
- cracked trace on the main board under the SIMM (not very likely unless some strong impact)

What implement and chemical should they be cleaned with ?

The most basic tool would be a new, clean toothbrush with firm bristles and some 70% or better IPA/alcohol. (If it's less than 99% make sure the board has been well dried for a good period before powering it back on, to make sure no water remains).

For the SIMM stick contacts you can use paper card soaked in some alcohol to rub the contacts clean. If the pads are gold this should be easy. If they are tinned, they could be dull and you'll see a lot of grey residue come off when cleaning. They should be shiny if possible or at least the residue should reduce the more you clean it.


Alcohol isn't always very effective at cleaning on its own and it doesn't itself remove stuff from the surface, just dilutes it and washes it around, then dries again - so you need to make an effort to wash the contacts off with some alcohol and e.g. absorbent thing like a cotton bud or paper towel.

There are other things you can use but will be more expensive and/or more trouble to clean up and adds some risk for other e.g. plastic components nearby.

Real contact cleaner for electronics is better for this. In some cases, I use WD40 (common gas-oil lubricant, close to petroleum) which can dissolve some things alcohol won't touch (like dried flux). I would only use that here if you have it handy and don't have any IPA/alcohol. (It stinks and leaves residue which isn't nice to handle but will clean very well - it can dissolve some plastics so watch out).

I'm not sure I should try to resolder them.

This step only requires heating the pins with fresh solder or flux. If nothing else works you could find someone near you who can do it.

[peters]

Thanks for the info.

I do have a Servisol Super 10 aerosol but I've only used it electrics and not electronics. It leaves oil behind.

I wasn't sure about cotton buds and pads due to the fibres they have behind.

I will have to strip the Milan to get at them and I've avoided that as much as possible.

[dml]

Thanks for the info.

I do have a Servisol Super 10 aerosol but I've only used it electrics and not electronics. It leaves oil behind.

I wasn't sure about cotton buds and pads due to the fibres they have behind.

I will have to strip the Milan to get at them and I've avoided that as much as possible.

You can use the Super 10 and you can use the cotton buds - and yes both require some cleaning when you're satisfied the problem is fixed (or not).

Really anything you use on the board should be followed by a wash with 99% IPA and a good long dry-out - but it's not really worth doing that until you're done with other experiments.

But do be careful not to power the board if it is not dry - especially anything with windings (inductors, transformers - anything which can trap liquids) - as it can cause worse problems than the original one. So use small amounts of cleaning fluid and give time to dry out if the board gets wet in places.

[peters]

I've ordered some isopropyl.

I'll restrict the cleaning to the memory slots using the Super 10 and cleaning it afterwards with the alcohol.

[dml]

I've ordered some isopropyl.

I'll restrict the cleaning to the memory slots using the Super 10 and cleaning it afterwards with the alcohol.

Ok, good luck with it. If you find the problem gets worse after a toothbrush+contactcleaner, it would point more towards dry joints on the pins themselves - (if there are joints, the more you mess with the sockets, the worse they will get). So you should get a bit more information on whats happening one way or another.

[peters]

I've ordered some isopropyl.

I'll restrict the cleaning to the memory slots using the Super 10 and cleaning it afterwards with the alcohol.

Ok, good luck with it. If you find the problem gets worse after a toothbrush+contactcleaner, it would point more towards dry joints on the pins themselves - (if there are joints, the more you mess with the sockets, the worse they will get). So you should get a bit more information on whats happening one way or another.

I got around to having a go at the RAM slots and the IDE while I'm at it.

I normally boot with 2 cards in the 4 slots (0-3).
128 + 0 + 64 + 0 = 256

After a bit of cleaning I noticed 2 small pieces of metal or solder stuck in slot 1 and 2. I managed to remove them, I think. However any card in slot 1 was ignored.

Then I got it to boot with 208:
128 + 0 + 64 + 16

And then with 256
128 + 0 + 128 + 0

and finally with 320:
128 + 0 + 128 + 64

This is great but slot 1 looks like it needs more work.

[peters]

I just got the Milan to boot with 3x 128 = 384 but Mint would freeze at different points during init the boot process.