A600 - Dead IDE?

[lesta_smsc]

Hello all,

So whilst trying to get WHDLoad to install (it ended up needing a different 'Installer' as the one I had caused some error), I switched off the Amiga as normal but on reboot the IDE is not working. No power or status LEDs show any activity on the converter itself.

I have an SD to IDE converter. It has not caused any issues in the past.

I can still boot from PCMCIA and use that using a boot disk that supports CF via PCMCIA.

I have swapped it out with a CF to IDE converter and after changing the jumpers I can get the red power light on the converter to show up but the CF card is not detected! The red light is solid and does not flicker e.g. to suggest disk activity.

HDD toolbox also shoes no attached devices.

I am a little concerned the IDE controller is dead but how!?

On pressing both mouse buttons during boot there is only DF0 and CC0.

The CF and SD card work fine in a PC in fact I could mount the drive in WinUAE and use it as normal!

Amiga I believe was recapped by original owner some time ago. Amiga has had seldom use and was stored for a while before I took it out a few weeks ago.

Any ideas what may have gone wrong?

Lesta_SMSC

[lesta_smsc]

Just to update. I opened it up and couldn't see any obvious caps leak although a fair amount of dust and possibly hair of the previous owner Lol!...

Anyway pressing firmly on IDE connector allowed the SD to IDE converter to work! Now the issue is, that also pressing on the metal shielding above the IDE port firmly also helps fix the issue. This corresponds to the RF port which actually looks like it's suffered some corrosion.

There must be a dry solder joint that caused the issue?

How best to test the IDE pin connectors without removing the board and/or the connectivity of the RF port pins though surely this would not have any effect on the IDE?

[solarmon]

I suspect pushing down on the RF modulator just has the same effect - as it is still flexing the motherboard around that area.

Sounds like capacitor leak corroding the IDE pins and joints. If it had been already recapped, then it might not have been cleaned up properly. I would recap again to be on the safe side.

Clean up the area with isopropyl alcohol and reflow the IDE pin joints.

[Daedalus]

It's possible that there's been capacitor leakage beside the IDE port that's corroded the pins on the IDE port, and flexing the board allows them to make contact. You really need to clean it up to make sure, and if it hasn't already been recapped, get that sorted.

[lesta_smsc]

This is how it looks. Sorry it seems great quality on phone but upload blurs the pic.

The Amiga now works flawlessly again! Also managed to get PCMCIA Ethernet working so it's now online!

I had some NOS D-Link from back when I was putting LAN on retro laptops! Fortunately it works with cnet... I'll probably sell the other 2. They are still wrapped in cellophane!

[Daedalus]

Glad it's working again! There are lots of signs of capacitor leakage and corrosion in the area though, so I would still suspect you've got a partial failure in the area that means it works sometimes and not others. It's possible for leakage to corrode the IDE port pins below the plastic base, being invisible until some point where a connection fails or pins come away when you try to remove a cable.

[lesta_smsc]

Do the capacitors look original or replaced? It had a new warranty sticker that I removed in order to see what was wrong.

The heat must have caused the issue. I had my PCMCIA card sitting over the airvent above but I hardly had it on long enough for the airflow to be significant in causing a solder melt.

If there was a point to point connectivity test then I could go through each pin in turn and give wiggle to see if there is a faulty there.

[solarmon]

It looks like there may have been a partial recap of some of the SMT caps, but regardless, the area looks to have a lot of cap leakage corrosion and needs to be cleaned up properly.

Use amigapcb.org to check the traces for the IDE pins. Although, this only has a rev 2 board for the A600.

Reflowing the pins on the under side of the board may not help overall, as it is usually the pin holes and traces on the top side of the board that mainly gets affected by the cap leakage corrosion. So it is best to get the IDE connector off for inspection and replacement.

[lesta_smsc]

Originally Posted by solarmon

It looks like there may have been a partial recap of some of the SMT caps, but the area looks to have a lot of cap leakage corrosion and needs to be cleaned up properly.

Use amigapcb.org to check the traces for the pins. Although, this only has a rev 2 board for the A600.

Reflowing the pins on the under side of the board may not help overall, as it is usually the pin holes and traces on the top side of the board that gets affected by the cap leakage corrosion. So it is best to get the IDE connector off for inspection and replacement.

I suspect corrosion is old... but maybe this didn't help with heat induced flex of motherboard.

The unit was unopened and working fine until it just decided to pack up.

[solarmon]

Originally Posted by lesta_smsc

I suspect corrosion is old... but maybe this didn't help with heat induced flex of motherboard.

The unit was unopened and working fine until it just decided to pack up.

All the flexing of the board may have got it to work for now, but is only a matter of time until it fails again if the uderlying issues are not addressed.

[lesta_smsc]

I think I'll avoid messing with the machine too much just yet but if it dies again then I'll review the pins in more detail... not the best solution but I'm so excited with having internet on it that I don't want to risk losing it all just yet.

I have a suspicion it may be the black large capacitor near the RF shield that has suffered damage but pure speculation.

[Daedalus]

That capacitor is just a smoothing capacitor on the 5V rail - you can remove it completely and the Amiga should still work. I don't want to scaremonger, but I would suggest getting it recapped and repaired sooner rather than later. With capacitor leakage, the longer it's left, the more extensive the damage will be and the more expensive complex the repair.

[solarmon]

"A stitch in time saves nine"

[lesta_smsc]

Yes ... you are right. I was so convinced it was recapped as that's what I was told when I bought it. I have a feeling the corrosion is from previous leakage but now I have my doubts too.

I'm keen to attempt this myself. I can see the solder pads for the SMD caps and I suspect the bigger ones are standard caps with the pads being underneath.

I've done some console missing and so familiar with fine soldering... not sure if the skill is transferable to replacing caps though.

What should I do about the IDE port? If due to motherboard flex is there a way to ensure pads are okay e.g. just testing so continuity would prove the pins/pads are in contact?

[Daedalus]

It may have been recapped, but if the original leakage wasn't cleaned up at the time then it will still be there, slowly corroding the board. I have my doubts it was done at all - was it an original warranty sticker you had to remove? I suspect some people will just say it's been recapped in order to get a higher price unfortunately. The larger, through-hole capacitors at least are the very same make as commonly used by Commodore, and the dull solder joints on some of the capacitors would suggest they've been that way for a long time...

[lesta_smsc]

It was a warranty sticker over the original and I was told it was fitted with internal SD to IDE and had recap. The former turned out to be true... seems latter maybe was not.

Great little machine. I love the portability lol.

Slightly odd topic. But I have a small TFT screen that draws under 2A in power. I think it's either 5v or 12v. Can I tap off connection from Amiga? That would make it truly portable as I could mount it to the Amiga lol.

[Daedalus]

Ah, I see. Well, it is possible that it was recapped, but it doesn't look like it from those photos. To answer a previous question, to check the IDE port, have a careful look around the base of the pins for any sign the surface is less than perfect. But really, if the port is suspect, it would be wise to replace it and clean underneath it, as that's where the corrosion usually happens.

Hmmm, 2A is probably pushing it on 5V, unless you uprate the PSU. And 2A at 12V is well beyond what the original PSU can provide. The external floppy port is a useful place to obtain a 5V or 12V power at a heavier current than the other ports, though you may have to modify a 25-pin D-plug to suit.

[lesta_smsc]

Never though of floppy port. Could use a dirty hack and see how well it works. I'll dig out the TFT to see exactly how much power it draws.

I've checked IDE it appears to be 'firmly' adhered although there was a very subtle region where the header was just slightly above motherboard... however pressing the opposite side and the RF shield is what made it work again.

[lesta_smsc]

Originally Posted by Daedalus

Ah, I see. Well, it is possible that it was recapped, but it doesn't look like it from those photos. To answer a previous question, to check the IDE port, have a careful look around the base of the pins for any sign the surface is less than perfect. But really, if the port is suspect, it would be wise to replace it and clean underneath it, as that's where the corrosion usually happens.

Hmmm, 2A is probably pushing it on 5V, unless you uprate the PSU. And 2A at 12V is well beyond what the original PSU can provide. The external floppy port is a useful place to obtain a 5V or 12V power at a heavier current than the other ports, though you may have to modify a 25-pin D-plug to suit.

Sorry to revisit this but I could modify a 25 way connector or source an original 23 way. However, is there any info regarding the current output and what pins I need for the same voltages? I was literally going to wire the power plug into a 23/25 way connector and that should be all thats needed then.