PDA

View Full Version : Replace RLL Hard Disk flex?



gleegum
June 8th, 2016, 03:16 PM
I have this Seagate ST-238R Hard Drive with a broken flex cable.
I'm assuming this is whiy it isn't working on my IBM XT.
Is it possible to replace it? Not sure if the drive could be opened or if I do this the HD gets destroyed.

Some pics:
31577 31578

Chuck(G)
June 8th, 2016, 04:00 PM
Have you examined the cable under magnification to see if there are any broken conductors? You may be able to patch with some of the silver-based ink that's used to repair PCB traces and rear-window defrost conductors on cars.

I'd recommend that before opening the HDA.

gleegum
June 9th, 2016, 12:18 PM
Have you examined the cable under magnification to see if there are any broken conductors? You may be able to patch with some of the silver-based ink that's used to repair PCB traces and rear-window defrost conductors on cars.

I'd recommend that before opening the HDA.
I'll check it, not sure if the silver glue works, I tried to repair an Amiga keyboard membrane eith that thing and it didn't work.
Thank you.

abandonware
July 7th, 2016, 07:02 AM
Clearly the flex is cut and can not be repaired with conductive ink. In my view, you must open the hard disk and replace the flex. The problem is getting a new flex or one from a donor drive.

When you open the disc be very careful with parts and try to have the disc open the shortest possible time and do it in a clean environment and windless. Try first with the donor disk to learn how to open it and see how it is inside before opening the disk to repair.

Stone
July 7th, 2016, 07:37 AM
I have a Seagate ST-225 that's kinda flakey that you could use for parts. FWIW, the hardware is the same as the ST-238R.

Chuck(G)
July 7th, 2016, 08:43 AM
Clearly the flex is cut and can not be repaired with conductive ink. In my view, you must open the hard disk and replace the flex. The problem is getting a new flex or one from a donor drive.

The flex looks cracked to me. But I'm not sanguine about one's ability to replace the whole thing--and then, it's a craps shoot if the thing can be made to work. There are several good videos on repairing/splicing flex cables Here's one (https://www.youtube.com/watch?v=TwvelNOHioY).

If there's nothing important on the drive, I'd probably just bin it. The failure rate of old MFM/RLL drives is pretty high, particularly consumer-grade ones like ST238.

diskettenfett
July 11th, 2016, 12:39 PM
Don't open the drive! Scrape the cracked part down to bare copper and solder short thin magnet wire pieces onto it to repair it. Then apply multiple polyurethane coats or even a thin layer of hot glue to stabilize it and there you go. Been there, done that, worked fine. Takes an hour and might save a good drive with good bearings.

NeXT
July 11th, 2016, 01:28 PM
I wouldn't even bother with the magnet wire. I'd just butt the break as close as it can be to the other half and flow the solder over both.

lyonadmiral
July 12th, 2016, 04:40 AM
If there's nothing important on the drive, I'd probably just bin it. The failure rate of old MFM/RLL drives is pretty high, particularly consumer-grade ones like ST238.

Chuck,
Just because I'm curious, in your experience, is it a particular component that is failing across these drives or is each drive failure different?
-D

Stone
July 12th, 2016, 04:54 AM
I don't have any hard evidence but I suspect that many of them are dying because of track 0 issues.

compu_85
July 12th, 2016, 05:10 AM
In the case of stepper motor drives, I'd wonder if you could "bring them back" by changing the head / stepper position just a teeny bit.

-J

Stone
July 12th, 2016, 05:32 AM
I've heard that can work but I've never had any success with that method.

Chuck(G)
July 12th, 2016, 08:05 AM
Chuck,
Just because I'm curious, in your experience, is it a particular component that is failing across these drives or is each drive failure different?

You could well ask the same thing about people. But mostly, I think the heads start to fail, then perhaps the coating--servo tracks go bad, etc.

NeXT
July 12th, 2016, 10:19 AM
Chuck,
Just because I'm curious, in your experience, is it a particular component that is failing across these drives or is each drive failure different?
-D

RLL drives were always rather unreliable. They were often drives picked off the production line that had scored especially good on testing so they were more likely to handle the RLL encoding method more reliably. Of course after 20+ years that kind of reliability is not expected compared to the traditional MFM encoding.

I also suspect a lot of the older MFM drives are being disposed of because nobody wants to actually research why they are failing. Delamination? Open head coil? noisy data lines? Someone doesn't know how to wire the drive up? Nodoby knows. "Because it's old" sometimes isn't a valid answer. Hold onto your damn drives. Ffs.

MCbx
July 12th, 2016, 03:31 PM
Soldering to this ribbon?
I tried with many scrap ribbons. Always the same thing, substrate foil melts before soldering melts. What are they using for soldering them, Wood's alloy?

I had a CMI Octagon (?Kalok with re-branding stickers?) with dying ribbon (similar thing as in this Seagate, pushed by front panel bracket), it was possible to fix it a different way. I located two broken tracks, exposed them and used a wire glued to parts above/below the exposed area. Before, I inserted a piece of paper folded few times under ribbon. Next, I applied a thin piece of plastic and another layer of glue to keep exposed part from oxidation as copper passivates. After securing with sticky tape, I pushed the cable again using bracket to maintain a nice pressure and keep contact. Risky method as there is no second chance after applying glue (and no way to check that it works), but this Octagon is still working nicely (except few bad sectors at the end trimmed with partition) in one of my XTs.