Megastor 8" drive unit

Talk about non-Acorn classic computers/hardware/software here (including retro consoles)
User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Megastor 8" drive unit

Post by jonb » Sun Mar 03, 2019 4:58 pm

Following on from the "Magic Smoke" thread, I spent this afternoon restoring the dual 8" drive unit. It's in here, somewhere...


IMG_5250.JPG
Bit of a bootload of old cr@p...


I'd spent the day after collecting that lot in triage and basic cleaning, as every item was covered in dust and dead spiders. Today, it's the turn of the "Megastor", a chunky (and staggeringly heavy) dual 8" drive unit. Last time we saw it, the magic smoke had been released and was curling lazily up towards my kitchen ceiling, stinking the place out.


IMG_5269.JPG
Megastore

IMG_5268.JPG
A peek inside. Spot the grime / magic smoke.


So, a closer look at the neat ASTEK PSU reveals the dreaded RIFAs, one of which is popped, big time.


IMG_5265.JPG
RIFAs

IMG_5266.JPG
More RIFAs, with "Old Smokey" bottom left.


Some soldering is in order..
Last edited by jonb on Tue Mar 05, 2019 3:07 pm, edited 6 times in total.

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastore 8" drive unit

Post by jonb » Sun Mar 03, 2019 5:09 pm

By now, I'm sure you know the drill. Out with the old, in with the new.

IMG_5337.JPG
Astek PSU is a joy to work on. All cables are screw fit.

IMG_5354.JPG
Back in with new X2 caps


Next, I decided to give it all a jolly good clean and service the drives (clean as best I can, and lubricate them).


IMG_5342.JPG
Mains powered fan, caked in dust.

IMG_5341.JPG
Drive wiring. 2x240v + 2x24v looms.


These drives sounded very rumbly when first powered up, so I took the spindle bearings out and gave them a good oiling.


IMG_5346.JPG
FAG 609ZR spindle bearing


This is a fairly common bearing, still available today. The ones in the drives are tatered of course. They don't turn smoothly; there's a sort of grittiness in them, but I think I'll see if the drives actually work before spending any money. So for now, oil the bearings then clean and lubricate the other parts of the drive mechanism - head carriage worm drive, door mechanism, floppy eject and door interlock mechanism, Track 0 switch, etc etc. Lots of oil and grease. I took a few pictures to help with reassembly: https://drive.google.com/open?id=1-3H8O ... RqT6HeS87j

The drives have very interesting looking AC motors.


IMG_5349.JPG


The whole outer casing (the black part) rotates, and it acts as a fan for the coils. I think on of them has dry bearings but I can't get at it to oil it. So the drives occasionally make a very funny noise, like they're about to shake apart.
Last edited by jonb on Sun Mar 03, 2019 5:29 pm, edited 2 times in total.

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastore 8" drive unit

Post by jonb » Sun Mar 03, 2019 5:18 pm

Eventually it is back together.

IMG_5356.JPG

Note the key switch, used to prevent unauthorised access to the drive unit. A common thing in mid '70s micro computers, copied (rather unnecessarily) from mainframes, presumably to make the machine look more powerful for marketing purposes. Bit of a waste of time here; you'd have done better to lock the floppy disks themselves away.

The drive units are BASF 6104s, which have a capacity 1600k according to the datasheet, if run as DSDD. That is pretty impressive - and there are two of them.


IMG_5348.JPG
Drive identification label.


But does it work?

Alas, no; at least, not on my TRS-80 Model II. However it is not the end of the story. The drives have many jumpers and options and the TRS-80 is a bit finicky. I have a DSDD Shugart 8" drive that it doesn't like and I thought that was toast until I connected it to my MDS-230 and configured it properly. So there is more mileage in this unit, I hope. It's supposed to be connected to my S100 box (also in the haul), which is pending restoration, so I will leave this as-is for now on the grounds that if it has any special configuration, it will be for the S100's Telek FDC-1 board.

More about that later...
Last edited by jonb on Sun Mar 03, 2019 5:34 pm, edited 3 times in total.


User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastore 8" drive unit

Post by jonb » Mon Mar 04, 2019 11:22 am

Further investigation reveals:
  • Dead shorts across the 24v rail on both drives, and the 5v rail on one drive
  • A proliferation of deadly tanatalum capacitors on the drive boards

IMG_5359.JPG


For the uninitiated amongst you, the darker blue beads on that PCB are tantalum capacitors, which have a particular failure mode - short circuit - so if one is across a power rail and fails, well you can imagine the rest! Hopefully the Astek PSU has protection.

User avatar
flaxcottage
Posts: 3934
Joined: Thu Dec 13, 2012 8:46 pm
Location: Derbyshire
Contact:

Re: Megastore 8" drive unit

Post by flaxcottage » Mon Mar 04, 2019 8:11 pm

Looks like a job for 'Solder Sucker'. :D
- John

Image

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastore 8" drive unit

Post by jonb » Mon Mar 04, 2019 8:18 pm

Well, I found two tants on one board that were shorted pulling these has fixed the 24v short on the drive. All the other tants are fine. Yet I have shorts between the +5 and ground, and -5v and ground on both boards, and the ASTEX PSU wont start up with shorted rails. It seems to be able to withstand them and with nothing connected is outputting the correct voltages on all rails, which is some sort of good news at least.

User avatar
vanpeebles
Posts: 651
Joined: Wed Nov 28, 2012 10:01 am
Location: UK
Contact:

Re: Megastore 8" drive unit

Post by vanpeebles » Mon Mar 04, 2019 8:54 pm

What are you gonna do with Big Bess once it's up and running?

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastore 8" drive unit

Post by jonb » Mon Mar 04, 2019 9:53 pm

It belongs on my S-100 box. CP/M and all that.

Meanwhile, I tracked down the -5v shorts to a single chip - N8T20N multivibrator. It's the only place on the schematic where the -5v rail feeds a semiconductor directly (rather than via a resistor). Both drives had exactly the same problem except one IC had the 5v sorted to ground as well. I had hoped that's be that board sorted but it was not to be.

So, I have 5v shorts on both boards left and all tants are out or tested OK. It must be one of the other ICs. Luckily these drives are the "LSI" versions, which have 40 pin ICs to reduce the chip count (and they are socketed) so I've eliminated them as the source of the problem. Just another 15 ICs to go on board 1, 13 on board 2 (as I already pulled and tested the ICs that share the same circuit as the N820Ns).

I might go round snipping the 5v in of these ICs. They are mostly 7400 series...

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastore 8" drive unit

Post by jonb » Mon Mar 04, 2019 10:18 pm

Hmm, snipping pins caused more damage. Silly me, I should have desoldered the ICs to eliminate them as a source of shorts. Duh. :evil:

I'll switch to the other board now. Maybe I can get one of them going.
Last edited by jonb on Mon Mar 04, 2019 10:22 pm, edited 1 time in total.

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastore 8" drive unit

Post by jonb » Tue Mar 05, 2019 8:27 am

Um. Looks like I wired the 24v and 5v lines the wrong way round at the PSU, and all the ICs on the boards are probably toast. Darn it. No wonder I couldn't locate the short - every IC is probably fused. #-o

User avatar
oPless
Posts: 47
Joined: Thu Feb 14, 2019 1:12 am
Contact:

Re: Megastore 8" drive unit

Post by oPless » Tue Mar 05, 2019 8:36 am

Ah. Another thing to test on my torch boards.


... Dead capacitors. I'll add that to my to-do list! :roll:
BBC Model B ... W.E. Sideways Rom/Ram Board ... GIS Teletext Adapter ... Torch board (non-functional) Gotek Drive.
Currently into Teletext/ViewData stuff...

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastor 8" drive unit

Post by jonb » Tue Mar 05, 2019 3:09 pm

So, some desoldering and fitting of IC sockets later...

IMG_5363.JPG
Those old ICs look like dead insects. May as well be!

5v short has gone and the board measures 500k across the power rail. Once the replacement ICs turn up I should be good to go...

johnkenyon
Posts: 227
Joined: Wed Jul 20, 2011 2:21 pm
Location: Coventry
Contact:

Re: Megastor 8" drive unit

Post by johnkenyon » Tue Mar 05, 2019 5:02 pm

What's the 40 pin device?
(Edit: Cancel that question - found this https://www.mirrorservice.org/sites/www ... manual.pdf
and it says "Large Scale Integrated Circuit BASF 81041-001")

And have you located a replacement?
Last edited by johnkenyon on Tue Mar 05, 2019 5:05 pm, edited 1 time in total.

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastor 8" drive unit

Post by jonb » Tue Mar 05, 2019 6:12 pm

The 40 pin device is an ITT 81041-003 which implements much of the control circuitry. I found an 81041-001 on a UK eBay seller's page and took a punt (I've found in the past that the 3 digit suffix might be functionally significant though, hence "a punt". There is a chance that the one in the second drive is still working, as it is not showing low resistance in the board (whereas the one removed from the pictured board does).

If the -001 variant or my spare don't work, I will have to keep looking. Already I am having difficulty replacing the terminator resistor package (Beckman 898-5 R220/330 dual terminator). There's one in Poland but it is expensive, as are the US sourced ones (shipping).

That said it'd be pretty easy to recreate it as it looks like this (for each line shown in the document):

Input Termination.JPG
Input Termination.JPG (13.54 KiB) Viewed 1284 times
Last edited by jonb on Tue Mar 05, 2019 6:16 pm, edited 1 time in total.

SteveH
Posts: 3
Joined: Mon Apr 10, 2017 8:37 am
Contact:

Re: Megastore 8" drive unit

Post by SteveH » Tue Mar 05, 2019 9:09 pm

jonb wrote:
Mon Mar 04, 2019 11:22 am
For the uninitiated amongst you, the darker blue beads on that PCB are tantalum capacitors, which have a particular failure mode - short circuit - so if one is across a power rail and fails, well you can imagine the rest! Hopefully the Astek PSU has protection.
Do you use anything special to test for shorted tantalums or just a plain common or garden multimeter?

User avatar
1024MAK
Posts: 9728
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Megastor 8" drive unit

Post by 1024MAK » Tue Mar 05, 2019 10:08 pm

A normal multimeter on the ohms / resistance range will show up any shorted capacitors. But can’t detect any which only fail when the supply voltage is present. But this later symptom is much less common.

Mark

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastor 8" drive unit

Post by jonb » Wed Mar 06, 2019 7:54 am

Obviously lift one of the legs off the board before testing. I use the continuity checker (beeper) that's built into my multimeter.
Last edited by jonb on Wed Mar 06, 2019 7:55 am, edited 1 time in total.

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastor 8" drive unit

Post by jonb » Wed Mar 20, 2019 9:07 am

The resurrected board is back in the drive unit and it is showing activity. However, it is not able to move the head carriage. The BASF drives use big stepper motors having 4 input channels, driven by this circuit:

Stepper Driver Stage.JPG

I have got the 'scope out and measured the waveforms on the outputs of the transistors for channel A, B, C and D. This is what I saw:

Green measurement points.png
Yellow=C Cyan=B, Magenta=A, Blue=D

It's pointing to a problem with channel 4. I can see the motor trying to step. It makes a noise and begins to turn the spindle, but doesn't quite complete the action. Assuming the motor pulses should go in some sort of sequence (we do see this in the trace - C,B,A,D), it looks like the output of channel 4 is too weak to complete the sequence.

I've made further measurements at the other coloured dots on the schematic, and found that the channel 4 weak output is present at green and cyan points, but not at the red points. This suggests that there is a problem with (one of) the 7406s which are driving the transistors, but I have replaced them and it hasn't helped.

Is there anything else that might influence the operation of channel D? I did measure R11 at 4.2k which is a little bit low (R8, R9 and R10 are 5.2k), might this be it?

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastor 8" drive unit

Post by jonb » Wed Mar 20, 2019 9:28 am

Swapped R11 out, made no difference...

User avatar
flaxcottage
Posts: 3934
Joined: Thu Dec 13, 2012 8:46 pm
Location: Derbyshire
Contact:

Re: Megastor 8" drive unit

Post by flaxcottage » Wed Mar 20, 2019 9:30 am

I'd definitely look at the Q4 transistor. If it's gain has gone low then the problem you experience would be present. It is also likely that the base junctions are fried, which means that any drive to the transistor is not being amplified.

Also check the diode CR5 for short circuit as that could be clamping the collector voltage of Q4.
- John

Image

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastor 8" drive unit

Post by jonb » Wed Mar 20, 2019 10:09 am

flaxcottage wrote:
Wed Mar 20, 2019 9:30 am
I'd definitely look at the Q4 transistor. If it's gain has gone low then the problem you experience would be present. It is also likely that the base junctions are fried, which means that any drive to the transistor is not being amplified.

Also check the diode CR5 for short circuit as that could be clamping the collector voltage of Q4.
Tested with a continuity beeper (on my meter) and CR5 is not shorted out.
Swapped out Q4 and - hey presto - it looks like it's seeking. I'm a little surprised at this as the weak output was on both sides of Q4 but it could have been pulling its input down. Should've known really, it was obvious enough.. Thanks, John!

However, the test computer (a TRS-80 Model II) just blew up so I can't continue with my tests for now. Great day (not)!
Last edited by jonb on Wed Mar 20, 2019 10:10 am, edited 1 time in total.

User avatar
vanpeebles
Posts: 651
Joined: Wed Nov 28, 2012 10:01 am
Location: UK
Contact:

Re: Megastor 8" drive unit

Post by vanpeebles » Wed Mar 20, 2019 10:49 am

More magic smoke? We will have to change your forum tag to Gandalf :D

User avatar
1024MAK
Posts: 9728
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Megastor 8" drive unit

Post by 1024MAK » Wed Mar 20, 2019 11:53 am

If the transistor has been damaged internally, there is likely to be a short circuit between the base and the emitter or collector...

Mark

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastor 8" drive unit

Post by jonb » Wed Mar 20, 2019 12:44 pm

OK, the TRS-80 blew the 5A fuse in the plug when I turned it off accidentally. It's got a funny power switch that tends to make arcing sounds if you move its lever too slowly. New fuse and the old girl is up and running again.

It looks like the BASF drive is working now! Hooray!

One of the problems I encountered with this is that old chestnut, the floppy cable. It's a 50 way ribbon and it's a bit stiff. I'll have to replace it as it was obviously causing problems. For now though, a quick clamp in the mini vise to tighten the connections up has sufficed.

Must say, these BASF drives are swish. I need to resurrect the other one, now that I've proved they can be restored.

:D

User avatar
vanpeebles
Posts: 651
Joined: Wed Nov 28, 2012 10:01 am
Location: UK
Contact:

Re: Megastor 8" drive unit

Post by vanpeebles » Wed Mar 20, 2019 1:13 pm

Would dielectric grease help?

Kazzie
Posts: 1178
Joined: Sun Oct 15, 2017 7:10 pm
Location: North Wales
Contact:

Re: Megastor 8" drive unit

Post by Kazzie » Wed Mar 20, 2019 1:44 pm

I'd love to see these working in anger when you're got them all sorted...
BBC Model B 32k issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
Acorn System 1 home-made replica

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastor 8" drive unit

Post by jonb » Thu Mar 21, 2019 1:41 pm

vanpeebles wrote:
Wed Mar 20, 2019 1:13 pm
Would dielectric grease help?
What for?

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastor 8" drive unit

Post by jonb » Thu Mar 21, 2019 1:43 pm

Right, next up is Drive 1. This needs the full monty IC swap as well, but I can test all the ICs I remove now as I have a working drive to substitute chips into for testing. It's desoldering time (again)!

User avatar
jonb
Posts: 2590
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England
Contact:

Re: Megastor 8" drive unit

Post by jonb » Thu Mar 21, 2019 5:03 pm

Stage 1

Remove all the old ICs and fit IC sockets.

Drive 2 board.jpg
Drive 2's board prepped up

Got a sore back now! :)

Post Reply