IBM PS/2 model 30 - Floppy and HDD repair
Вставка
- Опубліковано 30 лип 2024
- Let's continue the repair/restoration journey of my IBM PS/2. On this video I'll try to repair the Floppy drive and the Hard drive.
Thanks to our sponsor PCBWay: pcbway.com/g/M525r4
Previous episodes:
Part 1: • IBM PS/2 Model 30 - Un...
Part 2: • IBM PS/2 Model 30 (808...
The IC I am looking for: M52800FP
Mentioned in the video
Vogons: www.vogons.org/
The Retro Web: theretroweb.com/
Epictronics: @Epictronics1
Macintosh SE repair: • Apple Macintosh SE Rep...
Electrolube SWAS: electrolube.com/product/swas-...
Glitchworks XT-IDE: github.com/glitchwrks/xt_ide
00:00 Intro
01:01 Capacitors
04:08 Erratic behaviour
07:08 The index pulse
09:20 Faulty logic IC 1
12:20 Faulty logic IC 2
14:30 Testing
16:00 The floppy connector
19:50 The Index pin and I'm stuck
22:45 The HDD
23:28 Oiling the bearing
24:11 The corroded ribbon cable
25:00 Spin the spindle with a rubber
26:16 Is the drive dead now?
27:00 Capacitors 2
27:42 XT-IDE
28:57 Install DOS 3.30 on CF
33:54 Washing the board
34:55 Outro - Наука та технологія
Checked a couple old drives in my spare parts bin for you but dont have the chip in question.
I didn't mind the voiceover, do whatever suits you😉
Thank you so much for checking, for the feedback and for watching, appreciated!
I have an IBM PS/2 here that’s probably not seen any power on for the last 20 years or so. I have already been warned about all the drives failing on these machines so I found this video. I will now replace the capacitors before testing the machine to prevent any possible damage to the ic’s
Good luck!
Awesome video Tony! Your patience and perseverance is incredible. I can't believe how many issues you discovered with that floppy drive. I hope you get that last IC replaced and the drive going again. I learned a great deal watching you work on it! 🙂
Thank you for your kind words! I'm still looking for one - that said the drive is now "reliable enough" for casual use so all that time was not entirely a waste!
Always a pleasure to watch your videos. As soon as I saw the speed on that tacho, I though "It's the motor controller caps". Had exactly the same issue on my JVC TD-V1010 cassette desk. This has direct drive capstans, so if the caps fail (as they often do) it plays at double speed. I replaced them with some tantalums in the end. As they fitted the SMD pads perfectly.
That would have been an easy repair - but the IBM "Complication dept." did their best to keep me entertained! :) Thanks for watching!
Great video Tony, I don't mind the voice-over style at all.
Thank you Ted!
Man, I have three of those drives in external cases her for repair. They are from a friend and I have done my head in over these things for 3 years now. From what I have seen and read you are the person who has gotten the furthest with diagnosing them. Sadly I cannot offer them to you as they are not mine, in fact he has contacted me to pick them back up next week.
I feel pretty damn bad about not being able to help him, but it is some sort of consolation that, even if I had gone a lot further than it did, it might still not have been enough.
Sad indeed, I hope your friend can fix them! Thanks for watching!
Rough week! But it made a good video :) Thanks for the mention. Your HDD sounds like my Model 50 HDD. I opened it up to have a look and the top layer on the platters had corroded
I'm planning to take a look too. "Week"? I think I worked on this thing for a month... So frustrating! The IBM "complication dept" did an excellent job here :)
@@tony359 Haha, yeah, those guys lol
@@Epictronics1 that day they were working on my floppy and your motherboard! :)
@@tony359 Did I mention my other board that still has a parity error even after I had replaced all the RAM chips? :D
Great troubleshooting! - Something I need to test om a few drives myself.
Thank you!
wonderful, just wonderful
Thank you!
Great trouble shooting video! Looking to the number of damaged ICs on the FDD, I guess +5 and +12 were once reversed. No idea how this can happen with the IBM connector, but I had an ordinary AT supply where one of the outlets cables had 5/12 swapped. I've killed 3 unknown HDDs one after the other before I've recognized the problem. Trying to repair them but all +5V logic ICs got 12 instead of 5V and I gave up.
Interesting theory. The connector is keyed so it should not be possible. Unless it was plugged into something else. The MB is working so that would rule out a PSU issue as well.
It could possibly be a "spare" PS/2 where IT parked a broken floppy drive.
Thanks for watching and for your kind words!
You can get the Mitsubishi chip from UT Source. Wouldnt be the first time ive had to change a main IC on a floppy drive. Had a PS/2 drive that had its motor servo fail as well.
Thanks I sent an email but I find those websites a bit unreliable. I somehow doubt they have 33.000 of those chips in stock - if they do, well they have lots of space :)
The page suggested a quantity of 100, again I don't think I want to pay hundreds for that drive which is working-ish. Mostly because I am not 100% sure the issue is that IC. But let's see what they reply, thanks for your help!
@@tony359 I find UT source to be a savior of these old machines, I've had to order from them multiple times to keep things going
@@THEtechknight That's amazing to know thanks. I thought they were all scams to be honest, those stock numbers cannot be right! Let's see what they reply. Thanks so far, I'll keep you posted.
@@tony359 crossing my fingers hope it works out! But I've ordered a handful of various 82C series chips from them in the past to fix dead motherboards
@@THEtechknight Unfortunately the quotation came up with no availability... Thanks for trying!
Hi Tony! This is by far the best and most thorough treatment of these drives I've seen. Thank you so much for posting! I have two of these I've been fighting with -- one is now working (touch wood) but the other has a stepper issue, either in the mechanics or the logic. It seems to step in one direction but not the other. I'm wondering if it thinks it's always at track zero... it might be time to buy the rite-of-passage oscilloscope, but I'm a little daunted :) I don't have your experience or expertise! Anyway, thank you again for your content, it's splendid stuff!
Thank you for your kind words! AFAIK the stepper would need both coils to step in either direction so that rules out a stepper issue.
Does it try to step at all? Which direction does it step?
@@tony359 Hi Tony! It's odd -- it will step upward (ie, away from track zero) pretty reliably but only intermittently will it step or seek back. When using diskImage I can request steps and seeks away from track zero but not back. On a repower, it will sometimes seek back to zero and sometimes not. I sprayed contact cleaner on the track zero sensor and it started working properly for about a minute, then failed again. (cooling effect?) I've reflowed the ics on the mainboard, with no specific effect. There is an adjustment screw on the back of the stepper motor -- do you happen to know what that controls? It does really seem similar to the logic IC issues you were having, with intermittent function.
Best wishes, A
The track zero is an IR sensor, did you try cleaning it? You can use your multimeter on the track zero pin on the connector to the MB to monitor when track zero activates (goes to 0V if not mistaken otherwise you should read a solid 5V).
The screw sets the friction on the stepper shaft. If it's too tight it might cause what you observed. My understanding is that it needs to be tight enough to prevent mechanical play but not so tight that prevents the motor from operating. You can loosen it a bit and see if the spindle operates both way - then you'll have to gently tighten it and lock it with some glue or similar.
The stepper is controlled by a driver, it would be interesting to see whether it's a stepper issue or a logic issue. But yes, it's a similar issue I had on mine. You could try swapping those logic ICs... But first check the track zero line, you don't need a scope for that.
@@tony359 Hi Tony! Thank you for this -- I took this info and was checking voltages on the IR sensor -- I hadn't yet found a reliable pattern, the voltages seemed to be floating a lot between 1V and 3V (it may have been fine, I was just trying to get the pattern of how it worked) when there was a mild burning smell and, sadly, the stepper no longer steps. I likely screwed it up by adjusting the stepper knob, tho it wasn't tight. Or it was marginal and I sent it over the edge. Anyway, the stepper, she be dead. No visible burns on any components elsewhere, so that's why I theorize the motor itself. Anyway, that ends that little adventure lol. I will keep it for a parts drive, or, if you'd like it, I'd be happy to ship it over! Might be a simple fix, or the parts may be useful to you.
awww, so sorry to hear that.
For the IR sensor, I would check the TRACK ZERO pin on the main connector. That it's 0V or 5V depending on whether the sensor is covered or not.
For the stepper, there are two coils. If you unplug the plug which goes to the stepper, you should be able to see the two coils in resistance mode.
Are you in the UK?
About that very early SELECT command, the only caveat of what you did is that it installs DOS in the main directory. There was a way in that command to make sure it would install into a dedicated directory such as C:\DOS. Of course, with DOS 4.0, this was fixed by a revamped SELECT command that was a genuine setup programme (followed by SETUP from DOS 5.0 on).
I did notice that. I believe I found instructions online to add the destination folder for the OS? It was an Olivetti user guide.
That said, DOS was indeed in the main folder on the PS/2 I used to use so that's also fine! Brings up memories! The A> prompt is also cute - even though it was always changed in autoexec.bat with Prompt $p$g :)
@@tony359 The prompt command wasn't even integrated in the very early versions of DOS. I don't know when it was integrated into the command interpreter, but I also get a bit triggered if it just shows A> instead of A:\>. I need that prompt $p$g. I play around on PCem, and when I install DOS 3.30, that command is one of the first in the new autoexec.bat.
@@tony359 I am going to have to go beyond asking you to use a pencil as a pointer. Graphene is an excellent conductor of electricity. and even a tiny particle from a simple tap of the pencil can cause shorts. Stop using a pencil as a pointer.
@@InconsistentManner I’ll use a biro next time! Thanks for watching!
@@bundesautobahn7 coincidentally I fired up the PS/2 earlier on, didn’t bother to change the prompt, navigated to a folder and I was like ‘why didn’t that work?’ - took me a few seconds to realise that the standard prompt doesn’t show the path! It’s so quaint!
Could you implement some kind of bodge for the index pulse, and wire the output of the AND gate to the input of the '368, bypassing the Mitsubishi chip entirely?
Assuming of course that it is the correct pulse and in the correct phase.
Yes, another viewer has mentioned that. That said, if you watch the pulse from the Panasonic, it's not there all the time. So there might be some logic happening behind the scenes. Still, could be something to be tested even though I would feel pretty bad if I broke the drive or - even worse - the motherboard! But good idea! Thanks for watching!
Fantastic! Liked. Isn't there a way to retrofit a standard floppy drive into the metal chassis of the PS/2 drive? I'll look through my stash of old floppy drives to see if I can find one with that IC, but I seriously doubt it.
There is! The interface is identical, I used a Panasonic on my previous video: ua-cam.com/video/x0zQYWQsnz0/v-deo.html
I can find 3D printed adaptors and everything.
However... I don't want that! I want the original IBM drive! :)
For the HDD is a different story, those WILL fail.
Let me know what you find in your stash and thanks a lot! :)
Did you try make this IC hot and cold as with other chips?
Yes, multiple times...
the counter on the motor is that a magnet. could be weekend to much. remagnify it :)
Maybe adding an 8087 coprocessor would make your life easier there by speeding up that slow boat, otherwise very nice video, and what if you bake the floppy board containing the 2 unable to find chips cause heat has an effect and correcting internal IC problems. Also they may come to their senses on their own with frequent usage of the drive!
The HDD sounded completely like an angle grinder, maybe opening it for us and see what happened?
the 8087 could be an idea! The final video is coming and will have more about the floppy and the cursed chip! That said, I did warm it up during that video and it would make it work/not work intermittently. I have used the drive a bit and... it usually works enough so it's usable. I still cannot format a disk in there but to transfer stuff it's fine.
Thanks for suggesting the HDD inspection video, I think it's a good idea! Nothing to lose I suppose.
@@tony359 Could this format situation be a termination problem with any jumper on the drive? Like mfm hdds? Just a guess... or cable or controller jumper something
No unfortunately, I tried that and compared to others.
The next video is half-done and... it's interesting! Watch this space!
I wonder whether you'd be able to pick up the index pulse directly from the coil, amplify it through a transistor if needed then feed it to the buffer chip.
Might require a bodge on breadboard to be left in there, but if it works it works
Someone indeed suggested that. That said, I believe I noticed on the Panasonic drive that the index pulse is not always present on the INDEX pin. I'm not sure why? That might be why the signal is coming from that IC and is not simply amplified and fed to the buffer via some simple logic ICs. But your idea is sound! Thanks for watching!
Strange indeed, you'd expect the conditional output of the signal to be switched by one rather than two logic gates.
There must be some needed extra logic for that pulse to be valid. That said I have a feeling that doing what you said might work. But I might get a replacement soon so finger crossed. 🙂
Hey Tony
I don’t know much about electronics, but can the Signal be kept low because of a fault on the mobo side?
I enjoyed the video, thank you!
Well the Panasonic drive works fine. The pulse is there, I can format and run the IBM diagnostic on it.
Someone mentioned the Drive Select pin - but that works fine.
Thanks for the input though.
just had a quick Google of the IC - found the website Digchip had reference to the IC and it said there is 5 suppliers (not sure how truthful that is!!) Worldway Electronics have 38 available - no idea on the price each but minimum order of 100 USD
Thanks, found that. I really doubt that someone has 40.000 of those chips in stock! It won't hurt to ask - with the inevitable spam coming back to my email address!:)
As much as I like my PS/2 though, I don't think I would spend 100USD on a chip which 1. might not work 2. might be fake 3. might not be the problem with the drive!
Thanks for your input though, appreciated! :)
Perhaps some mechanical problem with the stepper on the floppy drive?
Anything's possible. I did look into that - it didn't make into the video - but it looks fine. To be honest, the fact that the "index" pin is missing makes me think that something's wrong on the logic side. But who knows. Thanks for watching!
With this HDD I saw some chip with white sticker on it. I guess it was eprom. I've read that it could lost some bits after 30 years.
Yes, I believe it's called bit-rot. That's another option indeed. Thanks for watching!
Long time ago, i asked an American female classmate of mine for a rubber (meaning eraser) and she kinda looked at me in a strange way. 🙂
Oh, I could make a whole video on my language "gaffes" :)
Funny thing! Doing this work on 2 720k drives these days myself. One might be working as it should now, but the other has similar issues as this one. The stepper motor is not moving at all. Strange stuff
IBM drive?
@@tony359 yup. PS/2 Model 30 :)
@@tonybardalen Glad to hear I'm not the only one then! :)
@@tony359 we should get to the bottom of this! Tonys unite 😂
@@tonybardalen "There can be only one" (Cit. Highlander!) 😄
Great video and format. (Not a fan of the background music though) 👍
Thank you! I appreciate your feedback on the music. Is that particular song (which I keep re-using!) or music in general?
@@tony359 Repeating music is always grating.
@@tony359 Ah just the same song every time I guess. I've watched all your videos all the way through so it's stuck in my head now 😄
@@NiTye357 I very much understand! I used to look for a new one every time but it's not easy. It has to bee free to use and I need to like it! So I decided to keep this one as a "footprint" of my videos but I can also understand it might be boring in the end. I'll see what I can do thanks for your feedback!
@@No_soup_for_you Point taken! I'll see if I can find a few I like and cycle them!
The 32MB limit is actually a limit of DOS 3.30. If you want to have more space, there's a special DOS from compaq iirc, or you can use DOS 4.0 (which has its problems ...) or DOS 5.0+
I suspected that! I really would like to keep this beast at 3.30 so 32MB it is :) Thanks for watching!
I believe DOS 4.0 (or 4.01 if you want to avoid the issues of "vanilla" DOS 4.0) introduced modern FAT16 with up to 2 GB, aka FAT16B. The 32 MB limit was from FAT12. And about 32 MB vs 2 GB, even if you use a bigger harddrive, you can't use the full 2 GB anyway because of most AT based mobos having that 500 MB limit in the BIOS (with late 486s putting an end to it, I presume). And even if you wanted to use the full 2 GB partition size, you'd need SCSI for it.
I didn't even know "FAT12" existed! Thanks for that, it's nice to know. I am ok with 32MB on this machine. I'm not looking to improve it. I only would like to have it as it was back in 1992 when I was cycling to school in the afternoons to use the computer labs! :) Only things missing are: the squeaky HDD, the monitor and the keyboard!
Ah, and also some space to store/display the resulting useless system! :)
@@tony359 It's one of the early File Allocation Table versions. It went through different iterations until ending with FAT32 in Windows 98 and the eventual switch to NTFS by Windows XP or Windows 2000. FAT12 is an evolution of early 8 bit FAT, which had 6.3 file names before the more common 8.3 was introduced. The rest, as they say, is history.
looks like BudgetAdrian.
ahah yes, I'm just the padawan! Thanks for watching!
I really didn't want to give you a thumbs up because the video is ironically at 286 upvotes. So, I made it a 287.
Ahah now I need a math co-processor! I don’t mind if it becomes a 386 or even a 486 by the way 🙂 thank you!
A bit of teaser for your dead PS3 CELL:
ua-cam.com/video/38pVcxI5U8E/v-deo.html
I still need to try with non-enamelled wire. Does it say if it's enamelled or not in the video?
@@tony359 Maybe someone who speaks polish can tell, but looks like non enamelled.
@@worroSfOretsevraH It looks like that. Thanks for the link!
I think I found out from one of his other video:
ua-cam.com/video/Q_z3hkQG28A/v-deo.html
"MECHANIC FXV08 500m 0.008mm Superfine Silver Flying Line" on ali.
VERY nice, thank you!
I've ordered it, it's an order of magnitude smaller than the one I have! Finger crossed!
@@tony359 Nice.
Yes it is suuper thin. You need a micropencil iron for that. I hope you make a video. Good luck!
@@worroSfOretsevraH I can only imagine. I got the solder tips as well. I'm not sure my tweezers will be good enough! AND my hands!
The wire has arrived and... well I need to get a better microscope now I guess! I mean... wow!