Thanks for bring back memories. In 1987 I worked at Miniscribe Hong Kong as summer job and repaired a lot of 8425. After you power on the hard drive, it would probably do a self diagnostic like spinning up the disk, check it’s rpm, move the stepper motor for track 0 and the amplifier circuits etc. If any problem found it will flash error code via the LED. Didi you notice any error code? Besides, From my poor memory there used to be quite a lot of problems with those inductors as well(resistor like components in green).
@@tony359 You have the 8425S right? I found something here: 68kmla.org/bb/index.php?threads/how-can-i-troubleshoot-a-miniscribe-8425s-hard-drive.42497/
@@Epictronics1 Amazing thanks. I'll have to check that one, I think it showed 4 solid flashes - hence Error F: unexpected interrupt from processor? Uhm... Thanks for now!
I'll add this for anyone that reads the comments and comes across this in the future. These drives can also be put into a self test mode when you close j13. You can then power up the drive without being connected to a computer and maybe after a few power cycles it can come back. I've seen this referred to as exercising the drive. Adding a couple of drops of oil to the stepper can help also.
But don't put too much lubricant in, it can penetrate all the way to the platter -- a tragic fate for a drive which is actually in my experience pretty reliable! I have a couple still working, with no more bad sectors than are listed on the label...
Even though the repair is stalled for lack of direction at this point, I want to say thank you very much for the video, because it provides a guide of what to check and what else to check if nothing seems to be the problem. That is a big help to people like me who know how to use a multimeter, but don't know much on how to find the problem beyond checking all the capacitors.
Valiant and good effort! On the tantalum caps, they are almost always used for decoupling, meaning that they are connected to ground and Vcc. It really is totally meaningless to try to measure their capacitance in circuit, but measuring the resistance from Vcc to ground (without removing any component at first) can give you a valuable indication if one of the tantalums is busted. When they go bad, they become resistors. Then they heat up and the electrolyte inside (tantalums are electrolytic) can vaporise and steam out, either catastrophically, or more slowly, by breaking the seal around the leads. In this latter case, you can only see the physical damage around the leads under magnification. Regarding the short leads of your ESR tester, any longer and they become inductors, reducing the reliability of the measurement.
Thanks, that would totally explain why I had magic smoke without any visual clues! Question then is, are those smallest radial caps tantalums or ceramics? Can I trust that very small tantalum caps have polarity markings?
My first job in 1979 was for a Canon calculator dealer in the USA. By the time I got involved they had single chip multi-function display and printing calculators. The calculators they made in the early part of the decade used discrete components, some had multiple boards for the various functions such as addition and multiplication. They also had programable versions. The earliest used wire wrapping, then later had paper cards, and then magnetic stripe cards. The first computers of theirs evolved from their calculator lines, eventually looking more like the personal computers of the day. They also had a typewritter division that made dedicated word processors. It was interesting seeing the custom computer development in that area. The first MS-DOS/CPM-86 compatible PC was the AS100. It was not IBM PC compatible. The first PC compatible that I saw was the A-200. So it was interesting to see your EX version. The Canon CAT was developed by Jeff Raskin after he left Apple. I was no longer involved with the Canon at that point, but there is a lot of published information on this Mac cousin. It's a shame that a lot of Canon's PC history is diluted and unavailable. They had so much to do in technology, often in partnerships with other companies. They weren't just cameras, printers and copiers. They took the personal copier and made a personal laser printer used by Apple, HP and so many others. The made components such as motors and disk drives.
Yeah, Canon's history and the line-up is quite bloody impressive. I have had so many Canon products over the years and still do. Cool to hear that there was an A-200
Pieces of black debris would suggest a diode or transistor. The tantalums, in my experience at least, wouldn't let you off with a just a few pieces of debris and smoke, they'd go with a light show and leave a nasty skid mark behind, maybe even a proper burn hole. I'd check below the ICs, next thing. And also check the pins for proper values.
I don't think there are any data sheets for those custom ICs unfortunately. But I could definitely lift all black components of the board and make a visual inspection
Not sure if you did, but I'd do a low resistivity check across the capacitors etc... firstly as well as doing a capacitance check, since the motor spun up I wouldn't suspect that side of the circuit as most servo style motors/circuits probably wouldn't spin up with a blown MOSFET or be as quiet as you've mentioned if it was faulty. Like Dave Jones says away check your voltages, even without a schematic you can still test Vcc levels to 74xxx logic chips etc.. lastly like others have said the drive might be fine and known good system might be a better test bed if possible.
The drive may be fine. Boot from a DOS floppy containing the DEBUG program and launch DEBUG from the command prompt. Type in G=C800:5 This should bring you into the controller cards low level format program. Let it format the drive and then you should see it in FDISK so you can partition and then high level (DOS) format it.
The debris found in the board looks like the bottom bits of one of those IC's. I have had that one bite me before. Big mystery hidden behind a door! On another note, I have one of those drives and for some odd reason it was picky with the power supply that ran it (I figured cap issues in my case but the drive still "works").
Yes, could be. The problem is that those are custom chips and don't have a data sheet with the pinout to check. We might lift them off the board to have a look underneath
@@Epictronics1 I forgot to mention before, but have you tried a jumper on 13a and left the drive run for a few minutes yet? It is a simple internal test for the drive itself. Also, if you need readings on certain parts of the drive please let me know. I can scope a few things out for you. My drive has a dead head now but the PCB seems to function just fine. Your video made me dig it out of the stack.
I’ve got one that I haven’t tried powering up yet. I want to do a full disassembly, clean, and recap first. I’ve only done my 5151 so far. Hope it actually works! Looking forward to your video as a heads-up for what I should expect to find in there.
The readings you got from that tantalum remind me of a Vanguard arcade machine where the speech only worked sometimes. I pulled the tantalums in the op-amp and checked them with my Testofon (a device that transforms measurement current into audio frequencies, aka the best piece of test equipment after the oscilloscope) and I essentially got random music. It was random, it was unpredictable and measured between ~500 Ohm and 5kOhm (measuring the leakage, not the ESR!) and if I had left it running until the end of the universe it would eventually have played Never Gonna Give You Up from start to finish. Note again on the Testofon, a capacitor should make a quick descending sound and then fall silent.
haha, Rick bloody Astley, That Testofon is quite an interesting devise. I think I have seen you use it in a video. The best lead so far is that one of the caps has broken the seal around the leads and blew out the magic smoke without leaving much of a trace
@@Epictronics1 There really isn't* a lot of information on the Testofon, so I thought I'd create a video showing how to use one. The company (Taco-Nauert) that makes these (they're now called Contitest) thanked me with four free Testofon kits. Two went to Repair Cafes, one to my local pinball and video arcade and one to a less local pinball arcade. Btw. there were two tantalums in that circuit and they both did the same random music stuff. You got lucky the other tantalum you measured was fine, otherwise you would have thought that "weird jumping around measurements" are normal on these... *)correction: wasn't. Now that I look, I can see two more videos where people actually use those. But mine is the one on the For Amusement Only e.V. channel which is my local pinball and video arcade's channel. (btw. the guy behind that youtube channel has a second channel daddlertl3. He's created the BreadAmp (Winamp for C64) and a whole new C64 video format called Koalavideo which plays videos at decent quality with sound at almost FM radio quality - I asked him to convert Weird Al's "It's all about the Pentiums" music video so it now runs on the C64! Turns out the Commodore 64 IS really neato! 🙂) Sorry for that Wall Of Text.
@@Epictronics1 You'll probably have more fun with Koalavideo - he coded tools and wrote a manual in German and English on how to make your own Koalavideos. So maybe your next TH-cam video will stream from a C64 🙂
Cool diagnostic video! If my miniscribe has surface issues and can't be fixed (I'd like to give it another go at some point), maybe you could have my PCB if they are identical?
Thanks, that would be great. But I hope you can fix it! Hmm, I just had a thought... Did those old MACs have 8425s in them? I think I have at least three compact MACs with HDDs. I'd better check EDIT: I just realised, they all have SCSI, don't they?
Ah yes, they are SCSI... Bugger! Mine is a 8425 though. I'm not optimistic for mine. It went through hours of surface scan, bad blocks keep coming up and it's totally unreliable when I try to copy things on it... But I'd like to give it another go.
modern way to find shorts is to inject voltage see what is getting hot via soaking the board in ipa or heat flare cam at this point this is your only option
@@jacktheripper6716 I use this method frequently when messing around with Commodore 64s. I just power it on and use my fingertips as a sensor. If I burn myself, I'll replace that chip haha. I guess it's about bloody time I get a thermal camera :)
i have a dead pcb for one of those...if you need a part or custom ic etc feel free to ask i can give it for the price of shipping.....i also have a drive that works but has very worn bearings...i have it for parts for my other working drives...
Hook up an LED directly to the drive and see if it blinks an error code. If not, put the drive in another computer with another controller, at least see if the drive can seek, using software like Ontrack's Disk Manager, which comes with a DIAG program allowing you to seek the drive or format a high cylinder to see if it is responding at all. Speedstor might also come in handy. If you are willing to have data loss, try to low level format the drive with DM or Speedstor.
Common for those early 3.5" drives, it could be the interface chip next to the MFM connector, and if that is the case the drive hears nothing from the PC, Try replacing the SGCC 004460100 chip and see if anything changes.
Thanks, that would explain the small black piece of debris I found on the PCB. I'll pull it off and check underneath. Unfortunately, I can't find that chip on any of the usual places (eBay, Ali, amazon, or mouser
The Mobo Bios hard drive setting was not mentioned, and I am curious if the mobo has a hard drive detect bios or not, if not the hard drive could have been placed in a computer that does have a hard drive detect.
Unfortunately, I can't find any info about this Canon or its BIOS online. So far, we have just assumed that it's the original HDD. We have also assumed that it has an HDD auto-detect. Simply because we have no choice. Next, we are going to mess around with debug
Have you tried mashing the delete or one of the function keys as you switch the computer on to get into the bios to see what type of hard drive settings have been set? Anyway, most modern PC's have a section in the bios to auto detect and set the settings or allow you to select what the hard drive setting should be for it to be detected and therefore operate properly, So once you know what settings of the hard drive are by getting a more modern computer to detect it, I would make sure nobody inadvertently changed those bios settings in the Miniscribe. Just be aware some old hard drives can be detected incorrectly by newer PC's ergo sending you to pull out your hair to get the damn thing bootable... 😇
@@KorAllRBare Thanks, I have tried with all F-keys, ESC, and DEL. No luck. Next, we'll try to run debug from DOS. Maybe, we can check/set the parameters that way
@@Epictronics1 long shot as it's rare but did you try all the arrow or direction keys like it's an AT PC.. Does the Hard drive have its settings on it's label?
@@KorAllRBare This is the first I hear about arrow keys to enter setup. I thought AT had to have a diagnostics diskette? The settings are available online
I would try the controller on another machine without any other hdd. Just to make sure that the board is working, check with MSD if there is the option ROM from the board loaded. With debug can access LLF routine from its ROM (G=C800:5). If all ok, connect the HDD and try fdisk to see if is there. Also can use Disk Manager from Ontrack (WinWorldPC) for that. Try to see if any data is readable and if not, LLF it with any tools (debug or Ontrack DM). If all is ok, you WILL need the setup/config diskette for that computer. Alternatively, you may try XTIDE ROM/board to use an IDE HDD.
Most 16 bit MFM controllers don't have a format routine in ROM and were actually formatted with a software which was supplied with the hard disk, or with an AT BIOS which had such functionality integrated (AMI-BIOS comes to mind).
@@diskettenfett3161 Ok, thanks. So, do I need to low-level format the drive in a system with AMI BIOS? I'm pretty sure I have a few working 486 boards with AMI BIOS
When you checked the hard drive controller, and you said "...I see something suspicious...", I thought you were referring to the "Made in Ireland" sticker.
i dont see you try different hdd control cards if you did them reply to ic chip with sticker on it wide one it sent data to hdd control cards if control cards have issues it will cut off reading to hdd and replace hdd control card to see if still working with hdd if no them that ic chip with sticker on it just save yourself a headache from stressing out looking for issues judging of smoke coming out of psu feeding power to hdd control card to hdd could been burn out one way fix it just put in different hdd control card and ic chip wide one to see any different if yes there other way save it
I was expecting fdisk to show a drive, even if disfunctional. To me this implies the controller card might be faulty. But I assume you have good reasons to trust it? If so, to me, the next step would be to check why the controller card doesn't even detect the presence of a disk drive, so it can report it back to the PC in turn. Doing that by starting at the cable interface between the two and trying to understand those signals. However I have no idea how the MFM standard works, and this could be quite the steep learning curve and a lot of effort to hunt down the fault this way. This is still a starting point however.
Yes, It was pointed out to me that the controller should see the drive, whether it's broken or not. I'll bring out a tested controller for the next repair attempt. I didn't have one easily accessible while troubleshooting
I have a similar Miniscribe drive housed in Dataframe SCSI enclosure. It has a dedicated board that converts MFM to SCSI. Only problem is that the drive suffers from stiction. I have to remove the HDD cover and physically rotate the platters to get them going. After which they'll work fine. It'll even spin back up if turned off for a little bit. However, if it's unplugged for any significant amount of time, it won't spin up any more, and I have to give it a jump start. I don't know what causes stiction. I saw a TH-cam video some time back where the person investigated it. Since his drive was broken, he went ahead and disassembled it completely by removing the platters and everything to try and figure out what causes stiction. After it was completely disassembled, the platters spun fine. The bearings and grease seemed fine, and the platters spun freely. So he thinks it could be something metal become magnetized, or perhaps something has degraded allowing the neodymium magnets to interfere with the drive spinning up (do HDDs have a "parking brake" or some sort?) I know that quite a few HDDs suffer from stiction. It'd be great to definitively figure out why, and then have a fix.
@@unmanaged Oh, I hope you're wrong. Replacing a bearing in a cast aluminum case that small would require a high-precision press. I'm not even gonna bother checking what one of those would cost :o
@@Epictronics1 Andrew beat me to it, IIRC if that controller is expecting a drive, even with one unplugged DOS will see one present but error out when trying to access it. FDisk instead didn't see one at all so I'm betting getting into the controller's setup via debug will let you restore the conf.
You have to know the specifics for a given controller, but the basic idea is you use debug in DOS to manually jump to the setup util or low level format routine in the controller's BIOS. So, debug followed by -G=C800:5 for example. I found the WD1003WAH manual online, looks like it has a low level format routine, but not a setup routine in it's ROM, it relies on the host PC to determine drive parameters. So this means a setup disk for that system is likely going to be needed unfortunately.
Could be. I'll check with the scope in the next attempt, along with the voltages on the 74 logic chips. Unfortunately, we don't have the pinout on the custom chips
First of all, there may be a problem in way too many places starting from an accidentally moved dip-switch. You've ruled out obvious problems. There is no reason to continue randomly checking every component. You need to get a working AT-computer with working MFM controller and drive. Those are not too hard to get. First try finding which component is bad. I'm going to guess it's the drive. If that's the case: it may be a VERY long job to diagnose a hard drive. extremely long without a repair manual. not sure if it is worth the time. there may be unique data on it still though. you can try getting an exact copy of this drive, but working, and replacing the board. First check that every motor works (from the looks of it - they do) And also the diode is most likely damaged. Trace what it's connected to, that is most likely damaged as well. Try disconnecting it from a PC first and scoping the pins. You will need to assert DRIVE SELECT 0, or whatever it is configured as (see drive manual/controller jumpers) is READY asserted? is INDEX getting a stable signal? Is its frequency consistent with RPM? try to STEP (set DIRECTION). does that work? etc..
Good to hear that there are knowledgeable people out there, this drive might have a chance with some help. It was mentioned that the diode's black legs were normal but I'd be happy to replace it in the next attempt. (and checking what it was connected to) I have a few systems with working MFM and plan on bringing one out to test the drive/controller individually. Yes, getting another miniscribe 8425 would be immensely valuable for this repair, especially since it seems to have custom chips
@@Epictronics1/videos Actually, I'm not *that* knowledgeable myself, just what I heard/seen/understand. But I remember seeing diodes like that, all of which got overheated and destroyed. It's not a far stretch to assume it may measure fine on the multimeter's microamps but not work properly under load. Maybe in your case it's not dead, but suspect nonetheless, so still It would be worth to replace it at least to make sure it doesn't fail catastrophically in the future damaging something else. And if that diode overheated, it's reasonable to assume whatever it is connected to caused it to become like that or was damaged as well. I'm guessing it might be related to stuck spindle but who knows?
@Mr Guru I guess it's a thing :) Trying to guess something that actually makes sense lol. To be honest, I don't think for long, but I like to have a reason to try something in a troubleshooting
@@Epictronics1 havent watched the video yet, most common failure components on drives are diodes , small ones (smd) and inductors or resistor arrays. if the drive spins and seeks then most likely it is fine. if it has a problem it will throw an error by blinking the led repeatedly on a pattern...if no problem the led should blink then stop after the drive is ready..... if you test the drive on an AT system make sure to disable all ide controllers or you will get conflicts and the drive wont work. if you use a super IO card enable just the floppy controller and disable the IDE by means of its jumpers...if your board has integrated IDE then disable them in the BIOS...make sure to set all the drives in BIOS to "not installed". the mfm controller bios will takeover...but for maximum compatiblity i would definetely test this drive on a XT or clone...i use my amstrad pc1512 to test the mfm drives.... sometimes smd components look like resistors and are diodes, sometimes they look like transistors and are diodes or inductors...etc...read the silkscreen and figure out before testing......if the drive spins, seeks and the leds blinks then goes off and stays off then its time for low level format , then fdisk and format again....then after its working use spinrite on it on level 5 to exercise the drive a bit and "refresh" the magnetic surface
I vaguely remember there were issues swapping drives between different brands/models of controller cards. Possibly something to do with the interleave that the controllers support, requiring a low-level format to make it work. But I don't know if these are the symptoms you would see if you used the wrong controller with your drive.
I have considered low-level formatting but held off. Perhaps I'm optimistic but I want to see if that drive has any cool software before I do a clean install. Apparently, that controller should see the drive if it's configured correctly, whether it's broken or not
Well, I'll leave it up to you to judge whether it's remarkable or not. I think it's a good system. Solid, well built with good specs. I think It's quite rare with a seemingly unknown system made by one of the best-known tech brands in the world.
you have at much time as you need, you dont have to pay any rent or anything anymore, be self-supported, best be God supported, I dont care which, as long as you are self-supported, and not supported by any other people, even through money, time for you to go self-supported and independent, and money economy trading for "support" is not it, freedom and then give freely, says the Lord to you, and Lord is Spirit
content is fine, but it must be free content, not made in any way "to have to", anything, even based on illegal country laws demand obligations (against-God given freedom), and anyone who claims to own anything, keep what you have
![โอ้เธอช่าง... - บี้เดอะสกา ft. ต๋อง เทวัญ (Prod. by ป๋าเพชร) [Official MV]](http://i.ytimg.com/vi/zbWhiLFmVL0/mqdefault.jpg)
Thanks for bring back memories. In 1987 I worked at Miniscribe Hong Kong as summer job and repaired a lot of 8425. After you power on the hard drive, it would probably do a self diagnostic like spinning up the disk, check it’s rpm, move the stepper motor for track 0 and the amplifier circuits etc. If any problem found it will flash error code via the LED. Didi you notice any error code? Besides, From my poor memory there used to be quite a lot of problems with those inductors as well(resistor like components in green).
What a great summer job : ) Thanks for the suggestions. I'll check those inductors and connect an LED to read the fault codes
Ah yes that is what I saw with the one I found in my IBM. Is there a cheat sheet with some details on those error codes?
@@tony359 You have the 8425S right? I found something here: 68kmla.org/bb/index.php?threads/how-can-i-troubleshoot-a-miniscribe-8425s-hard-drive.42497/
@@Epictronics1 Amazing thanks. I'll have to check that one, I think it showed 4 solid flashes - hence Error F: unexpected interrupt from processor? Uhm... Thanks for now!
I'll add this for anyone that reads the comments and comes across this in the future. These drives can also be put into a self test mode when you close j13. You can then power up the drive without being connected to a computer and maybe after a few power cycles it can come back. I've seen this referred to as exercising the drive. Adding a couple of drops of oil to the stepper can help also.
But don't put too much lubricant in, it can penetrate all the way to the platter -- a tragic fate for a drive which is actually in my experience pretty reliable! I have a couple still working, with no more bad sectors than are listed on the label...
Thanks, I'll definitely try that
Even though the repair is stalled for lack of direction at this point, I want to say thank you very much for the video, because it provides a guide of what to check and what else to check if nothing seems to be the problem. That is a big help to people like me who know how to use a multimeter, but don't know much on how to find the problem beyond checking all the capacitors.
Thanks : )
Valiant and good effort! On the tantalum caps, they are almost always used for decoupling, meaning that they are connected to ground and Vcc. It really is totally meaningless to try to measure their capacitance in circuit, but measuring the resistance from Vcc to ground (without removing any component at first) can give you a valuable indication if one of the tantalums is busted. When they go bad, they become resistors. Then they heat up and the electrolyte inside (tantalums are electrolytic) can vaporise and steam out, either catastrophically, or more slowly, by breaking the seal around the leads. In this latter case, you can only see the physical damage around the leads under magnification. Regarding the short leads of your ESR tester, any longer and they become inductors, reducing the reliability of the measurement.
Thanks, that would totally explain why I had magic smoke without any visual clues! Question then is, are those smallest radial caps tantalums or ceramics? Can I trust that very small tantalum caps have polarity markings?
@@Epictronics1 I think so, they should have polarity markings, even if in really tiny silkscreen!
@@thebyteattic Ok, I'll check, thanks
My first job in 1979 was for a Canon calculator dealer in the USA. By the time I got involved they had single chip multi-function display and printing calculators. The calculators they made in the early part of the decade used discrete components, some had multiple boards for the various functions such as addition and multiplication.
They also had programable versions. The earliest used wire wrapping, then later had paper cards, and then magnetic stripe cards.
The first computers of theirs evolved from their calculator lines, eventually looking more like the personal computers of the day. They also had a typewritter division that made dedicated word processors. It was interesting seeing the custom computer development in that area.
The first MS-DOS/CPM-86 compatible PC was the AS100. It was not IBM PC compatible. The first PC compatible that I saw was the A-200. So it was interesting to see your EX version.
The Canon CAT was developed by Jeff Raskin after he left Apple. I was no longer involved with the Canon at that point, but there is a lot of published information on this Mac cousin.
It's a shame that a lot of Canon's PC history is diluted and unavailable. They had so much to do in technology, often in partnerships with other companies. They weren't just cameras, printers and copiers. They took the personal copier and made a personal laser printer used by Apple, HP and so many others. The made components such as motors and disk drives.
Yeah, Canon's history and the line-up is quite bloody impressive. I have had so many Canon products over the years and still do. Cool to hear that there was an A-200
That machine is identical to Olivetti PC's of the time in the UK
Pieces of black debris would suggest a diode or transistor.
The tantalums, in my experience at least, wouldn't let you off with a just a few pieces of debris and smoke, they'd go with a light show and leave a nasty skid mark behind, maybe even a proper burn hole.
I'd check below the ICs, next thing. And also check the pins for proper values.
I don't think there are any data sheets for those custom ICs unfortunately. But I could definitely lift all black components of the board and make a visual inspection
Not sure if you did, but I'd do a low resistivity check across the capacitors etc... firstly as well as doing a capacitance check, since the motor spun up I wouldn't suspect that side of the circuit as most servo style motors/circuits probably wouldn't spin up with a blown MOSFET or be as quiet as you've mentioned if it was faulty. Like Dave Jones says away check your voltages, even without a schematic you can still test Vcc levels to 74xxx logic chips etc.. lastly like others have said the drive might be fine and known good system might be a better test bed if possible.
I checked for shorts across all the caps but they were not shorted. I didn't think to check the voltages I have on the logic chips. I'll check, thanks
The drive may be fine. Boot from a DOS floppy containing the DEBUG program and launch DEBUG from the command prompt. Type in G=C800:5
This should bring you into the controller cards low level format program. Let it format the drive and then you should see it in FDISK so you can partition and then high level (DOS) format it.
Thanks, I'll try that
If one tantalum cap is bad you ought to replace them all... same for the electrolitics...
Yes, they will be replaced. If we can make it work
The debris found in the board looks like the bottom bits of one of those IC's. I have had that one bite me before. Big mystery hidden behind a door!
On another note, I have one of those drives and for some odd reason it was picky with the power supply that ran it (I figured cap issues in my case but the drive still "works").
Yes, could be. The problem is that those are custom chips and don't have a data sheet with the pinout to check. We might lift them off the board to have a look underneath
@@Epictronics1 I forgot to mention before, but have you tried a jumper on 13a and left the drive run for a few minutes yet? It is a simple internal test for the drive itself.
Also, if you need readings on certain parts of the drive please let me know. I can scope a few things out for you. My drive has a dead head now but the PCB seems to function just fine. Your video made me dig it out of the stack.
@@jakethetech4958 Thanks for the offer. I'll try that :)
Enjoy the wine! What a doozy! Looking forward to the 5154 repair!
Thanks, I did, the whole bloody bottle haha. I'll start ordering parts for the 5154 :)
@@Epictronics1 my 5154 has been sorted a bit but never enough. It has a weird colour skew,. especially in text mode
@@AlsGeekLab I have only restored slightly newer color displays but I expect the EGA to be similar enough. You may need to adjust RGB cut & gain
I’ve got one that I haven’t tried powering up yet. I want to do a full disassembly, clean, and recap first. I’ve only done my 5151 so far.
Hope it actually works! Looking forward to your video as a heads-up for what I should expect to find in there.
@@nickwallette6201 Yeah, I should have known better : ) I have restored all my 5151s too. I might restore my 5153s first. I have not decided yet
The readings you got from that tantalum remind me of a Vanguard arcade machine where the speech only worked sometimes. I pulled the tantalums in the op-amp and checked them with my Testofon (a device that transforms measurement current into audio frequencies, aka the best piece of test equipment after the oscilloscope) and I essentially got random music. It was random, it was unpredictable and measured between ~500 Ohm and 5kOhm (measuring the leakage, not the ESR!) and if I had left it running until the end of the universe it would eventually have played Never Gonna Give You Up from start to finish. Note again on the Testofon, a capacitor should make a quick descending sound and then fall silent.
haha, Rick bloody Astley, That Testofon is quite an interesting devise. I think I have seen you use it in a video. The best lead so far is that one of the caps has broken the seal around the leads and blew out the magic smoke without leaving much of a trace
@@Epictronics1 There really isn't* a lot of information on the Testofon, so I thought I'd create a video showing how to use one. The company (Taco-Nauert) that makes these (they're now called Contitest) thanked me with four free Testofon kits. Two went to Repair Cafes, one to my local pinball and video arcade and one to a less local pinball arcade.
Btw. there were two tantalums in that circuit and they both did the same random music stuff. You got lucky the other tantalum you measured was fine, otherwise you would have thought that "weird jumping around measurements" are normal on these...
*)correction: wasn't. Now that I look, I can see two more videos where people actually use those. But mine is the one on the For Amusement Only e.V. channel which is my local pinball and video arcade's channel.
(btw. the guy behind that youtube channel has a second channel daddlertl3. He's created the BreadAmp (Winamp for C64) and a whole new C64 video format called Koalavideo which plays videos at decent quality with sound at almost FM radio quality - I asked him to convert Weird Al's "It's all about the Pentiums" music video so it now runs on the C64! Turns out the Commodore 64 IS really neato! 🙂)
Sorry for that Wall Of Text.
@@senilyDeluxe Haha, ok, now I have to download BreadAmp :)
@@Epictronics1 You'll probably have more fun with Koalavideo - he coded tools and wrote a manual in German and English on how to make your own Koalavideos. So maybe your next TH-cam video will stream from a C64 🙂
@@senilyDeluxe yeah, 8-bit YT sounds cool to me : )
Cool diagnostic video! If my miniscribe has surface issues and can't be fixed (I'd like to give it another go at some point), maybe you could have my PCB if they are identical?
Thanks, that would be great. But I hope you can fix it! Hmm, I just had a thought... Did those old MACs have 8425s in them? I think I have at least three compact MACs with HDDs. I'd better check
EDIT: I just realised, they all have SCSI, don't they?
Ah yes, they are SCSI... Bugger! Mine is a 8425 though. I'm not optimistic for mine. It went through hours of surface scan, bad blocks keep coming up and it's totally unreliable when I try to copy things on it... But I'd like to give it another go.
@@tony359 I'm actually watching the video right now! You interrupted me from watching your video, this is hilarious haha
@@Epictronics1 sorry sorry sorry LOL :)
@@tony359 🤣
modern way to find shorts is to inject voltage see what is getting hot via soaking the board in ipa or heat flare cam at this point this is your only option
That's would be a good approach, I didn't think to try that.
@@Epictronics1 just be careful when rising voltage after watching thecod3r hes got a few decent tips
@@jacktheripper6716 I use this method frequently when messing around with Commodore 64s. I just power it on and use my fingertips as a sensor. If I burn myself, I'll replace that chip haha. I guess it's about bloody time I get a thermal camera :)
i have a dead pcb for one of those...if you need a part or custom ic etc feel free to ask i can give it for the price of shipping.....i also have a drive that works but has very worn bearings...i have it for parts for my other working drives...
I have the drive on the bench right now. I will do some more testing this week. Thanks for the offer :)
Thanks for sharing. 😉👌🏻
Hook up an LED directly to the drive and see if it blinks an error code. If not, put the drive in another computer with another controller, at least see if the drive can seek, using software like Ontrack's Disk Manager, which comes with a DIAG program allowing you to seek the drive or format a high cylinder to see if it is responding at all. Speedstor might also come in handy. If you are willing to have data loss, try to low level format the drive with DM or Speedstor.
Thanks, I'll try that
When in doubt, always follow your nose.
You know, I forgot to sniff that damn thing :)
Common for those early 3.5" drives, it could be the interface chip next to the MFM connector, and if that is the case the drive hears nothing from the PC, Try replacing the SGCC 004460100 chip and see if anything changes.
Thanks, that would explain the small black piece of debris I found on the PCB. I'll pull it off and check underneath. Unfortunately, I can't find that chip on any of the usual places (eBay, Ali, amazon, or mouser
@@Epictronics1 I bet you'll need a donor card, as those ICs looks like they were custom manufactured
@@cocusar yeah, if those custom chips are bad, we're in trouble
@@Epictronics1 at that point it may be a better idea to find same (or similar enough) drive and pull the whole board from it....
@@jwhite5008 Yes, if I can find one for a reasonable price, I'll grab it
The Mobo Bios hard drive setting was not mentioned, and I am curious if the mobo has a hard drive detect bios or not, if not the hard drive could have been placed in a computer that does have a hard drive detect.
Unfortunately, I can't find any info about this Canon or its BIOS online. So far, we have just assumed that it's the original HDD. We have also assumed that it has an HDD auto-detect. Simply because we have no choice. Next, we are going to mess around with debug
Have you tried mashing the delete or one of the function keys as you switch the computer on to get into the bios to see what type of hard drive settings have been set?
Anyway, most modern PC's have a section in the bios to auto detect and set the settings or allow you to select what the hard drive setting should be for it to be detected and therefore operate properly, So once you know what settings of the hard drive are by getting a more modern computer to detect it, I would make sure nobody inadvertently changed those bios settings in the Miniscribe.
Just be aware some old hard drives can be detected incorrectly by newer PC's ergo sending you to pull out your hair to get the damn thing bootable... 😇
@@KorAllRBare Thanks, I have tried with all F-keys, ESC, and DEL. No luck. Next, we'll try to run debug from DOS. Maybe, we can check/set the parameters that way
@@Epictronics1 long shot as it's rare but did you try all the arrow or direction keys like it's an AT PC..
Does the Hard drive have its settings on it's label?
@@KorAllRBare This is the first I hear about arrow keys to enter setup. I thought AT had to have a diagnostics diskette? The settings are available online
I would try the controller on another machine without any other hdd. Just to make sure that the board is working, check with MSD if there is the option ROM from the board loaded. With debug can access LLF routine from its ROM (G=C800:5). If all ok, connect the HDD and try fdisk to see if is there. Also can use Disk Manager from Ontrack (WinWorldPC) for that. Try to see if any data is readable and if not, LLF it with any tools (debug or Ontrack DM).
If all is ok, you WILL need the setup/config diskette for that computer. Alternatively, you may try XTIDE ROM/board to use an IDE HDD.
Thanks, I'll try to get around the lack of a setup diskette with a generic BIOS
@@Epictronics1 I can´t wait to see the follow up!
@@sebastian19745 I have gotten lots of help in the comment section. There will definitely be a follow-up on this project
Most 16 bit MFM controllers don't have a format routine in ROM and were actually formatted with a software which was supplied with the hard disk, or with an AT BIOS which had such functionality integrated (AMI-BIOS comes to mind).
@@diskettenfett3161 Ok, thanks. So, do I need to low-level format the drive in a system with AMI BIOS? I'm pretty sure I have a few working 486 boards with AMI BIOS
could it be the firmware chip ?
I hope not. It's impossible to find replacements for those custom chips
are you sure it's the drive that's bad? Did you try a different controller on another computer or even a different drive?
I don't have any tested MFM drives or controllers easily accessible right now. I'll see what I can do about that
@@Epictronics1 fair enough.
Why not test this drive in some more modern PC? I have a 386 with WD1003 MFM controller and ST-251 MFM drives.
Sure, I have a few PCs with MFM controllers but they are all in storage. I'll bring one of them out and test the controller/drive separately.
When you checked the hard drive controller, and you said "...I see something suspicious...", I thought you were referring to the "Made in Ireland" sticker.
lol
MiniScribe apparently made some nice SCSI drives in their later years.
If it's nice and quiet as this one, I'd love to have one : )
@@Epictronics1 Unfotunately not. And one thing was common among the Miniscribe drives I've had in the past- they weren't very quiet.
i dont see you try different hdd control cards if you did them reply to ic chip with sticker on it wide one it sent data to hdd control cards
if control cards have issues it will cut off reading to hdd and replace hdd control card to see if still working with hdd if no them that ic chip with sticker on it just save yourself a headache from stressing out looking for issues
judging of smoke coming out of psu feeding power to hdd control card to hdd could been burn out one way fix it just put in different hdd control card and ic chip wide one to see any different if yes there other way save it
I don't really have a tested MFM controller or HDD available right not. I'll see what I can do about that
I was expecting fdisk to show a drive, even if disfunctional. To me this implies the controller card might be faulty. But I assume you have good reasons to trust it?
If so, to me, the next step would be to check why the controller card doesn't even detect the presence of a disk drive, so it can report it back to the PC in turn. Doing that by starting at the cable interface between the two and trying to understand those signals.
However I have no idea how the MFM standard works, and this could be quite the steep learning curve and a lot of effort to hunt down the fault this way. This is still a starting point however.
Yes, It was pointed out to me that the controller should see the drive, whether it's broken or not. I'll bring out a tested controller for the next repair attempt. I didn't have one easily accessible while troubleshooting
I have a similar Miniscribe drive housed in Dataframe SCSI enclosure. It has a dedicated board that converts MFM to SCSI. Only problem is that the drive suffers from stiction. I have to remove the HDD cover and physically rotate the platters to get them going. After which they'll work fine. It'll even spin back up if turned off for a little bit. However, if it's unplugged for any significant amount of time, it won't spin up any more, and I have to give it a jump start.
I don't know what causes stiction. I saw a TH-cam video some time back where the person investigated it. Since his drive was broken, he went ahead and disassembled it completely by removing the platters and everything to try and figure out what causes stiction. After it was completely disassembled, the platters spun fine. The bearings and grease seemed fine, and the platters spun freely. So he thinks it could be something metal become magnetized, or perhaps something has degraded allowing the neodymium magnets to interfere with the drive spinning up (do HDDs have a "parking brake" or some sort?)
I know that quite a few HDDs suffer from stiction. It'd be great to definitively figure out why, and then have a fix.
Let's hope someone figures it out!
@@Epictronics1 If i had to guess, flat spots on the bearings...
@@unmanaged Oh, I hope you're wrong. Replacing a bearing in a cast aluminum case that small would require a high-precision press. I'm not even gonna bother checking what one of those would cost :o
@@Epictronics1 in a past life I repaired copiers and the like and this was something we had seen for devices ... YMMV...
@@Epictronics1 also get some lock graphite powder and get in there if you can ...
You can usually DEBUG into a setup utility on the MFM controller card.
I'll try that. Thanks
@@Epictronics1 Andrew beat me to it, IIRC if that controller is expecting a drive, even with one unplugged DOS will see one present but error out when trying to access it. FDisk instead didn't see one at all so I'm betting getting into the controller's setup via debug will let you restore the conf.
@@771racing Thanks guys, this gives me something to work with
@@771racing how do you do it though?
You have to know the specifics for a given controller, but the basic idea is you use debug in DOS to manually jump to the setup util or low level format routine in the controller's BIOS. So, debug followed by -G=C800:5 for example. I found the WD1003WAH manual online, looks like it has a low level format routine, but not a setup routine in it's ROM, it relies on the host PC to determine drive parameters. So this means a setup disk for that system is likely going to be needed unfortunately.
I hope the wine was good!
haha, yes it is, an excellent Sangiovese from Piemonte, cheers!
crystal maybe bad?
Could be. I'll check with the scope in the next attempt, along with the voltages on the 74 logic chips. Unfortunately, we don't have the pinout on the custom chips
First of all, there may be a problem in way too many places starting from an accidentally moved dip-switch.
You've ruled out obvious problems. There is no reason to continue randomly checking every component.
You need to get a working AT-computer with working MFM controller and drive. Those are not too hard to get.
First try finding which component is bad. I'm going to guess it's the drive. If that's the case:
it may be a VERY long job to diagnose a hard drive. extremely long without a repair manual.
not sure if it is worth the time. there may be unique data on it still though.
you can try getting an exact copy of this drive, but working, and replacing the board.
First check that every motor works (from the looks of it - they do)
And also the diode is most likely damaged. Trace what it's connected to, that is most likely damaged as well.
Try disconnecting it from a PC first and scoping the pins.
You will need to assert DRIVE SELECT 0, or whatever it is configured as (see drive manual/controller jumpers)
is READY asserted?
is INDEX getting a stable signal? Is its frequency consistent with RPM?
try to STEP (set DIRECTION). does that work?
etc..
Good to hear that there are knowledgeable people out there, this drive might have a chance with some help. It was mentioned that the diode's black legs were normal but I'd be happy to replace it in the next attempt. (and checking what it was connected to) I have a few systems with working MFM and plan on bringing one out to test the drive/controller individually. Yes, getting another miniscribe 8425 would be immensely valuable for this repair, especially since it seems to have custom chips
@@Epictronics1/videos Actually, I'm not *that* knowledgeable myself, just what I heard/seen/understand.
But I remember seeing diodes like that, all of which got overheated and destroyed. It's not a far stretch to assume it may measure fine on the multimeter's microamps but not work properly under load.
Maybe in your case it's not dead, but suspect nonetheless, so still It would be worth to replace it at least to make sure it doesn't fail catastrophically in the future damaging something else.
And if that diode overheated, it's reasonable to assume whatever it is connected to caused it to become like that or was damaged as well. I'm guessing it might be related to stuck spindle but who knows?
@Mr Guru I guess it's a thing :) Trying to guess something that actually makes sense lol. To be honest, I don't think for long, but I like to have a reason to try something in a troubleshooting
@@jwhite5008 Yes, the most likely scenario is that the motor got stuck and caused over-current in a component. Let's hope it's a generic component
@@Epictronics1 havent watched the video yet, most common failure components on drives are diodes , small ones (smd) and inductors or resistor arrays. if the drive spins and seeks then most likely it is fine. if it has a problem it will throw an error by blinking the led repeatedly on a pattern...if no problem the led should blink then stop after the drive is ready..... if you test the drive on an AT system make sure to disable all ide controllers or you will get conflicts and the drive wont work. if you use a super IO card enable just the floppy controller and disable the IDE by means of its jumpers...if your board has integrated IDE then disable them in the BIOS...make sure to set all the drives in BIOS to "not installed". the mfm controller bios will takeover...but for maximum compatiblity i would definetely test this drive on a XT or clone...i use my amstrad pc1512 to test the mfm drives.... sometimes smd components look like resistors and are diodes, sometimes they look like transistors and are diodes or inductors...etc...read the silkscreen and figure out before testing......if the drive spins, seeks and the leds blinks then goes off and stays off then its time for low level format , then fdisk and format again....then after its working use spinrite on it on level 5 to exercise the drive a bit and "refresh" the magnetic surface
I vaguely remember there were issues swapping drives between different brands/models of controller cards. Possibly something to do with the interleave that the controllers support, requiring a low-level format to make it work. But I don't know if these are the symptoms you would see if you used the wrong controller with your drive.
I have considered low-level formatting but held off. Perhaps I'm optimistic but I want to see if that drive has any cool software before I do a clean install. Apparently, that controller should see the drive if it's configured correctly, whether it's broken or not
I kind of don't think this computer is "rare".... (maybe these days) but possibly it's just terminally "unremarkable".
Well, I'll leave it up to you to judge whether it's remarkable or not. I think it's a good system. Solid, well built with good specs. I think It's quite rare with a seemingly unknown system made by one of the best-known tech brands in the world.
why did you not filter the high pitch noise, if any
you have at much time as you need, you dont have to pay any rent or anything anymore, be self-supported, best be God supported, I dont care which, as long as you are self-supported, and not supported by any other people, even through money, time for you to go self-supported and independent, and money economy trading for "support" is not it, freedom and then give freely, says the Lord to you, and Lord is Spirit
there is no option to be people supported, you have all you need, for free, not to pest-er anyone, have sabbath as ordered
content is fine, but it must be free content, not made in any way "to have to", anything, even based on illegal country laws demand obligations (against-God given freedom), and anyone who claims to own anything, keep what you have
Take care
Not to worry, I'm having a laugh about it : )