The 1050 Repair Saga continues: Drive #4 on the bench now...

[SlagOMatic]

I'm continuing my learning process of repair/refurb/upgrade on my 1050s -- three complete, one on the bench (and the subject of this post), one waiting on the shelf. I'm not an electrical engineer. I've been learning as I go, mainly through YouTube videos and the kind folks here.  🙂

 

Initial symptom was it would power up (got the power LED) but it wasn't seen by the computer and the drive wouldn't initialize at power-up or spin if the lever was turned down (the busy LED would not turn on). As I did with the other three I disassembled it, pulled the five chips out of their sockets, give the board a bath in the ultrasonic cleaner, rinsed with distilled water, rinsed with 99% IPA, dried, removed all the caps and regulators, tested new caps and regulators, installed new caps and regulators, reinstalled the five chips, double-checked all my work, connected everything, and tested -- and got nowhere. The drive still wasn't seen by the computer and the drive wouldn't initialize at power-up or spin if the lever was turned down. Since I now have known-good drives to test against I verified that the mechanism wasn't at fault; connecting Drive 4's mechanism to my known-good board let the drive spin, so that's definitely an issue on the board. I went over the whole board with a magnifier to see if anything obvious stood out but found nothing. Many YouTube videos pointed me towards several diodes and resistors that are known to have issues; I tested those and they were good.

 

Since I was approaching the end of my engineering diagnostic skills I decided to test the five chips. After a bit of swapping out with known-good chips I discovered the U10 ROM chip was bad. With a known-good U10 the computer would see the drive (or at least, the board without the drive attached) and give BOOT ERRORs. Since the U10 has to be removed for the Happy 1050 upgrade anyway I just pulled it and installed the Happy upgrade. With the Happy upgrade the board without the drive attached was seen by the computer and gave BOOT ERRORs.

 

Reconnecting the drive mechanism to the Drive 4 board showed that it does spin, so yay. When I close the lever all the way the drive engages and spins at (to my eye) the same speed as my known-good drive. But when I put in a disk and close the lever, the disk does not spin. If I manually hold the lever in the partially-down position the disk does spin but not enough to get it to read or do anything else useful; it's clearly struggling. Running Happy diagnostics on the drive shows the upgrade is working properly. Trying to run the RPM test or the write test fails; the RPM test says it can't read the drive speed, the write test fails immediately after starting. I did a thorough cleaning of the drive, lubricated the mechanical parts, and installed a new drive belt; the issue persists.

 

I swapped the mechanism with one of my known-good drives. The known-good drive connected to Drive 4's board spins up without a problem, but the RPM test in Happy diagnostics still fails (can't read drive speed).

 

So I'm thinking I may have two problems here. One, the mechanism is either worn out (unrepairable?) or there's some kind of power issue within the mechanism itself. And two, whatever circuitry on the board that measures drive RPM is having issues. It's this second one that bothers me as it seems more logical to me that RPM circuitry would be on the mechanism, not on the board, but the fact that I get "can't read drive speed" using a known-good mechanism suggests it's on the board. I also think it's particularly odd that the U10 chip went bad. I've been reading a lot of internet and watching a lot of YouTube to help me learn about the repair process here and I've yet to come across a failed 1050 that had a bad U10. Closest I got was a video where the 1050 in question inexplicably had a ROM from an Atari 5200 installed where the 6810 memory chip was supposed to go, but nothing about a failed U10. So I'm also wondering if whatever killed the U10 could also have killed something else that's causing my current issues, maybe?

 

Right now I'm at an impasse. As always, any hand-holding would be appreciated. Thanks.

Edited February 25, 2023 by SlagOMatic

[BillC]

1 hour ago, SlagOMatic said:

If I manually hold the lever in the partially-down position the disk does spin but not enough to get it to read or do anything else useful; it's clearly struggling.

1 hour ago, SlagOMatic said:

One, the mechanism is either worn out (unrepairable?) or there's some kind of power issue within the mechanism itself.

Sounds to me like the drive belt is stretched/slipping and needs replacement. Console5 has 2 replacement belts of different quality/price.

 

https://console5.com/store/catalogsearch/result/?q=atari+1050+drive+belt

[Kenshi]

The manual mentioned in the previous thread has some good suggestions to check for drive motor problems, especially "drive motor doesn't rotate when latch is shut" and "drive motor operates sporadically" on 4-1. Since you completely replaced the drive mechanism and still got an RPM fault, it seems like that issue is on the board. If you have a known good LM2917 (U5) chip, you could try swapping that out. You can check pin 11 to make sure the tachometer signal is making it to the chip. An oscilloscope is ideal, but if you don't have one, you can switch your multimeter to VAC mode and it'll tell you if you're getting a changing voltage and some will give you the frequency. It'll be less than what's in the manual since it shows peak voltage and your multimeter will try to show RMS, but if you're only getting a few millivolts, it's background noise. If Q4 or Q6 is failing or R80 has too much resistance, it may make the motor weaker.

[sup8pdct]

RPM is generally checked by reading sector 1 continuously and timing how long it takes.

It means that something is wrong with the read circuit. There is quite a lot to it.

Maybe the FDC is the problem.

If someone has had a play with the adjustable resistors and adjustable Cap, then reading a disk will be very problematic. Usually a scope is required to set those.

The adjustable Cap is used for the Data separator circuit. If it is out, no reading of the disk will happen.

Reading of the disk is required before any write can happen.

 

James

 

[SlagOMatic]

14 hours ago, BillC said:

Sounds to me like the drive belt is stretched/slipping and needs replacement. Console5 has 2 replacement belts of different quality/price.

 

https://console5.com/store/catalogsearch/result/?q=atari+1050+drive+belt

Per my original post, the belt has already been replaced with a new one.

[SlagOMatic]

I was playing with this some more today (even though I promised myself I would take a day to relax, oops) and the more I play with it the more I think it's the actual drive mechanism at fault. Remember: Drive 4 is the problematic drive.

 

With Drive 4 assembled (Drive 4 board + Drive 4 mechanism) and not connected to anything other than power, I was just inserting and removing a (blank) floppy disk to see what would happen. I noticed that sometimes it spun without complaint, but other times when it spun it sounded like the motor was struggling and I noticed the media (the actual literal floppy part of the disk) was wobbling, and sometimes the media wouldn't spin within the disk at all and looked like it was actually binding inside the floppy jacket. Curious, I transplanted Drive 4 mechanism into a known-good chassis and repeated the experiment and got the same results.

 

You can watch the 1050 basically destroy a floppy disk here:

 

You'll see in the video a couple times when the media has trouble spinning, wobbles, or stops spinning and warps in the floppy window. This isn't me doing something, this is the media getting hung up inside the floppy jacket.

 

That got me thinking. If there was a component that was excessively worn or otherwise misaligned in the mechanism, that could account for the disk wobbling and binding as it is. And if RPM readings are determinate upon reading sector 1, and the media is wobbling enough to prevent reliable reading of sector 1, that could account for the RPM test failure.

 

So, two questions:

 

1. Does this sound like a reasonable hypothesis?

 

2. How in the name of Mount Fuji would I go about addressing and correcting this?

Edited February 26, 2023 by SlagOMatic

[macsonny]

1 hour ago, SlagOMatic said:

I was playing with this some more today (even though I promised myself I would take a day to relax, oops) and the more I play with it the more I think it's the actual drive mechanism at fault. Remember: Drive 4 is the problematic drive.

 

With Drive 4 assembled (Drive 4 board + Drive 4 mechanism) and not connected to anything other than power, I was just inserting and removing a (blank) floppy disk to see what would happen. I noticed that sometimes it spun without complaint, but other times when it spun it sounded like the motor was struggling and I noticed the media (the actual literal floppy part of the disk) was wobbling, and sometimes the media wouldn't spin within the disk at all and looked like it was actually binding inside the floppy jacket. Curious, I transplanted Drive 4 mechanism into a known-good chassis and repeated the experiment and got the same results.

 

You can watch the 1050 basically destroy a floppy disk here:

 

You'll see in the video a couple times when the media has trouble spinning, wobbles, or stops spinning and warps in the floppy window. This isn't me doing something, this is the media getting hung up inside the floppy jacket.

 

That got me thinking. If there was a component that was excessively worn or otherwise misaligned in the mechanism, that could account for the disk wobbling and binding as it is. And if RPM readings are determinate upon reading sector 1, and the media is wobbling enough to prevent reliable reading of sector 1, that could account for the RPM test failure.

 

So, two questions:

 

1. Does this sound like a reasonable hypothesis?

 

2. How in the name of Mount Fuji would I go about addressing and correcting this?

Isn't repairing 1050 drives relaxing?

 

On a more serious note, I have this exact same issues with two of my 105 drives. One needed the belt replacing and the other I needed to replaced the set pad on top of the read/write head. I found that these get old and hard/worn and prevent the disk from spinning. Have you replaced yours?

 

FYI - you can make them yourself or buy them. Either way they work fine.

[_The Doctor__]

My check list.

In order of occurrence.

 

1) Eject slide/catch/spring mechanism bent needs slight adjustment? Spring tension? Did the eject cam bend the ascender tab?

2) Head pressure/ pressure pad  adjust pressure/ clean or replace pad.

3) Spindle bearings need clean and lube.

4) Disk hub gripping dome needs cleaning / sticky stuff/ protect tabs, labels, glue, oils etc.

5) Dirty or worn belts/cog/flywheel.

6) Disk jacket stabilizers foam block degraded or fallen off the tandon mech.

 

Looks like #1 to me mostly and you can mix and match the other contributing factors

 

for 1050

for 810

 

Edited February 26, 2023 by _The Doctor__

[SlagOMatic]

On 2/26/2023 at 2:56 PM, macsonny said:

Isn't repairing 1050 drives relaxing?

 

On a more serious note, I have this exact same issues with two of my 105 drives. One needed the belt replacing and the other I needed to replaced the set pad on top of the read/write head. I found that these get old and hard/worn and prevent the disk from spinning. Have you replaced yours?

I once had to have a broken tooth removed. It got infected and hurt like hell, waking me up in the middle of the night to soak it in warm salt water, not allowing me to eat anything solid, making me feel like a hot shit sandwich. This went on for about two weeks before I broke down and went to the dentist. I had to wait another two weeks on antibiotics to clear up the infection before they could remove it.

 

Those were good times, in comparison.

 

Anyway, I already replaced the belt on this drive. I didn't think of the pad but I just ordered some replacement pads from Amazon and will see what happens there.

[SlagOMatic]

On 2/26/2023 at 3:42 PM, _The Doctor__ said:

My check list.

In order of occurrence.

 

1) Eject slide/catch/spring mechanism bent needs slight adjustment? Spring tension? Did the eject cam bend the ascender tab?

2) Head pressure/ pressure pad  adjust pressure/ clean or replace pad.

3) Spindle bearings need clean and lube.

4) Disk hub gripping dome needs cleaning / sticky stuff/ protect tabs, labels, glue, oils etc.

5) Dirty or worn belts/cog/flywheel.

6) Disk jacket stabilizers foam block degraded or fallen off the tandon mech.

 

Looks like #1 to me mostly and you can mix and match the other contributing factors

 

Since the weather is crap right now this is going to be my Saturday. Please check my understanding.

 

1. The mechanism seems to be working properly, insofar as there are no problems with inserting, holding, or releasing disks. But if I'm understanding the process correctly, if the tab that touches the disk jacket is bent forward (towards the front of the drive) then perhaps that could be buckling the disk a bit and therefore causing the issues I'm having. In which case, bending that tab out a fraction of an inch could solve the issue. True?

2. I have new pads on order.

3. Spindle bearings have already been cleaned and lubed.

4. Disk hub has been cleaned.

5. Drive belt has been replaced with fabric-reinforced unit from Console5.

6. This is new to me, and my Google skills are failing. Can you elaborate?

[SlagOMatic]

On 2/26/2023 at 3:42 PM, _The Doctor__ said:

My check list.

In order of occurrence.

 

1) Eject slide/catch/spring mechanism bent needs slight adjustment? Spring tension? Did the eject cam bend the ascender tab?

2) Head pressure/ pressure pad  adjust pressure/ clean or replace pad.

3) Spindle bearings need clean and lube.

4) Disk hub gripping dome needs cleaning / sticky stuff/ protect tabs, labels, glue, oils etc.

5) Dirty or worn belts/cog/flywheel.

6) Disk jacket stabilizers foam block degraded or fallen off the tandon mech.

 

Looks like #1 to me mostly and you can mix and match the other contributing factors

 

Following up on this.

 

It seems like the culprit is #1. See attached pics. The first pic shows the disk inserted normally and doing it this way gives me mixed results as seen in my earlier video. The second pic shows me pulling the eject slide (correct term?) further back by about 2mm using a pair of tweezers. If I insert a disk, then pull the slide back this way, then close the drive lever, the disk spins without incident 100% of the time. (I'd show you in video but it's a two handed operation so I don't have a third hand to hold the camera.) I don't get any wobbling or binding this way.

 

I have NOT replaced the felt pad on the head yet.

 

It seems that the only way for me to correct this is to disassemble the drive, bend the tab out a tiny bit, put it back together again (assuming that I can), and give it a try. I'm hesitant to just bend it in place as I don't know how strong the anchor points are that would be stressed during the bending process, and I'm worried that the disassembly/reassembly process might throw some critical alignment out of spec that I won't be able to recover from. Advice?

Edited March 4, 2023 by SlagOMatic

[SlagOMatic]

So, update...

 

Upon disassembling the drive I found a mystery shim that turns out was part of the cone lever assembly. Since I never disassembled that specific component I'm blaming a previous owner of the drive. I put the shim back where it belongs.

 

While I had the drive disassembled I very carefully bent that tab out a tiny bit. After reassembling everything I had "mostly" success, which is to say the disks were no longer binding when inserted. But I may have bent the tab out too much as I now have to push the disk much further in before it clicks into place. It works and passes all diagnostics, but it's annoying to have that little quirk. I also noticed that the felt on the drive head was degraded and fuzzy, probably from all the floppy carnage, so I replaced it. That brought up a new issue: the new felt is thicker than the old felt and it would occasionally hang up on the window when ejecting a disk. I solved this with a drop of structural epoxy on the pivot point of the drive lever. The head now lifts up another mm or so higher than it did before, allowing the felt to raise higher and not catch on the window.

 

I also accidentally damaged the wires to the "busy" LED so I'm gonna have to address that too.

 

So right now everything works, albeit with some annoying quirks.