1050 drive woes

[cwilbar]

I've been attempting to run the 1050 diags on a few 1050 drives I have..... and I'm experiencing some odd problems:

 

1. On a USD 1050 the diags don't work at all.  I presume that the US Doubler is not compatible with the 1050 diags ????  I have not tried a rom swap back to stock yet.

2. On another 1050 it fails the motor start test.  The drive appears to work fine in all other respects so far.  Is it just not coming up to speed quick enough to pass the test ???? 

3. On another 1040 it is failing the step/settle test.  The same mechanism on another 1050 control pcb passes the test fine, so it does appear to be something on the PCB and not the mech.

 

I did fix a track zero issue on one PCB.... thought it was Q5, but it turned out to be C52.  Once I replaced that it was able to find track zero again.  Not sure what type of capacitor it is (polypropylene ?), its one of those small yellow ones.

I can't say I've seen many issues with them previously to this.

 

Looking for any tips on the 3 above to help me focus what I should look at.

 

Thanks,

 

-- Curt

 

[TGB1718]

What software are you using to test the drives ?

 

I have a USD 1050, so could try and see what results I get, also a Happy, can test that too 

[TGB1718]

I just used the standard Atari Diag Disk on my 1050

 

It reports: Motor Start Test Fails !!

 

Moter Speed too high (but the motor is not running)

 

Track 0 fails

 

But this drive boots and seems to work normally, so I think the softwar is incompatable with USD.

 

On my Happy drive it failed to boot, but I think the head may need aligning as it will format and read

disks created on this drive, but "unhappy" with most disks from the other drive (including old ones from the 80's)

[Rybags]

Doesn't USD use a standard sized Rom?  (ed- 4K so looks like it does)  Possibly they had to omit some of the lesser used stuff (like the built in diag stuff) to get their features to fit.

Edited February 18, 2020 by Rybags

[cwilbar]

I was using the 'standard' CPS 1050 diag.  It is amazing how tests can fail here but the drive still boots disks fine.

 

I swapped all the socketed chips from a working one to the one with the step/settle test failure with no change.  So by swapping chips I believe I've ruled out the chips and the mech (mech tests fine on the pcb I used as the chip donor for testing purposes).

 

Maybe something with the 2 8 pin DIP chips that operate the stepper ????

 

I haven't done anything more with the motor start test drive, but I have to try a mech swap there to determine if it is the pcb or the mech.

 

-- Curt

 

 

[+kheller2]

1050 diags will not work with a USD upgrade, AFAIK.  If the same mech passes fine in a stock 1050 (Assuming the ROM and Mech match: tandon -> tandon mech, vs WST->WST mech) I would lean towards something on the board. 

[cwilbar]

3 hours ago, kheller2 said:

1050 diags will not work with a USD upgrade, AFAIK.  If the same mech passes fine in a stock 1050 (Assuming the ROM and Mech match: tandon -> tandon mech, vs WST->WST mech) I would lean towards something on the board. 

I haven't further tested the USD 1050 yet.  I plan to drop in a stock ROM to run the diags so that I'm checking everything in that drive but the ROM.  It wasn't used extensively (the USD 1050 is my original 1050 that I bought for DD operation, as my primary drive was my original 810).   I am trying to diag all of my drives.

 

The motor start and the step/settle failures though are perplexing me.... especially as the step/settle one is coming from the PCB (and not any of the socketed chips), yet it boots disks fine.  Very strange.

 

[+kheller2]

4 hours ago, cwilbar said:

I haven't further tested the USD 1050 yet.  I plan to drop in a stock ROM to run the diags so that I'm checking everything in that drive but the ROM.  It wasn't used extensively (the USD 1050 is my original 1050 that I bought for DD operation, as my primary drive was my original 810).   I am trying to diag all of my drives.

 

The motor start and the step/settle failures though are perplexing me.... especially as the step/settle one is coming from the PCB (and not any of the socketed chips), yet it boots disks fine.  Very strange.

 

FSM says that could be mech OR defects in the read circuit. And since you ruled out the mech, I would let others chime in on that.

[Marius]

As long as you do not go altering those  blue components with the yellow screw in them. Except the one for drivespeed all others should be untouched forever unless you have a scope or something.

 

I have learned this from @tf_hh. He also might be able to assist you in troubleshooting your drives.

 

 

[cwilbar]

5 hours ago, Marius said:

As long as you do not go altering those  blue components with the yellow screw in them. Except the one for drivespeed all others should be untouched forever unless you have a scope or something.

 

I have learned this from @tf_hh. He also might be able to assist you in troubleshooting your drives.

 

 

I do have a scope... but I've never attempted (yet) to ever check/adjust anything beyond the motor speed control.  If I understand correctly, they only need to be adjusted if you changed the FDC chip ?

 

I wonder if the head settling one is a funky capacitor somewhere.... as the one PCB I fixed that had track 0 sensor failure symptoms, was flaky at first, then couldn't detect track 0 other than sometimes at initial power on.... and that was a "473" capacitor..... do these little caps have an issue with failing ?  Do they become leaky ?  or is it value drift ?  esr change ?

 

 

[TGB1718]

Usually its the elecrolytic capacitors that have problems, I use the general rule, if one has gone it's highly

likely others will fail also, I just a complete set ( 2 actually) and replaced all the electolytics

in both my 1050's.

 

Other capacitor types don't have that problem and are less likely to fail.

[xrbrevin]

as TGB1718 stated, electrolytic caps are the ones that can leak. tantalums and ceramics dont usually have any visible symptoms if they fail (apart from physical blunt force trauma). But the sizes of these used in the 1050 dont handle particularly big loads so if they look intact then put your time into researching other possible issues

the 3x big caps in the power circuit are more commonly known to need replacing due to age and workload. altho that pesky little C70 failed by dead short in one of mine and made the heatsink cook super-quick...!

[cwilbar]

1 hour ago, xrbrevin said:

as TGB1718 stated, electrolytic caps are the ones that can leak. tantalums and ceramics dont usually have any visible symptoms if they fail (apart from physical blunt force trauma). But the sizes of these used in the 1050 dont handle particularly big loads so if they look intact then put your time into researching other possible issues

the 3x big caps in the power circuit are more commonly known to need replacing due to age and workload. altho that pesky little C70 failed by dead short in one of mine and made the heatsink cook super-quick...!

electrolytics can leak (electrically), and leak (physically).  Other caps can leak electrically.

 

Tantalums are very fond of shorting.  If you're lucky it will be on a power rail with a bit of current, and it will fail spectacularly and you'll know which failed by the physical damage.  Other times they isn't enough current or the safety features of the power supply (overcurrent protection) will make it a bit more work to find the offender.  On one 1050 a tantalum shorted.... it was in a spot where an electrolytic normally is.... so I replaced it with an electrolytic.  On a Qume 5.25" "PC" floppy drive it caused the power supplies protection to kick in.  Figuring that was the scenario I was able to test the tantalums at the input and found the offender. 

 

 

[+Nezgar]

With the US Doubler ROM in place, you can use the RPM.COM command included with various versions of SpartaDOS 3.2 to test RPM.

 

You cannot run the Atari 1050 diagnostics disk with US Doubler ROM installed. If you swap an Atari stock ROM into the drive (at U10), you can still leave the ICD 256 byte RAM stack in place (at U8) and then you can run the Atari 1050 diagnostics.

 

If you are changing between mask ROM, and 2732 EPROM, you'll also have to switch the jumpers above U10. (JP1 and JP3 for mask ROM configuration, JP2 and JP4 for EPROM configuration)

 

If your drive has a WDC 2793 controller, you can use all known Atari Stock ROM revisions (H, J, K, L, WSTR5)

If your drive has a WDC 2797 controller, you can only use the Revision "L" ROM.

 

Yes, as others have mentioned, you don't want to change any of the rotatable adjustments around the WDC controller chip other than RPM, and the WDC chip must stay with it's original PCB as they are calibrated together.... Otherwise you can read the gory details of how to re-calibrate here: https://atariage.com/forums/topic/282626-howto-adjust-analogue-part-of-fdc-279x-in-a-1050-disk-drive/

[cwilbar]

While working with the drives, I accidentally powered off the 1050 while the cps diag disk was in the drive.

 

This did harm the disk... it booted and all, but I started having more step/settle errors on other drives that hadn't had them before.  I do still have the one drive that has a step/settle problem, and it is associated with the elctronics and not the mech.

 

I also have one that fails writing.... and but boots disks fine.  These are the 2 problem 1050 PCBs I need to figure out what is up with them.  Looks like the remaining 2 drives and PCBs are fine.

 

 

 

 

[+tf_hh]

On 2/18/2020 at 5:06 AM, cwilbar said:

1. On a USD 1050 the diags don't work at all.  I presume that the US Doubler is not compatible with the 1050 diags ????  I have not tried a rom swap back to stock yet.

2. On another 1050 it fails the motor start test.  The drive appears to work fine in all other respects so far.  Is it just not coming up to speed quick enough to pass the test ???? 

3. On another 1040 it is failing the step/settle test.  The same mechanism on another 1050 control pcb passes the test fine, so it does appear to be something on the PCB and not the mech.

 

Hi, some hints...

 

1. Yes. The CPS1050 test runs only with genuine Atari firmwares. For example, when a Speedy is installed, the inserted disk will be formatted during the main tests ?

2. & 3. Does the motor start spinning anyway? Did you test with a well-known working mechanic?

 

Most issues around spin motor fails or step motor fails are based on instable voltages. First check if the output of the 7812 linear voltage regulator is stable between 11.90 and 12.10 volts, even when any of the both motors starting action. Best is checking with an oscilloscope to see if the input (approx 25 volts DC) and output (12 volts) is stable and show a flat line. When the input shows sine curves, replace the three big caps (C67, C68, C71) immediately. In rare cases one of the six major rectifier diodes is bad.

 

When input and output voltages are clean and stable, replace the four caps involved in the 12 volts motor power circuit. These are: C70, C21, C42 and C43.

 

In mose cases the spindle motor starts now flawless. Now use CPS1050 tests with a fresh single-density formatted disk (you should never use the CPS1050 disk for these tests...). Format the disk in a well-known working 1050.

 

Step motor issues could be of course a defect step motor or typically a defect at one of the both 5713 chips. Also have a look at the 47R resistors R56, sometimes it was slightly burned, visible by a brown/dark smokish surface.

 

Good luck!

 

 

[cwilbar]

Thanks for the tips.  I have a bit more testing to do.... however, one of the problems I had with motor start was issues with the diskette.  That left the head step/settle test.... of which that also appears to have been from the diskette.... although the diskette it fails the head step/settle test on works fine when testing other drives.   And I believe the issue is related to the controller PCB.... so I'm wondering if someone messed with the "don't touch" adjustments, or did some FDC swapping at some point.... as it really wants a near 'perfect' diskette much more so than other PCBs with the same mechanism.

 

And the last 1050 problem (for now) is a drive that fails write testing.  It even scrambles the track for other drives.  The write fails after the first write 'beep'.  I'm wondering if there is something in the circuitry that does an erase? before write... so a write attempt either fails that or scrambles the sector when it does the write.....

 

I'm going to try a logic comparator on the TTL logic in that circuit (based on the chips identified by the manual).  If anyone has seen this sort of symptom (fail on first write attempt resulting in a scrambled disk), let me know if you found the culprit in your drive.

 

Thank you all for the tips, etc. 

 

[1050]

Yes, erase is a function of the write cycle. U21 a 3086 transistor array is called the write amp in the Sams Photofact. On signal comes out at pin 8 to be amplified further by Q1 2N4404 while the signal is actually supplied by U17 a 7406 hex inverter using two gates for more drive.

 

But the problem isn't erasing, it does that. The problem is writing understandable data back to the disk. Those list of chips is then more than the above. For the moment changing the 3086 at U21 has fixed some write issues in the past as well as U17 the hex inverter. 3086 are known weaklings at both the write amp and the SIO input uses (U1) in the 1050. Bad silicon in the early days is my guess, new ones don't have the problem. Both these chips do double duty in erase drive and data writing.

 

Sams Photofact 1050

http://www.atarimania.com/documents/Atari_1050_Disk_Drive_Sams_Computerfacts_Technical_Service.pdf

 

I would change those two chips just for the heck of it. Then see if symptoms are different at that point.

[cwilbar]

16 hours ago, 1050 said:

Yes, erase is a function of the write cycle. U21 a 3086 transistor array is called the write amp in the Sams Photofact. On signal comes out at pin 8 to be amplified further by Q1 2N4404 while the signal is actually supplied by U17 a 7406 hex inverter using two gates for more drive.

 

But the problem isn't erasing, it does that. The problem is writing understandable data back to the disk. Those list of chips is then more than the above. For the moment changing the 3086 at U21 has fixed some write issues in the past as well as U17 the hex inverter. 3086 are known weaklings at both the write amp and the SIO input uses (U1) in the 1050. Bad silicon in the early days is my guess, new ones don't have the problem. Both these chips do double duty in erase drive and data writing.

 

Sams Photofact 1050

http://www.atarimania.com/documents/Atari_1050_Disk_Drive_Sams_Computerfacts_Technical_Service.pdf

 

I would change those two chips just for the heck of it. Then see if symptoms are different at that point.

Thanks for that link.  My spare (used for testing) 1050 PCB wasn't communicating via SIO.  I didn't have another 3086, so I found the bad transistor in the array, and I used a DIP socket, stacked it with the chip and put a transistor (3906 ?  I forget the #) "on board" to replace the blown one in that 3086.  Worked a treat ! 

 

Hopefully I'll get some time tomorrow to mess with that drive.

 

When I figure it out, I'll report here so hopefully the log of my adventure will help someoone else out in the future.