Yet another 1050 with boot errors

[Retrobecanes]

I believe that I read all the posts in this forum that deal with 1050 boot errors yet I have not found one that is identical to the problem that I am experiencing.  When I turn on my drive, the drive spins and the head moves and parks itself correctly like it does in my "good drive".  When I turn on the computer, the drive spins but the head never moves and I get the the "slow" boot error messages on the screen, and the disk keeps spinning...

 

Note that I have also tried this drive assembly with the board from my "good" drive and it works well.  Therefore it is fair to conclude that the problem lies with the board and not with the drive assembly.  On this board I have replaced the two 3086 and swapped all the socketed chips with those of my "good" board.  Nothing made a difference, still the same boot boot errors.  I am really not sure where to go from here.

[_The Doctor__]

it leaves you with basic support components and RPM resistor and FDC capacitor adjustments... these may require a strobe or oscilloscope however.

[Retrobecanes]

I have able to adjust the RPM with the diagnostic disquette when I connected the drive assembly to my 'good' board.  How do I check the FDC capacitors?

[_The Doctor__]

You can't adjust the RPM for the mech that way, the RPM has to be adjusted for each PCB and mech combination. Same at FDC adjustment cap for each FDC pcb combo.

[Rybags]

In the modern day, you might be able to use a strobe app on mobile phone that operates the flash.

I've done that in addition to the high speed camera which can do 240 FPS which allows measuring such things.

[TZJB]

On 5/21/2023 at 3:10 AM, Retrobecanes said:

I believe that I read all the posts in this forum that deal with 1050 boot errors yet I have not found one that is identical to the problem that I am experiencing.  When I turn on my drive, the drive spins and the head moves and parks itself correctly like it does in my "good drive".  When I turn on the computer, the drive spins but the head never moves and I get the the "slow" boot error messages on the screen, and the disk keeps spinning...

 

Note that I have also tried this drive assembly with the board from my "good" drive and it works well.  Therefore it is fair to conclude that the problem lies with the board and not with the drive assembly.  On this board I have replaced the two 3086 and swapped all the socketed chips with those of my "good" board.  Nothing made a difference, still the same boot boot errors.  I am really not sure where to go from here.

 

The drive is passing it's self-test. I don't know why, but I just thought of drive ID. Are the faulty board drive select switches set to Drive 1 position? If not it won't boot.

 

Just so that you know for the future, it is not advisable to swap the WD2793 FDC as it is calibrated to the board it is fitted in.

 

Other than that I think there may be a communications failure. SIO goes through U1 CA3086 to U7 the 6532 PIA.

 

 

 

[Retrobecanes]

@TZJB The Drive ID switch works and the computer recognizes the changes (that is actually one of the few parts of the 1050 diagnostic disquette that is passes).  I have also replaced the CA3086 with a know-to-be-good part and I had swapped the 6532 from a good drive.

 

@_The Doctor__ I have now adjusted the RPM with a strobe light and I believe that the motor speed is good.  

 

At this point here are the good news and the bad news:

- at startup, the drive seems to pass the self test, the head moves and parks itself as expected

- the drive assembly is good as it was tested with another board and works fine, I therefore assume that the issue in on the board with this "bad" drive

- the drive spins at the correct RPM and the heads moves and seeks the disk tracks

- I swapped all the sockets chips from my "good" drive into the bad one but that didn't make any change.  I put the originals back in.  I also replaced the two CA3086 on the board, no change.  Voltages are good at the TPs.

- if I attempt to format the drive, the drive goes through the motions, steps through all the tracks but it eventually results in an error 144 

- if I attempt to do a DIR on a good disk, the drive spins and the head movies, but it eventually results in an error 163

- if I attempt to copy files onto a disk, the drive goes through the motions but it results in an error 169

 

Really not quite sure where to go from here....

[TZJB]

8 hours ago, Retrobecanes said:

@TZJB The Drive ID switch works and the computer recognizes the changes (that is actually one of the few parts of the 1050 diagnostic disquette that is passes).  I have also replaced the CA3086 with a know-to-be-good part and I had swapped the 6532 from a good drive.

 

@_The Doctor__ I have now adjusted the RPM with a strobe light and I believe that the motor speed is good.  

 

At this point here are the good news and the bad news:

- at startup, the drive seems to pass the self test, the head moves and parks itself as expected

- the drive assembly is good as it was tested with another board and works fine, I therefore assume that the issue in on the board with this "bad" drive

- the drive spins at the correct RPM and the heads moves and seeks the disk tracks

- I swapped all the sockets chips from my "good" drive into the bad one but that didn't make any change.  I put the originals back in.  I also replaced the two CA3086 on the board, no change.  Voltages are good at the TPs.

- if I attempt to format the drive, the drive goes through the motions, steps through all the tracks but it eventually results in an error 144 

- if I attempt to do a DIR on a good disk, the drive spins and the head movies, but it eventually results in an error 163

- if I attempt to copy files onto a disk, the drive goes through the motions but it results in an error 169

 

Really not quite sure where to go from here....

 

That sounds like progress.

 

You hadn't said that it could pass any test or attempt a format previously. Is the drive un-modified?

 

Error 144 Device Done error. Could be a write protect error, but you said the mech works on the other 1050 PCB so it is a write error.

 

Error 163 Unrecoverable system I/O error. DOS or disk may be bad. A read error.

 

Error 169 Directory Full. Probably another read error.

 

So the actual fault is with reading/writing which could be to do with RPM not being near 288 RPM or the WD2793 being out of calibration.

 

From FSM:-

 

 

From SAMS:-

A bad connection at J6 would also cause your read/write issues, check for bad solder joints, although the end of the last paragraph points to U13 WD2793 which you say you changed, and changed back, but may be out of calibration. You would need an oscilloscope to be able to proceed.

 

Here are the details from TFHH:-

 

 

[Retrobecanes]

@TZJB, the drive is indeed unmodified, it seems completely stock to me.  

A few other points and comments based on your post:

- I adjusted the RPM with a strobe light.  It is interesting to note that the 1050 diagnostic disk was not able to read the RPM of the drive.  It would just beep and display 0.

- With the 1050 diagnostic disk, it passed "MPU and controller tests", "executing invalid command test", and "write protect detection test".  It failed "motor start test", "motor speed test", and "head stop/settle test".  I don't know how reliable those tests are under these circumstances as the motors seem to be working well.

- Last night I replaced all the caps for good measure, just in case, no change.

- I will double check the solder joints of J6

- I have an oscilloscope (although I am not an expert with them) and will follow the instructions that you provided.  Fingers crossed!

 

 

Edited May 23, 2023 by Retrobecanes

[xrbrevin]

is it worth trying a replacement eprom? you may have to change the jumpers if your drive has a PROM as standard tho

i suggest this because a USD modded 1050 also fails the RPM test of the 1050 diagnostic disk

[TZJB]

14 hours ago, Retrobecanes said:

@TZJB, the drive is indeed unmodified, it seems completely stock to me.  

A few other points and comments based on your post:

- I adjusted the RPM with a strobe light.  It is interesting to note that the 1050 diagnostic disk was not able to read the RPM of the drive.  It would just beep and display 0.

- With the 1050 diagnostic disk, it passed "MPU and controller tests", "executing invalid command test", and "write protect detection test".  It failed "motor start test", "motor speed test", and "head stop/settle test".  I don't know how reliable those tests are under these circumstances as the motors seem to be working well.

- Last night I replaced all the caps for good measure, just in case, no change.

- I will double check the solder joints of J6

- I have an oscilloscope (although I am not an expert with them) and will follow the instructions that you provided.  Fingers crossed!

 

 

 

The drive cannot read the RPM because it is not reading the disk.

 

You already established that the MPU and controller were fine as you had physically swapped them out. These tests don't involve reading the disk.

 

Failing the motor start test is anomalous though but could still be to do with reading the disk. And although the head stop is performed by a different motor and detector, again it may rely on reading the disk track 0.

 

In conclusion, the frequency modulated (FM) read data is not reaching or being interpreted by the controller WD2793. Hopefully a calibration will fix it.

[Retrobecanes]

9 hours ago, xrbrevin said:

is it worth trying a replacement eprom? you may have to change the jumpers if your drive has a PROM as standard tho

i suggest this because a USD modded 1050 also fails the RPM test of the 1050 diagnostic disk

The drive originally came with the "K" ROM.  I found a .bin file for this version online and programmed a replacement EPROM, just in case.  That did not solve my issue.

I also took the ROM from my 'good drive' and modified the jumper settings to match the good drive but that didn't help.  I then reverted to the original ROM and original jumper settings.

[Retrobecanes]

17 minutes ago, TZJB said:

 

The drive cannot read the RPM because it is not reading the disk.

 

You already established that the MPU and controller were fine as you had physically swapped them out. These tests don't involve reading the disk.

 

Failing the motor start test is anomalous though but could still be to do with reading the disk. And although the head stop is performed by a different motor and detector, again it may rely on reading the disk track 0.

 

In conclusion, the frequency modulated (FM) read data is not reaching or being interpreted by the controller WD2793. Hopefully a calibration will fix it.

So last night, I attempted to follow the 2793 adjustment instructions.  I closed TP7 and TP8 with a jumper, but the first test, which is to read pin 16 of the 2793, did not produce the intended result.  I did not get a square waveform at all but some form of 'wiggly' signal (sorry I am that technical) on the screen.  So I took out the board from my "good drive", put it in test mode as well, and obtained the exact same signal with the oscilloscope on pin 16.  Pins 29 and 31 did however match the instructions a lot better.  Not quite sure what all of this means.

[Retrobecanes]

@TZJB In addition to the test above, I also wanted to let you know what I carefully checked J6, all its solder joints and all the traces leading to it and could not find any issues.

[blacka013]

Retrobecanes,

 

Can you post photos of the board, and the oscilloscope traces, so we can see what you are getting.

 

Regards

[TZJB]

3 hours ago, Retrobecanes said:

So last night, I attempted to follow the 2793 adjustment instructions.  I closed TP7 and TP8 with a jumper, but the first test, which is to read pin 16 of the 2793, did not produce the intended result.  I did not get a square waveform at all but some form of 'wiggly' signal (sorry I am that technical) on the screen.  So I took out the board from my "good drive", put it in test mode as well, and obtained the exact same signal with the oscilloscope on pin 16.  Pins 29 and 31 did however match the instructions a lot better.  Not quite sure what all of this means.

 

This is puzzling, the waveform needs to be square. Did you connect the ground connector on the oscilloscope probe to the board ground plane? Also pin 16 is available on TP10, Pin 29 on TP9 and Pin 31 on TP11 so easier to access.

 

 

I did expect J6 to be fine, but it was worth checking. However please recheck your readings for the WD2793.

 

After that it needs to get complicated. The Read head amplifier circuit needs to be checked as this may have an issue. In theory you could start a RPM test and follow the signal through with your oscilloscope, first checking on the state of N WRT. As you can see Atari had thought of this and added TP1 to TP6 for this purpose. Luckily you have a working drive to compare it to.

 

[Retrobecanes]

I get the same result at TP10.  I was able to contact Jurgen, the gentleman that wrote those WDS adjustment instructions, and I am in communication with him now.  We will see what transpires.

So what path on this schematic would you recommend that I follow in the troubleshooting?

 

[TZJB]

1 hour ago, Retrobecanes said:

I get the same result at TP10.  I was able to contact Jurgen, the gentleman that wrote those WDS adjustment instructions, and I am in communication with him now.  We will see what transpires.

So what path on this schematic would you recommend that I follow in the troubleshooting?

 

 

You really need the 1050 Field Service Manual referenced here:-

 

 

It shows the waveforms you would expect to see at TP1 onwards. From this you should be able to determine which IC may be at fault.