Atari 1050 UDN5713M chips question

[+Nezgar]

2 hours ago, kbj said:

 

I took the EPROM that I had burned originally and wrote a batch file to continuously read its contents and compare it to the known working ROM dum

What programmer are you using? I notice in the picture above your EPROM is an ST 2732A that requires 21V programming voltage. The TL866 for instance maxxes out at about 18V in practice I've found, so it may have programmed "weakly" hence the occasional read errors. Better to get the ones that need 12.5Vpp.

 

Also, your picture shows the EPROM installed in the 1050 without the window covered - I have seen read errors caused by room light affecting the chip. Put a sticker over it to block light getting in.

[kbj]

3 hours ago, Nezgar said:

What programmer are you using? I notice in the picture above your EPROM is an ST 2732A that requires 21V programming voltage. The TL866 for instance maxxes out at about 18V in practice I've found, so it may have programmed "weakly" hence the occasional read errors. Better to get the ones that need 12.5Vpp.

 

Also, your picture shows the EPROM installed in the 1050 without the window covered - I have seen read errors caused by room light affecting the chip. Put a sticker over it to block light getting in.

Hey @Nezgar

 

Ahh, that was a pic from a different test 2732 that also didn't work - final, working one has a new sticker on top ?

 

I have a GQ4x4 and a TL866 II - used the GQ4x4 with an external PSU to do the programming on those 21v EPROMS..

 

Cheers,

 

Mark.

[kbj]

Hey All,

 

Another update again..

 

So after checking the calibration (and a drive that was working), all values seemed to be in tolerance according to @tf_hh's guide except for the first measurement (on pin 16 of the FDC) - it was flipping between 4/8uS - adjusted it, tested the drive and now it won't read disks or format them.  Strangely, I get error 139's and 142's in SpartDOSx now (using my SIDE3 cart).  I went back and remeasured Pin16 and all looked fine.

 

I ran the 1050 diagnostic disk and now it gives me motor errors - and when I check the RPM, it just gives me a constant 0 returned although the drive motor is spinning up etc.

 

Another weird thing is that you know when you format a disk, the head moves from track 0 up to the centre of the disk, then steps its way back down to track 0 during the format?  Well it now only does half of that - it formats and steps from track 0 to the centre of the disk and at that point the format stops with an error, motor stops spinning and the head doesn't move back to track 0.

 

So... I took a working drive mech from another drive, connected it to this drive's motherboard and all is good... so I'm not thinking that I have either a bad connection somewhere on the plugs/cables or a spindle motor ?

 

What do you guys think before I take this drive outside and drive back and forth over it with my car? ?

 

Cheers,

 

Mark.

 

 

[_The Doctor__]

sounds like a bad connector or wire, since it was working and then was disturbed and now does not. It's possible the intermittent condition was there all along being your gremlin. So, sounds like a broken wire or connector/connection to me... I once had to solder a wire exactly at connector that plugged the pins at the pcb for the head and at the spot where wire went into motor... seems they were stressed when they went searching about, a slight tug did them in. Looks ugly but works. You may have similar issues.

Edited March 16, 2021 by _The Doctor__

[kbj]

5 minutes ago, _The Doctor__ said:

sounds like a bad connector or wire, since it was working and then was disturbed and now does not. It's possible the intermittent condition was there all along being your gremlin. So, sounds like a broken wire or connector/connection to me... I once had to solder a wire exactly at connector that plugged the pins at the pcb for the head and at the spot where wire went into motor... seems they were stressed when they went searching about, a slight tug did them in. Looks ugly but works. You may have similar issues.

Yeah.. I was thinking the same myself but wanted to check that you guys didn't know something I didn't think of - like 'Ahhh, it's a diode or resistor blah..' I'll have a look at the schematics and identify the wire(s) that take the motor pulse back to the mainboard (think it arrives via Q5 or some transistor) and check for pulses there.. I gotta hope that its a break away from the motor closer to the pin header I guess.. 

 

Thanks for the insight @_The Doctor__... I'll report back ?

 

Cheers,

 

Mark.

[+kheller2]

The RPM tool reads sectors to figure out speed.

If it can't read (head issue) then you'll probably get wrong RPMs.  Assuming you are reading a properly formatted SD disk.

 

 

[+tf_hh]

13 hours ago, kbj said:

So after checking the calibration (and a drive that was working), all values seemed to be in tolerance according to @tf_hh's guide except for the first measurement (on pin 16 of the FDC) - it was flipping between 4/8uS - adjusted it, tested the drive and now it won't read disks or format them.

 

This can - normally - only happen when any disk was inserted before the bridge between TP8/TP9 (testmode ON) is made. 4 uS = 125 KHz (FM - Single Density) and 8 uS = 250 KHz (MFM - Double Density, which includes Atari´s "Medium Density" or "Enhanced Density" also). So if you measure 8 uS / 250 KHz and NO medium density formatted disk was inserted before, there´s a fault at the FDC or 6532 (RIOT). The only parts involved here.

 

[kbj]

3 hours ago, tf_hh said:

 

This can - normally - only happen when any disk was inserted before the bridge between TP8/TP9 (testmode ON) is made. 4 uS = 125 KHz (FM - Single Density) and 8 uS = 250 KHz (MFM - Double Density, which includes Atari´s "Medium Density" or "Enhanced Density" also). So if you measure 8 uS / 250 KHz and NO medium density formatted disk was inserted before, there´s a fault at the FDC or 6532 (RIOT). The only parts involved here.

 

Hey @tf_hh,

 

Thanks for that great explanation.  Can I just confirm the procedure then?

 

Should I insert a single density disk first, put the TP8/TP9 jumper on, then adjust for 4uS/125khz and then power down and power up everything and then insert a MFM disk and then check to make sure it's measuring 8uS/250KHz - or do I not need to do that at all?

 

I thought that all I needed to do was to power up without a disk in the drive, and adjust for 4uS/125Khz .?

 

Cheers Jurgen,

 

Mark.

[+tf_hh]

34 minutes ago, kbj said:

Should I insert a single density disk first, put the TP8/TP9 jumper on, then adjust for 4uS/125khz and then power down and power up everything and then insert a MFM disk and then check to make sure it's measuring 8uS/250KHz - or do I not need to do that at all?

 

I thought that all I needed to do was to power up without a disk in the drive, and adjust for 4uS/125Khz .?

 

Yes, this is the right way. No disk in drive, power on, set jumper between TP8 and TP9. I would suggest to wait 5 minutes after power-on until all chips have warmed up. Then set jumper and do measurement.

 

When you get 8 uS at Pin 16 of the FDC without any disk inserted before, there must be a fault. After power on, Pin 13 of the RIOT (6532) is set to high-level. This I/O is directly connected to Pin 37 of the FDC, which is /DDEN. A high-level here forces the FDC to work in FM mode, a low-level means MFM mode. When here´s a low level without any disk inserted after power-on, the trace is broken, or RIOT defect, or FDC.

 

[kbj]

On 3/17/2021 at 1:02 PM, tf_hh said:

 

Yes, this is the right way. No disk in drive, power on, set jumper between TP8 and TP9. I would suggest to wait 5 minutes after power-on until all chips have warmed up. Then set jumper and do measurement.

 

When you get 8 uS at Pin 16 of the FDC without any disk inserted before, there must be a fault. After power on, Pin 13 of the RIOT (6532) is set to high-level. This I/O is directly connected to Pin 37 of the FDC, which is /DDEN. A high-level here forces the FDC to work in FM mode, a low-level means MFM mode. When here´s a low level without any disk inserted after power-on, the trace is broken, or RIOT defect, or FDC.

 

Hey @tf_hh

 

Thanks for the great info Jurgen!  

 

Turns out that I had 2 issues.  I had a broken wire in the motor tacho return (yellow wire) and the calibration was out.  Repaired that and then ran the calibration steps again and all was great.  Drive is now working a treat!

 

Thanks again everyone - and hopefully this time the drive will go on to work for a long time to come.

 

Cheers,

 

Mark.