Still Trying

[justacruzr2]

2 hours ago, Stuart said:

Step and Direction look OK. They will read opposite on the WD chip pins and on the floppy connector as there are inverters on the output of the FDC card.

 

Does Step keeps pulsing even when the head has visibly reached track 00? See if you can work out how track 00 is detected - often by the head carriage operating a microswitch or breaking an LED sensor beam.

 

Also see if you can follow the track 00 output on the edge connector back to a logic chip on the floppy drive circuit board - the 'mechanism' may be fine but a faulty logic chip on the floppy output. Track 00 is active low - so should be high until it reaches track 00 then go low.

 

Have you got more than one drive you can test?

Glad to hear that.  This just got narrowed down to one thing.  The track 0 sensor consists of 2 parts, the sensor itself which is "u" shaped and is located in a bracket at the end of the head travel.  I believe it works like an "electric eye".  The other part is a finger extending from the side of the head carriage.  I think what happens is that when the head position reaches the end of its' travel to the outer edge, the finger breaks the LED sensor beam and should then set the track 0 register to LO.  According to the datasheet, if this input is not received it sets a seek error.  And that sounds very much like what's happening.  OK, I'll have to follow the circuit traces downstream to see what other devices (ICs, caps, resistors, transistors, diodes, etc) are involved.  The STEP test stopped pulsing when it reached track 0.  And yes I do have another drive.  2 more in fact.  One of the 2 is the brother of the drive I'm currently testing.  It was in the PEB at the same time as the spike so probably has the same problem.  I will do those tests again on it if you think I should.

 

PS  Love using that logic probe.  Should have got one years ago.

Edited May 15, 2023 by justacruzr2
more info

[Stuart]

Worth testing the other drives. To have all 3 exhibit the exact same fault would make me a bit suspicious.

 

Does STEP stop pulsing the moment the head stops moving? If the FDC is not getting the TRK00 indication, I think the datasheet said that it tries 255 steps then stops, so I would expect STEP to keep pulsing for a short time even though the head has reached the end of travel. (But it may take under a second to do all those 255 steps so might be difficult to be sure what is happening!)

 

Also important (if you didn't know) to only check the signal states while the drive is being accessed (the drive LED is on). Once the FDC stops trying to access the drive (drive LED goes off), the drive isolates its outputs and they're pulled high by resistors on the FDC.

Edited May 15, 2023 by Stuart

[justacruzr2]

6 minutes ago, Stuart said:

Worth testing the other drives. To have all 3 exhibit the exact same fault would make me a bit suspicious.

 

Does STEP stop pulsing the moment the head stops moving? If the FDC is not getting the TRK00 indication, I think the datasheet said that it tries 255 steps then stops, so I would expect STEP to keep pulsing for a short time even though the head has reached the end of travel.

 

Also important (if you didn't know) to only check the signal states while the drive is being accessed (the drive LED is on). Once the FDC stops trying to access the drive (drive LED goes off), the drive isolates its outputs and they're pulled high by resistors on the FDC.

That's what I read too...255.  I should probably try that again since I don't remember if it continued pulsing or not.  I thought I saw it stop but maybe I should have left the probe on there a little longer.  Yes, I'm only checking with the LED on and what gives me the longest time with the LED on is going into Extended Basic and typing in OLD DSK1.LOAD.  XB looks for a long time before it errors out.

[+Ksarul]

6 hours ago, Stuart said:

Step and Direction look OK. They will read opposite on the WD chip pins and on the floppy connector as there are inverters on the output of the FDC card.

 

Does Step keeps pulsing even when the head has visibly reached track 00? See if you can work out how track 00 is detected - often by the head carriage operating a microswitch or breaking an LED sensor beam.

 

Also see if you can follow the track 00 output on the edge connector back to a logic chip on the floppy drive circuit board - the 'mechanism' may be fine but a faulty logic chip on the floppy output. Track 00 is active low - so should be high until it reaches track 00 then go low.

 

Have you got more than one drive you can test?

On the floppy controller, the only thing between PIN 26 and the TR00* input of the chip is a 3.3KOhm SIP resistor with +5V on the other side. Here's the CorComp Schematic for the 2793 board.

[justacruzr2]

18 hours ago, Ksarul said:

On the floppy controller, the only thing between PIN 26 and the TR00* input of the chip is a 3.3KOhm SIP resistor with +5V on the other side. Here's the CorComp Schematic for the 2793 board.

Thanks and here's the results of last nights test.

 

 

Retesting the 1st drive for Track 00 at the edge connector:

With the head all the way to the center the probe showed a HI condition.  When I pressed ENTER, in XB, on the command OLD DSK1.LOAD, there was 1 pulse and it switched to LO, 3 more pulses while LO and when the head retracted all the way there was 1 pulse and it switched back to HI.  Did this 3 times to be sure.  One thing I didn't mention before is that when the head has retracted all the way it steps in and out the distance of about 1 track rapidly about 20 times and then there is a slight pause and then it repeats that 1 more time and stops.  It's like it vibrates or chatters.

 

Another interesting thing was the PCB of the Track 00 sensor.  There are 4 wires soldered to it at its' edge that terminate in a 4 pin connector that connects that small PCB to the main PCB.  I have circled it in red in the picture.  Visible is the "finger" (black, straight at the back and tapered in the front), the sensor itself (black, about a 1/2" to the left of the "finger" at the end of the rails, the wires I'm talking about and the connector.  With everything re-assembled and the drive connected and electrically active, but not accessing, when I touched the probe to the white wire, with the head fully retracted, it read LO.  I moved it out of that position and the probe registered HI.  Of the other 3 wires, 2 remained LO no matter where the head was positioned and 1 had no reading at all (still maybe a problem?).

 

Here are the results of testing the 2nd drive (the other one that was in the PEB at the time of the spike):

==============================================================
                 WD2793 Controller Test

Pin         Function           Signal Type              Test Result
===      ========     ===========     ===========
15         Step                  Output                     LO - no change no pulsing
16         Direction            Output                     LO - no change no pulsing
34         Track 00             Input                      *see below
35         Index Pulse        Input                       HI - no change no pulsing

==============================================================
                  Floppy Drive Test

Pin         Function            Signal Type              Test Result
===      ========     ===========     ===========
8           Index Pulse       Output                     HI - no change no pulsing
18         Direction           Input                       HI - no change no pulsing
20         Step                 Input                       *see below
26         Track 00           Output                     **see below

==============================================================

FYI I haven't replaced any components on the 2nd drive like I did on the 1st drive which may explain why the results are different.

 

*Testing the 2nd drive for STEP at the edge connector and testing the WD2793 pin for TRACK 00 (These 2 different tests gave the same results):

Started HI before accessing the drive.  When accessing the drive it pulsed once switched to LO, 1 pulse then a fast switch to HI (more like a blink) and back to LO then another pulse and back to HI.  It did this 2 times and then shut the drive down.  Checked this 3 times too.

 

**Testing the 2nd drive for Track 00 at the edge connector:

Did this the same way as I did above for the drive 1 retest.  Results were a bit different.  It started HI then when I entered OLD DSK1.LOAD it pulsed and switched to LO and went from HI to LO 5 times with a pulse before every switch.  The head does not chatter on this one.

 

 

[Stuart]

Hmmm ... so the TRK00 signal from the drive is definitely not 'stuck'. Possibly the head is getting back to track 00, then the FDC detects a read error so repeatedly steps the head in a track then back out to reposition it and tries reading again, then finally gives up with a read error?

 

The 2793 board has a couple of variable resistors and a variable capacitor which might possibly require adjusting to give a good read signal. Did you say you have a 1773 board as well - I don't think that has any adjustable components on it?

 

I think at this point I'd start to think about either trying to borrow a known working drive with a known working disk to try, or if you could post your FDC, one of the drives and a disk to a kindly samaritan to test with their known working FDC and drives. I would offer but I'm in the UK so not really practical.

 

 

 

[justacruzr2]

Sounds like a reasonable conclusion.  Yes, this board has 2 adjustable resistors and the one you may be referring to is called the Data Separation resistor.  I have not adjusted it firstly because I thought I would first see if the drive would read data and see if I get an error message (maybe that's what's already happening as you say) and secondly, and I do have the datasheet and instructions on how to adjust it, it suggests using a scope.  That's not cheap and I really wouldn't have any further use for it when done.  So I thought what I might do is adjust it on the fly.  And to do that I would have to get the drive to search long enough to see if turning the screw makes a difference.  That's a bit tricky.  You usually don't get more than about 3 or 4 seconds.  And yes I also have the Rev A board as well.  I think it did the same thing too but it's probably worth checking again with that board.  At least that would narrow it down to "it's definitely the drive".  Unlikely that both cards would have the same exact problem.  What do you think?  I am having a friend over tonight so I may only have time to check the Rev A board to see if there's any difference.  I appreciate all your assistance.

[+Ksarul]

Testing with a second controller will definitely help. . .

[justacruzr2]

19 hours ago, Ksarul said:

Testing with a second controller will definitely help. . .

There was no difference with the Rev A (WD1773) card.  So this definitely points to the drive.  And I guess that the data separation resistor hasn't come into play yet.  2 nights ago I did a little more testing of the main PCB on the drive, comparing it to a 4th drive I have.  I actually have 4 drives but the 4th drive never was in play because I bought it off eBay as a "AS IS for parts or repair".  And that's exactly why I bought it...for parts.  Well when testing one of the resistor packs, a 12 pin one, I found some interesting differences.  On the 4th drive, putting the multimeter negative probe on the pin with the dot, as I moved from pin to pin, the meter showed consistent resistance and the readings would start out low and then increase to a point and stop.  This told me that these pins are going thru caps and the caps are OK.  On the drive we've been working with, the readings were much higher and they did not increase as if going thru a cap.  So I'm thinking either the resistor pack is defective or there are one or more caps that are defective.  Sound right?  What do you think?

[Stuart]

Could just be down to different circuit designs. Unless you can find circuit diagram of those floppy drives ...

[justacruzr2]

2 hours ago, Stuart said:

Could just be down to different circuit designs. Unless you can find circuit diagram of those floppy drives ...

That 4th drive is the exact same model.  And I have tried that but no luck.  I haven't replaced the resistor packs yet or the caps so I think I will try that first and see what happens.  Fortunately I acquired (for free I might add) the new parts inventory of a TV and electronics repair shop that went out of business.  He was going to just throw it all out and I asked him if I could have it.  And it did include the Holy Grail of capacitors.....polystyrene caps.  Almost impossible to find these days.  These drives have 5 of that type on the main PCB.  Somewhere along the line I'm going to hit the bad part.

[+InsaneMultitasker]

Are you able to swap out the 9901 chip?   In my repair experience, this component is often the culprit with the Corcomp cards. It might not be your card's issue but it is worth checking into.  Multiple drives 'not working' is a bit suspicious beyond a cable or termination problem, IMHO.

 

Edit: Was the REV A card working with drives at one point?

[justacruzr2]

On 5/19/2023 at 12:18 AM, InsaneMultitasker said:

Are you able to swap out the 9901 chip?   In my repair experience, this component is often the culprit with the Corcomp cards. It might not be your card's issue but it is worth checking into.  Multiple drives 'not working' is a bit suspicious beyond a cable or termination problem, IMHO.

 

Edit: Was the REV A card working with drives at one point?

Everything on both cards was replaced with new parts.  And yes.