CBM 3040 Disk Drive - Page 4 - UK Vintage Radio Repair and Restoration Discussion Forum

SiriusHardware · 25th Jan 2024, 5:13 pm

My guess would be that the edge of the cable with the darker red line denotes the wire connected to pin 1, and then I would have said that pins 3, 5, 7, 9, etc are the ones going across from right to left on the upper row in your photo in #58, and the ones on the lower row are, from right to left, 2, 4, 6, 8, 10.

But like you I don't know for sure, so the thing to do is to unplug the connector from P7 which will presumably expose pin row connections on the PCB.

Draw an oversized pencil sketch of the overhead view of the P7 pin connections.

Look at the circuit diagram:

UA3 pin 11 goes to P7 pin 15. Put your meter into continuity testing mode, put one probe on UA3 pin 11 and touch the other probe to each of the P7 pin row connections in turn - whichever one is directly connected to UA3 pin 11, that is pin 15 of P7. Note that on your sketch of the connector.

The diagram says UB3 pin 3 goes to P7 pin 14. Use your meter to determine which physical pin of P7 UB3 pin 3 is connected to, and mark that P7 pin as pin 14 on your sketch.

Do the same for UC1 pin 5 (connected to pin 9 of P7)
Same for UC1 pin 1 (connected to pin 18 of P7)
Same for UC1 pin 13 (connected to pin 17 of P7)

And so on.

Once you've identified about half a dozen of the P7 pins, it should be fairly obvious what the overall numbering sequence for the rest of the pins must be.

SiriusHardware · 25th Jan 2024, 5:47 pm

Can we have a little bit more info on what look to be small PCBs located near the heads, each with one 2N3904 transistor and three diodes on them - they are contained within their own outlines on the diagram. What do they look like and where are they?

ScottishColin · 25th Jan 2024, 6:38 pm

Done that - it's neither of the responses I gave in posts 59 and 60 so I will re-test them.

For anyone who ever stumbles across this in future, the 20 pin connector P7 goes from pins 1-10 from top right along the top row to top left. Then 11-20 from the bottom right pin to the bottom left pin.

Colin.

Quote:

Originally Posted by SiriusHardware

My guess would be that the edge of the cable with the darker red line denotes the wire connected to pin 1, and then I would have said that pins 3, 5, 7, 9, etc are the ones going across from right to left on the upper row in your photo in #58, and the ones on the lower row are, from right to left, 2, 4, 6, 8, 10.

But like you I don't know for sure, so the thing to do is to unplug the connector from P7 which will presumably expose pin row connections on the PCB.

Draw an oversized pencil sketch of the overhead view of the P7 pin connections.

Look at the circuit diagram:

UA3 pin 11 goes to P7 pin 15. Put your meter into continuity testing mode, put one probe on UA3 pin 11 and touch the other probe to each of the P7 pin row connections in turn - whichever one is directly connected to UA3 pin 11, that is pin 15 of P7. Note that on your sketch of the connector.

The diagram says UB3 pin 3 goes to P7 pin 14. Use your meter to determine which physical pin of P7 UB3 pin 3 is connected to, and mark that P7 pin as pin 14 on your sketch.

Do the same for UC1 pin 5 (connected to pin 9 of P7)
Same for UC1 pin 1 (connected to pin 18 of P7)
Same for UC1 pin 13 (connected to pin 17 of P7)

And so on.

Once you've identified about half a dozen of the P7 pins, it should be fairly obvious what the overall numbering sequence for the rest of the pins must be.

ScottishColin · 25th Jan 2024, 6:55 pm

OK. Retest of P7 pins 2, 12 and 13

Good drive
Idle pin 2 - 0.1v
Idle pin 12 - 6.64v
Idle pin 13 - 0.14v

HEADER pin 2 - 1.0->4.6v
HEADER pin 12 - 0.14v
HEADER pin 13 - 1.7->5.6v

Bad drive
Idle pin 2 - 0.1v
Idle pin 12 - 6.77v
Idle pin 13 - 0.124v

HEADER pin 2 - 0->6.79v
HEADER pin 12 - 0.1v
HEADER pin 13 - 0.126->3.3v

Colin.

ScottishColin · 25th Jan 2024, 6:57 pm

The only other PCBs I can see are attached. There's one on the back of each drive. Not sure this is what you're looking for?

As a reminder, all the drives work fine when attached to the working analogue board.

Colin.

Quote:

Originally Posted by SiriusHardware

Can we have a little bit more info on what look to be small PCBs located near the heads, each with one 2N3904 transistor and three diodes on them - they are contained within their own outlines on the diagram. What do they look like and where are they?

SiriusHardware · 25th Jan 2024, 7:54 pm

Quote:

As a reminder, all the drives work fine when attached to the working analogue board.

I did remember that.

I've worked out where they are now - the dotted lines drawn around the 2N3904s and three diodes (image #1) seemed to suggest that those little circuit sections were on separate little PCBs, perhaps mounted closer to the heads - but on closer inspection I noticed these 'modules' or whatever they were had Ux numbers (like ICs) - so I looked at the layout diagram and found them occupying real estate originally intended to be populated by 2 ICs (image #2).

I was intending to follow up on RonBryan's observation that the wayward voltages on UC5 could be explained by a missing common connection from the 2N3904 transistors' emitters, but that is less likely with them being mounted on the PCB than it would be if they were mounted offboard and connected by wires.

Apologies, but I am going to have to look at your revised results later or tomorrow - being dragged away again.

SiriusHardware · 26th Jan 2024, 12:44 am

Referring to the results in #64, and the circuit diagram:

The upper 2N3904 (part of 'UD3')is the head enable / disable control transistor for head 1. By default, it is turned on because its base is pulled up to +12VB by a 5.6K resistor which is another part of the phantom 'UD3'. It can be turned off by taking P7 pin 2 low.

The lower 2N3904 (part of 'UD5') is the head enable / disable control transistor for head 0. By default, it is turned on because its base is pulled up to +12VB by a 5.6K resistor which is part of the phantom 'UD5'. That one can be turned off by taking P7 pin 12 low.

Pin 13 would appear to be the write / erase control line. When taken high, it:

-Turns on Q9, which turns on UC3 (10,9,8), switching +12VB onto UC3 pin 8

-Turns on Q10, which connects the head 'common' line directly to 0V (when not in write mode, the head 'common' line is connected to the cathode of the 6V zener).

-Also turns on the ERASE control transistor UC3 (5,6,7), via UD1 (9,8)

-Supplies power to the WRITE control transistors UC3 (1,2,3) and UC3 (12,13,14) via R24.

-The 'Q' and 'Q' outputs from IC UB3 are always in opposite states, so while power is applied to both the W1 and W2 write control transistors, only one of those two is ever turned on by their respective control signals from IC UB3 via UB1 (13,12) and via UB1 (1,2).

Code:

Good drive
Idle pin 2 - 0.1v - Head 1 is held 'off'
Idle pin 12 - 6.64v - Head 0 is allowed 'on'
Idle pin 13 - 0.14v - Write line is inactive

HEADER pin 2 - 1.0->4.6v - Head 1 is active
HEADER pin 12 - 0.14v - Head 0 is held 'off'
HEADER pin 13 - 1.7->5.6v - Write line is active

and:

Code:

Bad drive
Idle pin 2 - 0.1v - Head 1 is held 'off'
Idle pin 12 - 6.77v - Head 0 is allowed 'on'
Idle pin 13 - 0.124v - Write line is inactive

HEADER pin 2 - 0->6.79v - Head 1 is active
HEADER pin 12 - 0.1v - Head 0 is held 'off'
HEADER pin 13 - 0.126->3.3v - Write line is active.

Broadly speaking the states of these control lines are as would be expected on both drives. The highly variable levels (measured with a meter) on some of the lines is indicating the fact that the voltages there are not actually static voltages, but rather periods of rapid on / off digital activity. We may look at those more closely with a scope, in a bit.

Now that we know head 0 is enabled for read even when the drive is in 'idle' mode, we can focus more closely on the voltages around UC5 under purely DC conditions to see what's going on in that area. We already have the voltages for UC5 measured when the 'bad' drive is idle (in #49).
.
Unfortunately I'm out of time tonight, so will have to resume some time tomorrow - anyone who can look at it earlier, please jump in.

ScottishColin · 26th Jan 2024, 3:28 pm

Just poking around a little to find any differences. Both drives powered on but not plugged into a PET.

I went through the diodes and most are close enough to each other with the following two exceptions:

CR25 Bad drive - 0.09v
CR25 Good drive - 0.77v

CR22 Bad drive - 0.122v
CR22 Good drive - 0.77v

Just wondered if they might be another part of the jigsaw.

Colin.

SiriusHardware · 26th Jan 2024, 4:57 pm

Interesting observation. Power off and use the 'diode test' feature of one of your multimeters to check the diodes, initially in-circuit and always with the black meter lead to the 'banded' end of the diode and the red lead to the plain end.

In this mode the meter indicates the forward voltage drop across the diode under test, and for a healthy silicon diode that will be about 0.6V to 0.7V.

If you see a reading with a significantly smaller voltage drop or even no voltage drop at all that would suggest the diode has failed low resistance or short-circuit but to be sure of that you would need to lift one end of the suspect diode and measure only the diode, not the diode plus the surrounding circuitry.

Bottom line: Whatever forward voltage drops you see across CR25 and CR22 of the good drive, you should see virtually the same readings across CR25 and CR22 of the bad drive, if you don't, investigate.

ronbryan · 26th Jan 2024, 5:45 pm

It does look as if the LM592 UA5 has failed. If diodes CR25 and CR22 check out ok, it seems that the chip is taking no current and from post #47 readings, the inputs and emitters of the input long-tailed pair (ltp) and outputs are all stuck high at 8.9V, 10.8V or 11.4V respectively. This failure could be attributable to a loss of the current mirror that sinks all of the current from the ltp and output stages.

Ron

ScottishColin · 26th Jan 2024, 6:24 pm

Good drive CR22 - 0.64v
Good drive CR25 - 0.64v

Bad drive CR22 - 0.707v
Bad drive CR25 - 0.662v

Colin.

Quote:

Originally Posted by SiriusHardware

Interesting observation. Power off
and use the 'diode test' feature of one of your multimeters to check the diodes, initially in-circuit and always with the black meter lead to the 'banded' end of the diode and the red lead to the plain end.

In this mode the meter indicates the forward voltage drop across the diode under test, and for a healthy silicon diode that will be about 0.6V to 0.7V.

If you see a reading with a significantly smaller voltage drop or even no voltage drop at all that would suggest the diode has failed low resistance or short-circuit but to be sure of that you would need to lift one end of the suspect diode and measure only the diode, not the diode plus the surrounding circuitry.

Bottom line: Whatever forward voltage drops you see across CR25 and CR22 of the good drive, you should see virtually the same readings across CR25 and CR22 of the bad drive, if you don't, investigate.

ScottishColin · 26th Jan 2024, 6:26 pm

The other day, I bought a Phillips NE592N14 in case. Current chip is marked as a NE592N so that looks like a suitable replacement.

Votes for replacing it?

Colin.

SiriusHardware · 26th Jan 2024, 6:56 pm

The diodes seem OK, given that you are measuring them in-circuit.

You can try your new chip and maybe that will get you a step further forward.

ScottishColin · 26th Jan 2024, 6:57 pm

Thank you.

We have an 18-year old back from her second heart surgery so I'll probably try to find a bit of time tomorrow.

Colin.

Quote:

Originally Posted by SiriusHardware

The diodes seem OK, given that you are measuring them in-circuit.

You can try your new chip and maybe that will get you a step further forward.

SiriusHardware · 26th Jan 2024, 9:47 pm

We wish her well, of course.

ScottishColin · 27th Jan 2024, 4:11 pm

Thanks very much.

She's rushing around like she never had an operation. I guess that's what happens when you have youth on your side.

Colin.

Quote:

Originally Posted by SiriusHardware

We wish her well, of course.

ScottishColin · 27th Jan 2024, 4:14 pm

Replaced it and everything's working now - HEADERing works on both drives and I've run several of the Commodore disk diagnostics and it's all good.

Now I need to restore the case which has a bit of rust but the mechanics are all good.

In summary, 4 RAM chips replaced and the NE592N chip replaced.

Thanks for your help again.

Colin.

Quote:

Originally Posted by SiriusHardware

The diodes seem OK, given that you are measuring them in-circuit.

You can try your new chip and maybe that will get you a step further forward.

SiriusHardware · 27th Jan 2024, 6:48 pm

Glad she appears to be doing well, also glad that ronbryan's suspicions about a likely UA5 fault proved to be correct.

SiriusHardware · 28th Jan 2024, 12:04 am

I think there was also one dud 2332 ROM as well?

ScottishColin · 28th Jan 2024, 5:05 pm

You're correct. So to re-summarise:

On the main motherboard

- 4 x 2114s
- 1 x 901468-11 ROM replaced with a 217128 using the adapter below:
https://myretrostore.co.uk/product/23xx/

On the disk drive PCB
- 1 x NE592N replaced with a Phillips NE592N14

I'm lucky that I have another working device to compare it with; I think it made things easier for me.

Colin.

Quote:

Originally Posted by SiriusHardware

I think there was also one dud 2332 ROM as well?

25th Jan 2024, 5:13 pm	#61
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 13,865	Re: CBM 3040 Disk Drive My guess would be that the edge of the cable with the darker red line denotes the wire connected to pin 1, and then I would have said that pins 3, 5, 7, 9, etc are the ones going across from right to left on the upper row in your photo in #58, and the ones on the lower row are, from right to left, 2, 4, 6, 8, 10. But like you I don't know for sure, so the thing to do is to unplug the connector from P7 which will presumably expose pin row connections on the PCB. Draw an oversized pencil sketch of the overhead view of the P7 pin connections. Look at the circuit diagram: UA3 pin 11 goes to P7 pin 15. Put your meter into continuity testing mode, put one probe on UA3 pin 11 and touch the other probe to each of the P7 pin row connections in turn - whichever one is directly connected to UA3 pin 11, that is pin 15 of P7. Note that on your sketch of the connector. The diagram says UB3 pin 3 goes to P7 pin 14. Use your meter to determine which physical pin of P7 UB3 pin 3 is connected to, and mark that P7 pin as pin 14 on your sketch. Do the same for UC1 pin 5 (connected to pin 9 of P7) Same for UC1 pin 1 (connected to pin 18 of P7) Same for UC1 pin 13 (connected to pin 17 of P7) And so on. Once you've identified about half a dozen of the P7 pins, it should be fairly obvious what the overall numbering sequence for the rest of the pins must be.

25th Jan 2024, 5:47 pm	#62
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 13,865	Re: CBM 3040 Disk Drive Can we have a little bit more info on what look to be small PCBs located near the heads, each with one 2N3904 transistor and three diodes on them - they are contained within their own outlines on the diagram. What do they look like and where are they?

26th Jan 2024, 12:44 am	#67
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 13,865	Re: CBM 3040 Disk Drive Referring to the results in #64, and the circuit diagram: The upper 2N3904 (part of 'UD3')is the head enable / disable control transistor for head 1. By default, it is turned on because its base is pulled up to +12VB by a 5.6K resistor which is another part of the phantom 'UD3'. It can be turned off by taking P7 pin 2 low. The lower 2N3904 (part of 'UD5') is the head enable / disable control transistor for head 0. By default, it is turned on because its base is pulled up to +12VB by a 5.6K resistor which is part of the phantom 'UD5'. That one can be turned off by taking P7 pin 12 low. Pin 13 would appear to be the write / erase control line. When taken high, it: -Turns on Q9, which turns on UC3 (10,9,8), switching +12VB onto UC3 pin 8 -Turns on Q10, which connects the head 'common' line directly to 0V (when not in write mode, the head 'common' line is connected to the cathode of the 6V zener). -Also turns on the ERASE control transistor UC3 (5,6,7), via UD1 (9,8) -Supplies power to the WRITE control transistors UC3 (1,2,3) and UC3 (12,13,14) via R24. -The 'Q' and 'Q' outputs from IC UB3 are always in opposite states, so while power is applied to both the W1 and W2 write control transistors, only one of those two is ever turned on by their respective control signals from IC UB3 via UB1 (13,12) and via UB1 (1,2). Code: Good drive Idle pin 2 - 0.1v - Head 1 is held 'off' Idle pin 12 - 6.64v - Head 0 is allowed 'on' Idle pin 13 - 0.14v - Write line is inactive HEADER pin 2 - 1.0->4.6v - Head 1 is active HEADER pin 12 - 0.14v - Head 0 is held 'off' HEADER pin 13 - 1.7->5.6v - Write line is active and: Code: Bad drive Idle pin 2 - 0.1v - Head 1 is held 'off' Idle pin 12 - 6.77v - Head 0 is allowed 'on' Idle pin 13 - 0.124v - Write line is inactive HEADER pin 2 - 0->6.79v - Head 1 is active HEADER pin 12 - 0.1v - Head 0 is held 'off' HEADER pin 13 - 0.126->3.3v - Write line is active. Broadly speaking the states of these control lines are as would be expected on both drives. The highly variable levels (measured with a meter) on some of the lines is indicating the fact that the voltages there are not actually static voltages, but rather periods of rapid on / off digital activity. We may look at those more closely with a scope, in a bit. Now that we know head 0 is enabled for read even when the drive is in 'idle' mode, we can focus more closely on the voltages around UC5 under purely DC conditions to see what's going on in that area. We already have the voltages for UC5 measured when the 'bad' drive is idle (in #49). . Unfortunately I'm out of time tonight, so will have to resume some time tomorrow - anyone who can look at it earlier, please jump in.

26th Jan 2024, 3:28 pm	#68
ScottishColin Nonode Join Date: May 2012 Location: Perth, Scotland Posts: 2,445	Re: CBM 3040 Disk Drive Just poking around a little to find any differences. Both drives powered on but not plugged into a PET. I went through the diodes and most are close enough to each other with the following two exceptions: CR25 Bad drive - 0.09v CR25 Good drive - 0.77v CR22 Bad drive - 0.122v CR22 Good drive - 0.77v Just wondered if they might be another part of the jigsaw. Colin.

26th Jan 2024, 4:57 pm	#69
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 13,865	Re: CBM 3040 Disk Drive Interesting observation. Power off and use the 'diode test' feature of one of your multimeters to check the diodes, initially in-circuit and always with the black meter lead to the 'banded' end of the diode and the red lead to the plain end. In this mode the meter indicates the forward voltage drop across the diode under test, and for a healthy silicon diode that will be about 0.6V to 0.7V. If you see a reading with a significantly smaller voltage drop or even no voltage drop at all that would suggest the diode has failed low resistance or short-circuit but to be sure of that you would need to lift one end of the suspect diode and measure only the diode, not the diode plus the surrounding circuitry. Bottom line: Whatever forward voltage drops you see across CR25 and CR22 of the good drive, you should see virtually the same readings across CR25 and CR22 of the bad drive, if you don't, investigate.

25th Jan 2024, 6:55 pm	#64
ScottishColin Nonode Join Date: May 2012 Location: Perth, Scotland Posts: 2,445	Re: CBM 3040 Disk Drive OK. Retest of P7 pins 2, 12 and 13 Good drive Idle pin 2 - 0.1v Idle pin 12 - 6.64v Idle pin 13 - 0.14v HEADER pin 2 - 1.0->4.6v HEADER pin 12 - 0.14v HEADER pin 13 - 1.7->5.6v Bad drive Idle pin 2 - 0.1v Idle pin 12 - 6.77v Idle pin 13 - 0.124v HEADER pin 2 - 0->6.79v HEADER pin 12 - 0.1v HEADER pin 13 - 0.126->3.3v Colin.

26th Jan 2024, 5:45 pm	#70
ronbryan Nonode Join Date: Jan 2003 Location: Guildford, Surrey, UK. Posts: 2,014	Re: CBM 3040 Disk Drive It does look as if the LM592 UA5 has failed. If diodes CR25 and CR22 check out ok, it seems that the chip is taking no current and from post #47 readings, the inputs and emitters of the input long-tailed pair (ltp) and outputs are all stuck high at 8.9V, 10.8V or 11.4V respectively. This failure could be attributable to a loss of the current mirror that sinks all of the current from the ltp and output stages. Ron Attached Thumbnails Last edited by ronbryan; 26th Jan 2024 at 5:56 pm.

26th Jan 2024, 6:26 pm	#72
ScottishColin Nonode Join Date: May 2012 Location: Perth, Scotland Posts: 2,445	Re: CBM 3040 Disk Drive The other day, I bought a Phillips NE592N14 in case. Current chip is marked as a NE592N so that looks like a suitable replacement. Votes for replacing it? Colin.

26th Jan 2024, 6:56 pm	#73
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 13,865	Re: CBM 3040 Disk Drive The diodes seem OK, given that you are measuring them in-circuit. You can try your new chip and maybe that will get you a step further forward.

26th Jan 2024, 9:47 pm	#75
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 13,865	Re: CBM 3040 Disk Drive We wish her well, of course.

27th Jan 2024, 6:48 pm	#78
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 13,865	Re: CBM 3040 Disk Drive Glad she appears to be doing well, also glad that ronbryan's suspicions about a likely UA5 fault proved to be correct.

28th Jan 2024, 12:04 am	#79
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 13,865	Re: CBM 3040 Disk Drive I think there was also one dud 2332 ROM as well?