Commodore 8050 dual disk drive - Page 5 - UK Vintage Radio Repair and Restoration Discussion Forum

TonyDuell · 25th Aug 2023, 6:19 pm

I think that's rather over-stating it. Turns out there is a mask option to use 2 of the port pins as additional chip selects (polarity selectable in the mask, too). But that's hardly being able to put the ROM anywhere in the memory map.

Looking at the circuit diagrams, the 8050 seems not to do this.The only chip select for the ROM section is the RS0/ pin.

SiriusHardware · 25th Aug 2023, 6:24 pm

In which case it should be easier to read out the contents, should it prove necessary, than appeared at first glance. Does it actually need a clock for the purposes of reading out the ROM? ROMs are, by their very definition, static devices.

How do we solve the mystery of the four flashes? Colin can quite easily test the 2114s, I think.

TonyDuell · 25th Aug 2023, 6:33 pm

The datasheet is not clear as to whether you need the phi2 clock for reading the ROM. It's shown as going into the chip select block, so it might be needed.

[It's not going to apply here, but there most certainly are ROMs that need a clock input. Core-on-a-rope being the obvious one, along with the inductively coupled PCB ROM in an HP9100]

At this stage don't worry about the 6530. The diagnostics must test the RAM first and I don't think that's happening. Test the 2114s and we'll go on from there.

ScottishColin · 25th Aug 2023, 6:45 pm

OK. I'll make a start on the 2114s and report back - probably tomorrow.

ScottishColin · 26th Aug 2023, 8:50 pm

So two of the 2114s were indeed duff.

The left hand of the two drives now reads diskettes from 1984 which I do like a lot.the right hand drive doesn't spin up so there's still more to come but we've definitely moved forward today.

Colin.

SiriusHardware · 26th Aug 2023, 10:44 pm

Good show, Colin, and another good idea from Tony.

Mark1960 · 27th Aug 2023, 12:42 am

I guess first thing to try is swapping the left and right drives and see if the problem follows the drive, or even the connecting cables.

TonyDuell · 27th Aug 2023, 5:14 am

These drive units have a pair of 'bare' mechanisms with a common 'analogue board', so there aren't any swappable cables.

Which drive units do you have? The problem could be the spindle motor (which is a DC permanant magnet motor with brushes/commutator) or a fault on the analogue board.

ScottishColin · 27th Aug 2023, 11:58 am

Well according to the ROMs installed, it's DOS v 2.5 and they're Micropolis drives.

I think my next task is to get them out, check obvious wiring issues and libricate.

It's interesting that the one without the PCB works fine - I'm hoping that means the PCB is good and the fault is confined to the motor.

Although having said that, my experience with motors lately hasn't been so good.....

Colin.

Quote:

Originally Posted by TonyDuell

These drive units have a pair of 'bare' mechanisms with a common 'analogue board', so there aren't any swappable cables.

Which drive units do you have? The problem could be the spindle motor (which is a DC permanant magnet motor with brushes/commutator) or a fault on the analogue board.

TonyDuell · 27th Aug 2023, 12:59 pm

That looks like board covered by the circuit diagram on pages 12-14 of the .pdf service manual. Yes, your drives are Micropolis.

Page 13 shows the spindle motor control circuit, there are 2 indentical circuits, one for each drive. It is not an elegant or well-designed circuit, alas.

The spindle motor is a DC motor with an AC tachogenerator built in. The better designed spindle motor controllers take the frequency from that generator as the speed reference. This one simply half-wave rectifies it (CR20) and smooths it (C21). The voltage there is thus depends on the speed of the motor. It's compared against a reference voltage from the slider of R50 by the op-amp U12. Output of that drives the power darlington transistor Q1.

The current though the motor is effectively sensed by R47. If it's too high then Q3 reduces the drive to Q1. Also Q1 can be forced off via CR26 if the output of U11b (pin 6) is low.

The spindle motor itself has 4 wires and will connect via part of the molex connector on the appropriate edge of the PCB. Trace the wires to find it. Normally red and blue are the motor, yellow and green the tachogenerator.

To start with, with the unit off, no disks in, unplug the motor from the PCB. Connect 9V or so DC (a PP3 battery should do it) between the red and blue wires, Red +ve, but if you get it wrong the spindle will simply run backwards. The motor should then run smoothly, the spindle should spin. If it doesn't, time to repair/replace the motor. If it does then the fault is on the PCB, which is not hard to repair.

SiriusHardware · 27th Aug 2023, 1:02 pm

I would rather hope it is a fault on one channel of the 'analogue' board, since that is nearly all discrete electronics and quite fixable.

Assuming the working drive to be drive 0 and the non-worker to be drive 1, and now that we know the units are Micropolis, you can try measuring the DC voltage between pins 1 and 4 of analogue PCB connector P5 when the non-worker should be running. You should see about 12V there under those circumstances. Also try measuring the resistance of resistor R47 (in-circuit, with power off).

TonyDuell · 27th Aug 2023, 1:24 pm

What I normally do is to test the motor on a 9V battery (or bench supply). If that works, lift one end of CR26 (for drive 1), connect the motor to the analogue PCB, and power up. With CR26 open, the motor should run _at the correct speed_ all the time. If not, then debug the control circuitry. If it does then work back through the simple bit of digital electronics shown on that page of the circuit diagram.

ScottishColin · 28th Aug 2023, 11:44 am

Quick update - both drives out and visually inspected and they look ok.

9V DC applied to the wiring as in post 90 and both motors spin ok. Now to put it back together to try the next tests.

Colin.

TonyDuell · 28th Aug 2023, 11:48 am

That's good news. Replacement motors are semi-hard to find, but the motor control circuit on the PCB is all standard components thst you can get.

ScottishColin · 28th Aug 2023, 7:05 pm

R47 has 1.3Ω resistance.

I've attached the photo so I've got this right. P5 is on the left hand side of the board and that connects to the drive that works.

P8 is on the right hand side and (obviously) is connected to the non-working drive.

I'm assuming we're looking at P8 (which I know has the connectors on incorrectly - all fixed before I tested anything)?

While we're here, when I insert a disk into the working drive, it spins for a short amount of time as soon as inserted.

When I put a disk into the non-working drive (which is labelled Drive 0), there is no spin (although we know the motor works from me popping 9V up it).

Colin.

Quote:

Originally Posted by SiriusHardware

I would rather hope it is a fault on one channel of the 'analogue' board, since that is nearly all discrete electronics and quite fixable.

Assuming the working drive to be drive 0 and the non-worker to be drive 1, and now that we know the units are Micropolis, you can try measuring the DC voltage between pins 1 and 4 of analogue PCB connector P5 when the non-worker should be running. You should see about 12V there under those circumstances. Also try measuring the resistance of resistor R47 (in-circuit, with power off).

SiriusHardware · 28th Aug 2023, 8:04 pm

Assuming the applicable diagram to be on page 13 then it looks as though connector P5, rather than P8, is the one relevant to drive 1 (as opposed to drive 0). You should see ~12VDC between terminals 1 and 4 when the motor should be running, but since you have already proven the motor to be OK I think you will almost certainly not see that 12VDC on those pins.

Follow up Tony's suggestion of temporarily unsoldering one end of diode CR26 and power the drive up with that diode disconnected. Does the motor run continuously with that diode disconnected?

ScottishColin · 28th Aug 2023, 9:57 pm

Lifting CR26 makes the working drive (Drive 1) spin continuously at power on.

Colin.

ScottishColin · 28th Aug 2023, 10:06 pm

Interestingly (although I appreciate that might be the wrong word), my board is marked as 8050008 on the back.

Colin.

ScottishColin · 28th Aug 2023, 10:12 pm

And even more interestingly, this page states that the PCB with 8050008 on it is actually schematic 8050005 with PCB assembly 8050006-1.

https://www.amiga-stuff.com/hardware/cbmboards.html

I don't think I could be any more confused right now.

Colin.

SiriusHardware · 28th Aug 2023, 11:51 pm

Quote:

And even more interestingly, this page states that the PCB with 8050008 on it is actually schematic 8050005 with PCB assembly 8050006-1.

The bare board with no components fitted has a part number usually etched on the PCB at the same time as the rest of the printed circuit layout. The board + components fitted then gets given a a different part number which tends to be hand written on the top side of the PCB.

Quote:

Lifting CR26 makes the working drive (Drive 1) spin continuously at power on.

Right, so the dud drive is in fact drive 0? In that case - as you were, P8 is indeed the correct connector to be looking at and you now need to try the same dodge with CR22 to see if lifting one end of that diode makes drive 0's motor run continually. Ordinarily, we would expect that the first drive would be physical drive 0 and the second drive would be physical drive 1.

Sorry about the mix up, but I think both Tony and I were both under the impression that it was drive 2 (of 2) which was not working.

25th Aug 2023, 6:19 pm	#81
TonyDuell Dekatron Join Date: Jun 2015 Location: Biggin Hill, London, UK. Posts: 5,055	Re: Commodore 8050 dual disk drive I think that's rather over-stating it. Turns out there is a mask option to use 2 of the port pins as additional chip selects (polarity selectable in the mask, too). But that's hardly being able to put the ROM anywhere in the memory map. Looking at the circuit diagrams, the 8050 seems not to do this.The only chip select for the ROM section is the RS0/ pin.

25th Aug 2023, 6:24 pm	#82
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,086	Re: Commodore 8050 dual disk drive In which case it should be easier to read out the contents, should it prove necessary, than appeared at first glance. Does it actually need a clock for the purposes of reading out the ROM? ROMs are, by their very definition, static devices. How do we solve the mystery of the four flashes? Colin can quite easily test the 2114s, I think.

25th Aug 2023, 6:33 pm	#83
TonyDuell Dekatron Join Date: Jun 2015 Location: Biggin Hill, London, UK. Posts: 5,055	Re: Commodore 8050 dual disk drive The datasheet is not clear as to whether you need the phi2 clock for reading the ROM. It's shown as going into the chip select block, so it might be needed. [It's not going to apply here, but there most certainly are ROMs that need a clock input. Core-on-a-rope being the obvious one, along with the inductively coupled PCB ROM in an HP9100] At this stage don't worry about the 6530. The diagnostics must test the RAM first and I don't think that's happening. Test the 2114s and we'll go on from there.

25th Aug 2023, 6:45 pm	#84
ScottishColin Octode Join Date: May 2012 Location: Perth, Scotland Posts: 1,603	Re: Commodore 8050 dual disk drive OK. I'll make a start on the 2114s and report back - probably tomorrow.

26th Aug 2023, 8:50 pm	#85
ScottishColin Octode Join Date: May 2012 Location: Perth, Scotland Posts: 1,603	Re: Commodore 8050 dual disk drive So two of the 2114s were indeed duff. The left hand of the two drives now reads diskettes from 1984 which I do like a lot.the right hand drive doesn't spin up so there's still more to come but we've definitely moved forward today. Colin.

26th Aug 2023, 10:44 pm	#86
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,086	Re: Commodore 8050 dual disk drive Good show, Colin, and another good idea from Tony.

27th Aug 2023, 12:42 am	#87
Mark1960 Octode Join Date: Mar 2020 Location: Kitchener, Ontario, Canada Posts: 1,195	Re: Commodore 8050 dual disk drive I guess first thing to try is swapping the left and right drives and see if the problem follows the drive, or even the connecting cables.

27th Aug 2023, 5:14 am	#88
TonyDuell Dekatron Join Date: Jun 2015 Location: Biggin Hill, London, UK. Posts: 5,055	Re: Commodore 8050 dual disk drive These drive units have a pair of 'bare' mechanisms with a common 'analogue board', so there aren't any swappable cables. Which drive units do you have? The problem could be the spindle motor (which is a DC permanant magnet motor with brushes/commutator) or a fault on the analogue board.

27th Aug 2023, 12:59 pm	#90
TonyDuell Dekatron Join Date: Jun 2015 Location: Biggin Hill, London, UK. Posts: 5,055	Re: Commodore 8050 dual disk drive That looks like board covered by the circuit diagram on pages 12-14 of the .pdf service manual. Yes, your drives are Micropolis. Page 13 shows the spindle motor control circuit, there are 2 indentical circuits, one for each drive. It is not an elegant or well-designed circuit, alas. The spindle motor is a DC motor with an AC tachogenerator built in. The better designed spindle motor controllers take the frequency from that generator as the speed reference. This one simply half-wave rectifies it (CR20) and smooths it (C21). The voltage there is thus depends on the speed of the motor. It's compared against a reference voltage from the slider of R50 by the op-amp U12. Output of that drives the power darlington transistor Q1. The current though the motor is effectively sensed by R47. If it's too high then Q3 reduces the drive to Q1. Also Q1 can be forced off via CR26 if the output of U11b (pin 6) is low. The spindle motor itself has 4 wires and will connect via part of the molex connector on the appropriate edge of the PCB. Trace the wires to find it. Normally red and blue are the motor, yellow and green the tachogenerator. To start with, with the unit off, no disks in, unplug the motor from the PCB. Connect 9V or so DC (a PP3 battery should do it) between the red and blue wires, Red +ve, but if you get it wrong the spindle will simply run backwards. The motor should then run smoothly, the spindle should spin. If it doesn't, time to repair/replace the motor. If it does then the fault is on the PCB, which is not hard to repair.

27th Aug 2023, 1:02 pm	#91
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,086	Re: Commodore 8050 dual disk drive I would rather hope it is a fault on one channel of the 'analogue' board, since that is nearly all discrete electronics and quite fixable. Assuming the working drive to be drive 0 and the non-worker to be drive 1, and now that we know the units are Micropolis, you can try measuring the DC voltage between pins 1 and 4 of analogue PCB connector P5 when the non-worker should be running. You should see about 12V there under those circumstances. Also try measuring the resistance of resistor R47 (in-circuit, with power off).

27th Aug 2023, 1:24 pm	#92
TonyDuell Dekatron Join Date: Jun 2015 Location: Biggin Hill, London, UK. Posts: 5,055	Re: Commodore 8050 dual disk drive What I normally do is to test the motor on a 9V battery (or bench supply). If that works, lift one end of CR26 (for drive 1), connect the motor to the analogue PCB, and power up. With CR26 open, the motor should run _at the correct speed_ all the time. If not, then debug the control circuitry. If it does then work back through the simple bit of digital electronics shown on that page of the circuit diagram.

28th Aug 2023, 11:44 am	#93
ScottishColin Octode Join Date: May 2012 Location: Perth, Scotland Posts: 1,603	Re: Commodore 8050 dual disk drive Quick update - both drives out and visually inspected and they look ok. 9V DC applied to the wiring as in post 90 and both motors spin ok. Now to put it back together to try the next tests. Colin.

28th Aug 2023, 11:48 am	#94
TonyDuell Dekatron Join Date: Jun 2015 Location: Biggin Hill, London, UK. Posts: 5,055	Re: Commodore 8050 dual disk drive That's good news. Replacement motors are semi-hard to find, but the motor control circuit on the PCB is all standard components thst you can get.

28th Aug 2023, 8:04 pm	#96
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,086	Re: Commodore 8050 dual disk drive Assuming the applicable diagram to be on page 13 then it looks as though connector P5, rather than P8, is the one relevant to drive 1 (as opposed to drive 0). You should see ~12VDC between terminals 1 and 4 when the motor should be running, but since you have already proven the motor to be OK I think you will almost certainly not see that 12VDC on those pins. Follow up Tony's suggestion of temporarily unsoldering one end of diode CR26 and power the drive up with that diode disconnected. Does the motor run continuously with that diode disconnected?

28th Aug 2023, 9:57 pm	#97
ScottishColin Octode Join Date: May 2012 Location: Perth, Scotland Posts: 1,603	Re: Commodore 8050 dual disk drive Lifting CR26 makes the working drive (Drive 1) spin continuously at power on. Colin.

28th Aug 2023, 10:12 pm	#99
ScottishColin Octode Join Date: May 2012 Location: Perth, Scotland Posts: 1,603	Re: Commodore 8050 dual disk drive And even more interestingly, this page states that the PCB with 8050008 on it is actually schematic 8050005 with PCB assembly 8050006-1. https://www.amiga-stuff.com/hardware/cbmboards.html I don't think I could be any more confused right now. Colin.