Amiga A1200 Repair

[sutol]

Hi, I'm looking for a bit of advice/assistance.


I am trying to repair an A1200.
I was told that the Internal floppy was intermitent, but I can't get either the Internal or the external floppy to read. I have changed drives without success it simply says DH0:????.


The aim is simply to get a single disk to work again (Kwizoke Karaoke), so I either need to get the drive working again, or get the disk converted to an adf and load it into a Gotek drive, or on a hard disk.


I've run out of ideas, so any suggestions would be greatly appreciated
I am in the Cambridge UK area



Thank you

[cpiac64]

df1 without df0 work?

[sutol]

no, but Gotek on DF0 works

I have an internal SD card hardisk too, so I thought I would go through all combinations

DH0 | DF0 | DF1 | Gotek | Result
connected|connected|unplugged|unplugged|DF0:????
connected|connected|connected|unplugged|DF0:???? DF1:????
connected|unplugged|connected|unplugged|DF0:???? DF1:????
connected|unplugged|connected|connected|DF1:????
connected|unplugged|unplugged|connected|Working
connected|unplugged|unplugged|unplugged|DF0:????
unplugged|connected|unplugged|unplugged|No Boot
unplugged|unplugged|connected|unplugged|No Boot
unplugged|connected|connected|unplugged|No Boot
unplugged|unplugged|connected|connected|DF1:????

Floppy drives both had bootable disks, there is no combination that gives me a way of accessing the floppy.

Why DF0:???? even if the drive is not connected, is that normal?

[shuttlepilot]

Check your cables for faults or breaks and also put a multi-meter on the PSU to see if you are getting good and consistent voltages.

Take this opportunity to have a good look at the mainboard itself and check for any leaking or bulging capacitors

[sutol]

Thank you, I have check with a DMM, I'll have a look with an oscilloscope too in case they are noisy.

I have changed the floppy cable and had a good look at the solder joints.

[solarmon]

I have a lot of questions!

In your test table, I don't understand how the Gotek is connected, compared to DF0 or DF1 (or something else) - since you have the Gotek as a separate column.

For the external drive connections (DF1+) - how are these connected, using what adapters/enclosure, etc?

Have the physical drives been cleaned and have they been tested to be working on another Amiga?

What are the physical floppy drive models - are they Amiga or PC (converted?) drives?

The drive select setting on each drive (including the Gotek) matters, depending on what drive controller (whether internal or external) - so it is important to know exactly what each one is before we can start troubleshooting further and ruling things out.

If DF0 is working with a Gotek and not a normal drive, then that sounds that the normal drive is either not configured correctly for DF0 - i.e. the DS0 jumper should be set - or the floppy drive has some other issue with it.

If you can get DF0 to boot using a Gotek then I would suggest you boot ATK (Amiga Test Kit) to be able to perform some drive signal tests. Once ATK booted you then test DF0, DF1, etc - swap out the Gotek for a normal drive to test the drive in DF0.

https://github.com/keirf/Amiga-Stuff

[sutol]

Thank you solarmon for you list of questions, I will give you the full story!

The Amiga A1200 is owned by someone I know from the local pub, he uses it as part of his karaoke setup.
He told me that every now and then it fails to boot from the disk.
As I am an Electronics engineer he asked me if i could look into it for him. I have little knowledge of Amigas though
The machine is a standard A1200 with a CD drive attached to the IDE port (Not connected at present)

When I looked at the machine it was filthy, it had spent its entire life in pubs around the area!
My initial tests showed the floppy drive (Chinon FZ-354) was working ok but as it was reported as intermittant I thought that a clean would restore it to good condition.
I stripped the drive and cleaned it with isoprohenol, whilst I had the case apart, I inspected the pcb, caps etc and deep cleaned everything.
I reassembled it and it was working so I thought that was the job done. however the next day I found it failed to boot again.
I suspected the drive was worn out as it was so full of dirt and dust. I purchased a couple of Pc drives (Sony MPF920) and modified them for Amiga use.
I put one in the amiga, and intended keeping one as a spare. Again everything was working and I thought my work done.
The next day the machine again failed to boot.
My thought now was the floppy disk itself. I decided to swap the drive for a gotek, as just copying the disk would cause the new disk to fail again in the future.
but I had no way to either copy the disk or convert it to an adf.
So I bought a sdcard hard disk kit a dual IDE interface, gotek drive and an external floppy drive (big hammer for the problem, to give multiple solutions!)

However I have got to the point that I cannot get any floppy drives to work at all now!

In my test table the Floppy and the gotek are connected (not at the same time) to DH0 via a new ribbon, DF1 is an external amitek drive
Supplies to the drives look about right (12v & 5v)
The drives dont seem to spin very fast, but I understand that Amiga drives do run slower than pc floppies, so maybe that is ok
The gotek drive works great and I would like the adf file on a gotek to be the final solution (or run from hard disk.)

The Chinon drive has DS0 shorted and DS1,DS2 & DS3 unshorted, plus MM, RDY and TTL/CMOS shorted
The Sony drives have JC30 (SEL0) shorted and JC31 (SEL1) unshorted, plus pin 34 rerouted.
The Gotek drive has S0 shorted and S1 unshorted.

ATK
DF0 Id:00000000 (Present)
DF1 Id:FFFFFFFF (Present)
DF2 & DF3 Not present

Chinon DF0:Signal test - does sensible things
Chinon DF0:Read test - Cannot read track 0
Chinon DF0:Write test - OK
Chinon DF0: Head calibration test - Cyl 0: Side 0 0/11 okay, Side 1 11/11 okay
Cyl 40: Side 0 11/11 okay, Side 1 11/11 okay
Cyl 79: Side 0 11/11 okay, Side 1 11/11 okay

Both Sony DF0:Signal test - does sensible things
Both Sony DF0:Read test - Cannot read track 0
Both Sony DF0:Write test - OK
Both Sony DF0: Head calibration test - Cyl 0: Side 0 0/11 okay, Side 1 11/11 okay
Cyl 40: Side 0 11/11 okay, Side 1 11/11 okay
Cyl 79: Side 0 11/11 okay, Side 1 11/11 okay

DF1:Read test - Ready too fast,cannot read track 0
DF1:Write test - Ready too fast,cannot write track 158
DF1: Head calibration test - Cyl 0: Side 0 0/11 okay, Side 1 0/11 okay
Cyl 40: Side 0 0/11 okay, Side 1 0/11 okay
Cyl 79: Side 0 0/11 okay, Side 1 0/11 okay

[solarmon]

Hi sutol,

Thanks for the detailed explanation, it is getting much clearer now.

Very interesting story/history of this Amiga A1200.

I'm going to assume it hasn't been recapped? I wouldn't say that would be the cause of such an issue.

Can you perform those same ATK disk tests with the Gotek and see what you get. I know it should be good, but it is nice to have a complete comparison.

When you perform the floppy disk signals tests you should have a disk inserted. The important signals tests for me are:

1. If you have a disk inserted, set Motor to On, you should get index pulses incrementing and also the RDY signal should appear to indicate that the motor is spinning at an acceptable speed and that it is ready for reading/writing.
2. If you keep pressing "F6: Step" option it should step in and out of Track 0, so the TK0 signal should appear and disappear accordingly.

If there is no Index pulse and subsequently the RDY signal, and it cannot find TRK0 then that explains why it your floppy drives are not working. However, your Gotek floppy drive emulator is working so that doesn't completely add up.

The Gotek doesn't use/require the motor signal (since it doesn't have one) but all the other floppy drives and functions should still be work the same.

Out of your test results, the main thing that stands out is that it cannot find Track 0, so we need to figure out why this is the case. The floppy drive signals on the Gotek drive should show whether that signal is working or not.

Get the Kwizoke Karaoke floppy disk converted to ADF should not be an issue, if you have the right tools. Worse case, you can send it to me and I can convert it for you. But it would be great if we can solve this mystery!

[sutol]

Hi solarmon,
No it hasn't beek recapped, but there is no sign of damaged caps/swelling/leaking. aged caps still possible I know, but I dont think there are any electrolytics in the floppy interface so I didn't think that was a concern, I was just going to say it will need doing in the future.

The ATK test show that my external floppy drive is broken (Gotek doesn't work in DF1, so the interface has probably died), so I am ignoring that, I did take the drive out and put it into DF0 with the same results
The gotek drive works on ATK DF0

I think we might be getting somewhere (Learning fast!)
Now I can see how ATK works (using the Gotek) I can test with a real floppy drive a little better:
1. TK0 Toggles with STEP, Head is moving as expected
2. The motor doesn't spin up with MOTOR-ON/OFF,
3. I get TK0 but not TK0 RDY, MTROn-RDY1 = 0uS, MTROff-RDY1 = 0uS
4. No Index pulses
5. I get CHG when disk changed, motor spins for a second then stops

However when I do a read test the motor spins, I get ready in good time(161.6ms), Ready signal is oscillating, (hacked PC drive?) cannot read track 0
Write test also spins motor and passes
Head cal also spins motor

What tools do you recommend to convert the disk to ADF (assuming I get a drive to work!)

[solarmon]

Hi Sutol,

Yes, using the Gotek to test is good as you can see the track number as you step through. However, it doesn't use the motor signal so you cannot test for this using the Gotek.

If the motor is not starting then that would be a problem. However, you mentioned that it doesn't work in the signal tests, but does work when doing a read/calibration test - so this is a bit confusing.

It looks like there is something up with the Motor signal.

I would recommend getting a logic probe to easily test the signals on the wire itself. I remember having a similar issue, but I need to try to think back on what it was and whether I resolved it. My memory is failing me at the moment!

For converting the disk to ADF. You can do this on the Amiga, but you need a working drive - so that rules that out for now!

The other option is a using a Greasweazle on a PC to do a disk dump. Assuming that the floppy disk was clean with no errors, then the image can be converted to ADF format.

https://github.com/keirf/Greaseweazle/wiki
https://www.facebook.com/groups/greaseweazle

The Github prokject page has details on how you can build a cheap one using a Blue Pill board. Or you can request to be on the waiting list for a fully assembled proper board unit, or you can request a kit which might be more suitable for you. Request in this post:

https://www.facebook.com/groups/grea...15410935211174

Alternatively, send me the floppy disk and I will clean it (assuming there is no physical errors) and convert it for you. PM me if you are interested, or if you need any help with setting up Greaseweazle.

[sutol]

OK I'll dive into the signals. I have found some schematics, I dont know if they are the right revision, so I'll try to stick to the floppy drive connector as that is likely to be the same for all revisions.
I assume I need to look at (and compare with Gotek drive)
_MTR0 on Pins 16 and 4
_TRK0 on Pin 26
_RDY on Pin 34
Do _SEL0 (Pin 10) and _SEL1(Pin 12) act as a drive select?
Anything else I should be looking at? _DKRD, _SIDE (All errors appear to be on side 0)
Should I expect TTL signal levels?

[solarmon]

I've also mapped the floppy drive connections to help me troubleshoot such issues - so you might find this useful:

https://sites.google.com/one-n.co.uk...ve-connections

Check the signals at the motherboard connector and at the floppy drive side too.

Yes, _SEL01 and _SEL1 are the drive select. For the internal drive it should be _SEL0 (and the floppy drive/Gotek should be jumpered for S0).

I believe the Amiga mainly uses TTL signal levels.

[sutol]

This is getting frustrating now!

Using ATK When I perform the read test I get the waveforms in the attached ReadTest.png, which looks ok
When I press F5 to go into the signal Test I get the waveforms in the attached EntryToSignalTest.png
However, When I press F5 in Signal test to switch on the motors the MTR0 line does not go low! The step line in OK when I press F6

Can you have a look at the plots and see if you can see anything I'm missing

Thank you

[sutol]

Should _SEL0 always be low, even if I select DF1, DF2 or DF3?

a clue?

[sutol]

I retested on the Gotek Drive and SEL0 is normally high, and goes low just befre INUSE, so thats a difference, but where is the fault.
I'll check the SEL0 line on the drive isn't shorted to GND, but on 4 drives?

Attached are the Gotek plots for reference

[cpiac64]

check U5 and U8

[sutol]

Hi Cpiac64,
Can you advise what to check, especially given that DF0 works with a Gotek drive, but not with a floppy drive

[solarmon]

@sutol,

The first physical drive test - which drive was it? The Amiga Chinon or the PC converted Sony?

When you are doing the signal testing, I assume you have a floppy disk in the drive?

Are you using the same power and data cable when testing physical drive and when testing the Gotek?

When the _MTR0 signal goes low, the floppy LED on the A1200 should light up too - do you see this happening?

Where on the Amiga/drive are you capturing these signals?

[sutol]

The drive was a converted sony (The one I proved worked during my initial investigations.

There is a floppy in the drive (and CHNG is working). I am capturing using a Salae logic analyser. It is the same power and data cable. I have grafted a header onto the connector of the drives to connect the logic analyser. Now I have identified that SEL0 doesn't appear to go high I need to probe on the main pcb to prove the connection is getting through (even though it does for the Gotek, it might be intermitant

I have verified that the F.Disk LED is switched on when the MTR0 line is active (During the read test, and entry to the signal test, but not when F5 is used to switch the motor on) The SEL0 line is always low and SEL1 goes low briefly during entry to the signal test

Do you have details of U5 Gayle, I cant find it (is it on the underside?) I thought the SEL0 was used to clock the flip-flop to latch the MTR signal, but that doesn't stack up in my plots, unless the signals are ok on U5

[sutol]

found U5, I didn't expect it to be so far away!