CBM 3040 Repair

[Wreck]

So to start with, I am building a simple NOP tester as follows (6502)

Pin 33, 31, 29 to 21 (GND) and

Pin 32, 30, 28, 27, 26 to 8 (Vcc)

This will keep me busy for a while

 

[daver2]

The value of the decoupling capacitors are fairly irrelevant. As long as something is fitted - you are good to go.

I am not convinced a 2732 will work in here. A11 (on a 2732) is the same pin as /CE (on a 27128) and this could cause problems.

The ROM image is 1K ($000 to $3FF), a 2732 is 4K. What precautions did you use when you programmed the 1K image into the 4K EPROM? Did you program it four (4) times?

The 2732 is a 24 pin device whereas a 27128 is a 28 pin device.

How did you insert the 24 pin device into the 28 pin socket?

I would (personally) have use a 27128 EPROM and programmed the 1K image into the upper 1K of the 27128 (all of the 'unused' address lines on the 27128 (A10-A13) are all connected to +5V - so they are all HIGH).

I think there are so many variables here that you may have got one of them wrong - with the result that the adapter is not working as you expect.

Dave

 

[Wreck]

I am relieved regarding the capacitors. Regarding the 2732, I only programmed it once (so probably a mistake on my side), so I will try a 27128 (haven´t got any at the moment I think, so need to source). In the meantime I will try and figure out how to burn the 2732 four times; I haven´t used this particular software

I seated the 2732 according to the little picture on the PCB (essentially at the bottom).

I was able to "manufacture" the NOP tester and will try it out later today.

Joe

 

[Wreck]

Just out of curiosity; in a German service manual of the CBM the first troubleshooting step is to check the multiplexers (UD3/UE3/UF3) with the aid of a NOP tester. I just did and found UD3 PIN7(output) dead (no activity except 5V).

But I guess I am jumping around, I will start with the RRIOT first and see what is going on. Obviously I am open for any other inputs

Joe

 

[daver2]

>>> I seated the 2732 according to the little picture on the PCB (essentially at the bottom).

That is correct!

You can 'binary copy' the source image file four times to make a bigger file, then burn that into the 2732.

COPY /B infile+infile+infile+infile outfile

Where <infile> is the 1K file downloaded from Bo's website and <outfile> is the name of a 4K file that will be created.

EDIT: Ignore what I stated above about the 2732 not working. I re-looked at the website I was using and realised I was looking at the 2532 and not the 2732. The 2732 is OK.

Now I have the correct pinout of the 2732 (...) I have deduced that the 1K image can also be burnt into the top 1K of the 4K 2732 by telling the EPROM programmer to offset the binary file by 3K = 3072 (decimal) = C00 (hexadecimal). This means that the $000 to $3FF of the 1K image will occupy $C00 to $FFF of the 2732. The EPROM programmer can do this, you just have to sit down and read the manual. Alternatively, copying the 1K image file four times is usually easier...

Dave

 

[Wreck]

>>> I seated the 2732 according to the little picture on the PCB (essentially at the bottom).

That is correct!

You can 'binary copy' the source image file four times to make a bigger file, then burn that into the 2732.

COPY /B infile+infile+infile+infile outfile

Where <infile> is the 1K file downloaded from Bo's website and <outfile> is the name of a 4K file that will be created.

EDIT: Ignore what I stated above about the 2732 not working. I re-looked at the website I was using and realised I was looking at the 2532 and not the 2732. The 2732 is OK.

Now I have the correct pinout of the 2732 (...) I have deduced that the 1K image can also be burnt into the top 1K of the 4K 2732 by telling the EPROM programmer to offset the binary file by 3K = 3072 (decimal) = C00 (hexadecimal). This means that the $000 to $3FF of the 1K image will occupy $C00 to $FFF of the 2732. The EPROM programmer can do this, you just have to sit down and read the manual. Alternatively, copying the 1K image file four times is usually easier...

Dave

Thanks Dave! I will try this afternoon. I am using an old GalepIII with software running under WIN98.

Joe

 

[Wreck]

I was able to burn the 2732 from $C00 to $FFF, but no difference (actually it is worse now, as the drive motors won´t spin up at all). Also, 6502 PIN7(SYNC) is LO again...so back to square one. I will be getting a spare 901466-02 over the next few days, will see if this makes any difference.

In a nutshell; all LEDs illuminate, then both drive LEDs flash once but stay on.

This hints at faulty 6532s, or the MC3446 buffers if I am not completely wrong.

But first I need to figure out why SYNC is LO on the 6502, will check those sockets again...

Joe

 

[Wreck]

I was able to trace the machine back to his original owner; a certain Boris Eelzebub, it appears.
I put in the replacement 901466-02 today, no difference when turning on. Checked all sockets again, still have SYNC LO on the 6502. Whenever I type in any LOAD command and execute, the center LED comes on and stays on (both drive lights are permanently ON). Also, when turning the drive on the 6504 IRQ is HI until the 10-15 sec drive operation is done, then goes LO.
Also, I am unable to retrieve the DOS version through the CBM, no idea why it shouldn't work.
I asked myself today why were both 12V regulators failed when the machine arrived, and what else is hidden. In
any case the machine is beyond my capabilities to repair, as there appear to be various other issues making things complicated.

Joe