• Please review our updated Terms and Rules here

Cbm dual drive 3040 disk - two red lights

daver2 said:
I am now desperate :cry:!

Check all of the pins of UA3 then as the next best thing. Check with your oscilloscope (don't forget to vary the timebase) to see if the signal is permanently HIGH, permanently LOW or has some ACTIVITY and report back.

You can also check UH3 pins 5 through 14 (address lines A0 through A9) because they are actually wired to the pins of the UK3 adapter. Check the schematic - it is your friend! I think you can get access to these pins via pins 1-20 header of the RRIOT adapter.

If you can't measure them at one point, chase them around the PCB and measure them somewhere else...

Dave
Click to expand...
Here?
 

Attachments

  • Schermata 2022-08-14 alle 19.36.02.png
    Schermata 2022-08-14 alle 19.36.02.png
    214.8 KB · Views: 7
Yes, you can see A0 to A9 on the EPROM pins - perhaps the best place to check them out at. These pins are directly wired to the A0 to A9 pins of the 6504 CPU.

Dave
 
Measures on eprom s' adapter RRIOT with scope:

A0: HIGH
A1: HIGH
A2: HIGH
A3: HIGH
A4: HIGH
A5: HIGH
A6: HIGH
A7: HIGH
A8: HIGH
A9: HIGH
 
OK, so what I am thinking is to:

1. Remove the RRIOT adapter from UK3.

2. Make a NOP generator for the 6504 at UH3.

Obviously, this will require 2 off 28 pin sockets instead of 40 pin sockets (the 6504 is a 28-pin part)!

The data lines are on pins 18 to 25 of the 6504 (D7 to D0) so these will not have to be connected from the top socket of the sandwich (where the 6504 goes) to the bottom socket (where we plug into socket UH3).

The 6504 still executes the same instruction set as the 6502, so a NOP instruction is still $EA = 1110 1010 and the data lines of the top socket will need strapping as for a 6502 (but, obviously, taking into account the pinout differences of the data bus of the 6504).

If you plug the 6504 into the NOP generator, and then the NOP generator into socket UH3 - we should get activity on address lines A0 through A12 of the 6504 CPU (that you can measure with an oscilloscope in the same manner as the 6502).

All of the the outputs from UA3 (74LS42 pins 1 to 7 and 9) should now be activated in sequence (i.e. go LOW) for part of the NOP generator cycle.

This will give us confidence that the 6504 part (and UA3) are operating correctly - and we need to concentrate on the RRIOT ROM or the 6504 data bus.

We may also be able to do some checks of the 2114 SRAM and the address multiplexers?

Dave
 
Last edited:
daver2 said:
OK, so what I am thinking is to:

1. Remove the RRIOT adapter from UK3.

2. Make a NOP generator for the 6504 at UH3.

Obviously, this will require 2 off 28 pin sockets instead of 40 pin sockets (the 6504 is a 28-pin part)!

The data lines are on pins 18 to 25 of the 6504 (D7 to D0) so these will not have to be connected from the top socket of the sandwich (where the 6504 goes) to the bottom socket (where we plug into socket UH3).

The 6504 still executes the same instruction set as the 6502, so a NOP instruction is still $EA = 1110 1010 and the data lines of the top socket will need strapping as for a 6502 (but, obviously, taking into account the pinout differences of the data bus of the 6504).

If you plug the 6504 into the NOP generator, and then the NOP generator into socket UH3 - we should get activity on address lines A0 through A12 of the 6504 CPU (that you can measure with an oscilloscope in the same manner as the 6502).

All of the the outputs from UA3 (74LS42 pins 1 to 7 and 9) should now be activated in sequence (i.e. go LOW) for part of the NOP generator cycle.

This will give us confidence that the 6504 part (and UA3) are operating correctly - and we need to concentrate on the RRIOT ROM or the 6504 data bus.

We may also be able to do some checks of the 2114 SRAM and the address multiplexers?

Dave
Click to expand...
Hi Dave, i am desperate now!
i don't think i can make a nop for the 6504 !!!! :(
 
Desperado: Attached, a rough sketch for a NOP generator for 6504. Not very difficult to make.

People often make NOP generators just with wires taking the data pins to +5V or GND but I prefer to use low value resistors for the connections to GND and +Ve in case the CPU somehow gets switched into output mode - like maybe, I make an error in the construction and the CPU starts trying to execute write instructions instead of NOPs. If that happens the CPU will not be happy about having its data pins shorted to GND or hard +5V so the resistors are there as a safety feature for the CPU. The resistor value is not critical but I usually say in the low hundreds of ohms, not more than 470 ohms.

You can make this NOP generator using a single normal 28-pin socket, bend pins 18-25 straight out horizontally the same way you did when you made the reader adaptors, making sure that the pins are not touching the pins of the UH3 socket below and then connect the bent-out pins using resistors to +5V / GND as shown. If you want to be sure that pins 18-25 are not connecting to the UH3 socket, get another 28 pin socket and completely remove the metal contacts from pins 18-25, plug that into the UH3 socket and then plug this adaptor on top.

6504_NopGen.jpg
 
SiriusHardware said:
Desperado: Attached, a rough sketch for a NOP generator for 6504. Not very difficult to make.

People often make NOP generators just with wires taking the data pins to +5V or GND but I prefer to use low value resistors for the connections to GND and +Ve in case the CPU somehow gets switched into output mode - like maybe, I make an error in the construction and the CPU starts trying to execute write instructions instead of NOPs. If that happens the CPU will not be happy about having its data pins shorted to GND or hard +5V so the resistors are there as a safety feature for the CPU. The resistor value is not critical but I usually say in the low hundreds of ohms, not more than 470 ohms.

You can make this NOP generator using a single normal 28-pin socket, bend pins 18-25 straight out horizontally the same way you did when you made the reader adaptors, making sure that the pins are not touching the pins of the UH3 socket below and then connect the bent-out pins using resistors to +5V / GND as shown. If you want to be sure that pins 18-25 are not connecting to the UH3 socket, get another 28 pin socket and completely remove the metal contacts from pins 18-25, plug that into the UH3 socket and then plug this adaptor on top.

View attachment 1244879
Click to expand...
Wow!! Thanks so much Sirius, can i try to made it, but i don't have these resistors...can i use only piece of wires please?
 
What resistors HAVE you got that are close?

Technically, you would only require two (2) resistors - one pull-up to +5V (the red ones) and the other a pull-down to 0V (the black ones).

I wouldn't use just pieces of wire - even though this should (in theory) work. The data bus is being used to fetch (read) instructions only (no writing). However, if you screw up the wiring - all bets are off.

Think of it this way: this is the only 6504 you have - so what happens if you damage it? Using the resistors makes this scenario much less likely than not using the resistors...

Dave
 
The individual resistors are there so that if the data pins do somehow get into output mode and start outputting opposing states they will not fight with each other, which could be equally damaging if several are connected to the same end of a single resistor.

If the NOP gen is made right first time then it should be possible to use just wires, agreed? But I would not do it myself.
 
Last edited:
SiriusHardware said:
The individual resistors are so that if the data pins do somehow get into output mode and start outputting opposing states they will not fight with each other, which could be equally damaging if several are connected to the same end of a single resistor.

If the NOP gen is made right first time then it should be possible to use just wires, agreed? But I would not do it myself.
Click to expand...
My 6502 Nop doesn't have any resistors...only piece of wire....
 
You must log in or register to reply here.
Back
Top