• Please review our updated Terms and Rules here

KIM-1 Repairs continue

glitch said:
You leave it off in some cases if/when you expand the system. IIRC the KIM-2 RAM board I had required you to remove it.
Click to expand...

Hi glitch,
Yes, but typically a better way of doing this would have been to design it such that an open circuit at this pin K would allow internal operation and then a ground would be required to allow for expansion. This could be accomplished with an inverter and a pull-up resistor on the DECEN signal.
-Dave
 
dave_m said:
Hi glitch,
Yes, but typically a better way of doing this would have been to design it such that an open circuit at this pin K would allow internal operation and then a ground would be required to allow for expansion. This could be accomplished with an inverter and a pull-up resistor on the DECEN signal.
-Dave
Click to expand...

Of course, but this is old computers :) I'm getting ready to test a S-100 clock board that expects you to jumper whatever AC voltage is ahead of the +/-16V rectifiers onto the bus! "Sane defaults..." right?
 
falter said:
Darn.. couldn't be that easy. I replaced the broken 1k resistor but yeah, no change. Darn!
Click to expand...

Good try! That resistor was only used as a pull-up to +5V for one of the chip selects. If the resistor was open, the circuit would probable still work although noise could cause problems.

It's looking more and more like a bad 6530, but let's make sure. Hook up the USB scope and see if SYNC is pulsing. If it is, the CPU is hard at work fetching instructions. You can find SYNC at U26-pin 5.

I just noticed that the output of U26 is a signal called SST for Single Step. It will cause a non-maskable interrupt if switch K25 is closed. Is there a Single Step key on the KIM-1? Make sure it is not stuck by seeing that there is NO pulsing on the signal NMI. That would cause problems if it is not steady at +5V. NMI can be found at the CPU U1-pin 6 or at the P1 connector pin 6.
 
glitch said:
Of course, but this is old computers :) I'm getting ready to test a S-100 clock board that expects you to jumper whatever AC voltage is ahead of the +/-16V rectifiers onto the bus! "Sane defaults..." right?
Click to expand...

For use as a source of 60 Hz clock? I hope it is routed to an analog circuit to be attenuated and conditioned to +5V square wave ;). You are right. Old computers can be fun!
 
debug circuit.
I added a couple extra parts to have an indicator that can
change from code.

Oops I see an error in the schematic.
The bottom clock should be P1-Y, not P1-U.
A typical copy paste error!!! Argggg!
Dwight
 

Attachments

  • Debug.jpg
    Debug.jpg
    68.4 KB · Views: 2
Last edited:
I've purchased a pile of parts to make one of these diagnosers.
I have a number of 2764s so that is not an issue.
The .100 hole proto board was the most expensive part.
The logic could easily be put in a 16R4 but I'm too lazy
to deal with it.
Dwight
 
Hi Dave
It'll be a week or two. I have to wire board and drag my
KIM-1 out of storage. I've got to find my sock of dual
22 pin connectors. I hope I stored them in my 6502 box.

I would think the first select should be a supper simple program.
It should just toggle the light. It should use no RAM.
I've got to review the 6502 instructions to see if I can make a
delay with the CPU only. It has been a while and I know the 6502
likes to make a lot of use of the first part of RAM.
I know I can make it toggle the light from ROM code but
don't know about a delay, to make it something like every
0.5 seconds. I'm not sure if the internal register effect flags
on overflow.
Dwight

Ok, I just did a check. INX and INY effect both carry and zero
so this should be easy code.
Dwight
 
Last edited:
Dwight,
Yes, there are a lot of instructions that use 'zero page' ram (first 256 locations), but that is only for speed as the implied upper address keeps computer cycles to a minimum.

Besides the diagnostic code, perhaps a block of the 2764 can contain the boot code. If selected, the KIM-1 can run even with bad ROM if a pin on the 6530 is grounded. Does anyone have the KIM-1 binary image?
 
My thinking is the first code should be to just check that the bus is working
and the processor is running.
My next level I'd think would be to check the RAM ( a common failure point ).
Then do a CRC on the ROMs.
If the code is good there, one can use the code in the ROMs to run LEDs
and keyboard.
Replacing the ROM code is a little more tricky. It can be done but requires
more decoding than I'd planned.
As it is, I only decode one 8K block for read. I also have no option to decode
addresses at the locations of the ROMs in the two 6530's.
We can talk about doing this once I make progress with basic functions.
It is clearly possible with DECEN to mask anything on the KIM.
I only have to one masking now to take over the boot. I only use
it long enough to jump to a clear space in memory. I then disable
that masking so I can read and CRC the ROM that would normally be used
for boot.
Dwight
 
I went looking for my KIM-1 in my storage. So far, it is still hiding
someplace.
I'll need to look later for it.
I did find an interesting object or two that I'd forgot I had.
I found an SDK-51 and SDK-2920, as well as 2 SDK-85s.
Both SDK-85s were built into briefcases but done differently.
I found a VideoBrain that I thought I'd lost years ago.
I did find two dual 22 pin connectors so I can start on the project.
There is still depths that haven't been explored.
Worst come to worst, I'll borrow a KIM-1 from someone.
Dwight
 
Crystals tend to get knocked around a bit and break internally. Have you checked for a valid clock? The lack of a clock will cause the symptom you have. Use your logic probe to check for pulses on pin 10 of the 7404.
 
Inconsistent result for U26. First time I powered on I had a solid LO, no pulse. I hit reset, it began pulsing. I shut off and tried again.. this time it stayed LO. Tone only changed briefly after releasing ST. Turned on again.. had pulse.. hit reset.. no pulse. Hit reset again.. pulse.
 
You must log in or register to reply here.
Back
Top