• Please review our updated Terms and Rules here

KIM-1 replica not working

C

Chrille

New Member
Joined
Dec 26, 2024
Messages
6
Hi,

I have just completed building a new KIM-1 Rev.D replica board. This is a project I have been wanting to build for quite some time and now it’s finally finished (at least the soldering part) :).

I do however have some issues getting it to work. The problem is that I get no output on the display even after hitting the reset button on the keypad.

I have connected pin ”K” to ground to enable internal address decoding since I am not using anything other than the KIM-1 itself right now.

I have inspected and tested every TTL IC and RAM IC and they all work fine so they should not be the cause of my dead KIM-1. I have also verified that I have a working 1 MHz clock signal on the board using an oscilloscope, so the only IC's that have not been tested are the CPU and the 6530-replacement board I am using. The TTL IC’s on the 6530 replacement board have also been tested and I have tested two pairs of different 6532 IC's.

I know that the MAN72A displays are working since they are NOS parts and I verified all of them on a breadboard before soldering them to the board.

The power consumption of the board is about 0.8A with the dual 6532 -> 6530 replacement board connected.

I also have a question regarding resistors R18 - R23. According to the KIM-1 user manual they are supposed to be 220Ω resistors (which is what I used in my build), but I have noticed that later variants of the KIM-1 (Rev. D and newer) uses 100Ω resistors instead. Why did they change the resistor value and does it matter if they are 100Ω or 220Ω?

I would really appreciate any advice that would lead me to a functional KIM-1. I include an image of my build to make it easier for any of you KIM-1 experts to check if there is any obvious fault on my board.

Thanks!

KIM-1 replica.jpg
 
You might want to build the KIM1 diagnostic board as described at: http://retro.hansotten.nl/6502-sbc/kim-1-manuals-and-software/kim-1-diagnostic-board/
It will check out most of the typical problems on a KIM-1. It has relatively few parts required and could be done on a wire wrap board.
As well as the diagnostics, there are a few parts of the EPROM used to load the EEPROMs I used on my design for the 6532 replacement circuits. They will not do anything with your replacements but are not part of the diagnostics that are selected by the switches on the diagnostic board.
I believe Hans has all the information needed to use the diagnostics. The RAM test has one thing not mentioned in the instructions ( I'd not thought about it when I wrote the instructions ). When the address decoder fails, the RAMs will show alternating fail/pass indicators. This is because the test halts on the first failure. Since it uses a pattern of 55/AA hex, with no address decoder function, it will fail on the first pattern used ( I forget which, 55 or AA). Just be aware that it is more likely the address decoder or the data buffers used with the RAMs, when it indicates alternating RAMs failing.
A RAM test is a simple test as it has to run without depending on any RAM working and runs only on the internal registers of the 6502. A rather limited condition.
The test are sequenced such that each test must work in sequence before moving to the next test. If the bus is dead, it will fail the CPU test. This would need to be fixed before moving on. Since your board looks to be fully socketed, removing parts that could block a particular diag step is a good way to isolate bus issues.
Dwight
 
I should also note that the CPU test is a very minimal test and if the CPU has more than a simple problem it may pass the CPU test but fail later test.
Dwight
 
Welcome to VCFED (as this is your first post).

I would proceed as follows (all pins referenced on the 6502 CPU):

Check that pin 40 goes LOW when you operate the reset button and then goes high again when you release the reset button.

Check that pins 2, 4 and 6 are all HIGH.

Check that pin 7 is pulsing (using a logic probe or - better still - an oscilloscope).

Report your findings.

You will be under moderation for your first 10 posts, so let's take it slowly until the moderators are happy with you...

Dave
 
Thanks Dave!

Great advice, I will check the pins you refer to using my oscilloscope tomorrow.
 
Dwight,

Thanks for your input. I have read about your diagnostic board on Hans site. It looks like a very nice and useful tool indeed. I will build one of those if I can't get the board to work soon. I don't have tools to wire wrap a board but I will build one using a manufactured PCB instead if I need one.
 
Last edited:
I can't see how the resistor values will affect the circuit much (if anything).

They act to switch off the transistor connected to the anode of the display, thus blanking it.

A lower resistor value will switch the transistor off 'more' than the current 220 Ohm resistor.

Dave
 
Great news!

I finally found the cause of the non-working KIM-1. It turned out to be a ROM-related issue. I used a 27C512 EEPROM instead of a 27C64 with the dual 6530 adapter board. The two IC's are more or less pin-compatible (at least the data and adress lines up to the extra address pins available on the 27C512 so I just grabbed one of those when I burned the ROM image before. The problem is that the adapter board ties adress pin 14 to high (connected to VCC) since it is designed to use the 27C64 EPROM that does not have a A14 pin and because of that the address mapping does not work without some kind of adapter board on top of the adapter board. So I had to dig deep in my IC inventory and found an unused 27C64 and when I burned the ROM image to that IC and plugged it in it booted up right away.

Now I just have to have some fun with my new KIM-1! :)

Thanks everyone!

KIM-1 working.jpg
 
No problem.

You could also have either:

Burnt the image into the correct location in the larger device.

Burnt multiple copies of the same image into the larger device.

Either would also have worked.

Enjoy!

Dave
 
Yeah, you are right. I could have just burned the data to another location on the 27C512 but since I had the correct IC in my stash I might as well use it.
 
Thank you Hugo!

I am pleased at how it turned out in the end and I did put a lot o effort in it since I wanted it to look good.

When I have some time over I will probably design and 3D print an enclosure for the keypad to make it look less "naked" and more like the original keypad that had a black plastic enclosure and bigger keys that are easier to use.
 
You must log in or register to reply here.
Back
Top