How to eliminate dust from circuit on AIM 65?

andyd · Post by **andyd** » Sat Apr 23, 2005 10:48 pm

Well, I think I am getting somewhere. I put together a NOP generator to troubleshoot the AIM 65 and it really helps... I removed the ROM, RAM, VIA and RIOT and I turned it on and started probing around with a logic probe. On the CPU all the address ines change state as would be expected. ON the ROM, RAM and RIOT the A4 line is indeterminate,(neither high nor low), I found I can move the problem around. I had a can of compressed air which I thought I would try to blow away the dust. Well, guess what, the problem moved to A5, so I blew more back to A4. It seems that there must be something, somewhere on the board that I can move around, but not eliminate.

Any ideas how I can clean the board to eliminate the, whatever that is causing the problem?

Any ideas much appreciated, I'm stumped

Thanks,

AndyD

leeeeee · Post by **leeeeee** » Sun Apr 24, 2005 12:17 am

Use a long bristled soft brush, like a paintbrush, and a vacuum cleaner. Don't use the vacuum cleaner brush as it can be charged with static and damage things.

Work with the board vertical or slightly tilted towards you, start from the top of the board and brush down towards the vacuum cleaner pipe. Pay particular attention to just under the edges of sockets and, if possible, always work with the socket pins in a vertical line.

If the sockets are the skeleton type then a blast of compressed air through the centre while holding the vacuum pipe nearby can help shift dirt from under them.

Lee.

andyd · Post by **andyd** » Sun Apr 24, 2005 10:23 pm

Well, AIM 65 still not working.

Bummer.

Thanks for all the help and encouragement.

I used the NOP generator (Thanks Lee) to trace down problems with the Address lines. I was able to get all the address lines working by using the vacuum and and paint brush to get whatever was causing the stuck line out (thanks again, Lee) Finally got all the address lines working as they should (as near as I can tell with a logic probe).

Then I put all the removed components back, and well, no joy.

I then decided to make sure one more time that the address lines were OK, so back to the NOP generator. All looked OK.

I then started reassembling a piece at a time, first the ROM, and RAM, and display, no joy. So I thoght, well maybe the program uses a timer in the RIOT or VIA, so I put them back in one at a time, no joy. put the last VIA in, the rest of the RAM and tried one last time, no joy.

Well, I'm out of ideas, I think I'll put it away for a couple of days, then maybe revisit it with fresh eyes.

If anyone has any other ideas I can try, I'm willing to try most anything (that doesn't involve solder).

Thanks for any and all help,

ANdy

leeeeee · Post by **leeeeee** » Mon Apr 25, 2005 1:58 am

Something you may try is check that all the device enables are active at some part of the address range. You can do this with the NOP generator in place, just logic probe the CE and OE pins on addressed devices to see if they're active, usually active low, for some part of the range.

Another thing to check is that RESET is working and does reset the processor.

Check that the data bus goes to all parts correctly.

Can't think of much more that doesn't require other diagnostic tools.

Lee.

andyd · Post by **andyd** » Mon Apr 25, 2005 3:14 am

Good idea, I did not check all the CE and OE, just some I'll check the rest with the NOP generator in use.

The RESET appears to be resetting properly, at least the RESET signal is getting to the processor, and the address lines look like they are interrupted when RESET is applied, so I think RESET is working. Maybe I should look at the RESET signal on the VIAs and RIOT as well.

I guess there could be data bus problems like the address bus problems I had. Thanks I'll try to find a good way to look into that.

Thanks a lot Lee, your help is much appreciated.

Andy

Nightmaretony · Post by **Nightmaretony** » Mon Apr 25, 2005 5:04 pm

If you have a flukje unit, you can always check it out with a 6502 pod.

For cleaning, I would have washed the board with hot water and dishwashing liquid then let it dry.

With the NOP generator, make sure to be checking out the enables for everything, they should be flashing.

since you are in sothern california, if you are anywhere near riverside county, drop me a line, bring it over, see if I can help out there....

GARTHWILSON · Post by **GARTHWILSON** » Mon Apr 25, 2005 8:04 pm

> would have washed the board with hot water and dishwashing liquid

Doesn't the AIM-65 have the printer built right onto the board? Can it be removed? Just don't wash the printer and wash out all the lube and get detergent or soap (which is usually corrosove) into the motors or other parts that you can't get it all out of. I don't know if that would include the display. (It has been over 20 years since I've seen an AIM-65.) Otherwise a ride in the dishwasher can clean up a board to some extent.

andyd · Post by **andyd** » Mon Apr 25, 2005 9:28 pm

Just got done with more checks. I put th NOP generator back in, checked all address lines everywhere, OK, phase 2 to everywhere OK, RESET to everywhere OK, All chip selects on ROM, RAM, VIA, RIOT, PIA and CPU, all OK.

Put minimum system back together, Display, ROM, and first 1K RAM, no joy.

Any other ideas?

Thanks,

Andy

Nightmaretony · Post by **Nightmaretony** » Mon Apr 25, 2005 10:10 pm

Are the roms ok?> I can program eproms if needed, 2716 or greater, we can test for that here.

andyd · Post by **andyd** » Tue Apr 26, 2005 9:23 pm

Thanks Tony!! My AIM 65 lives again!

Wow! I had it in the box for 20 years, took it out and it didn't work..

Here's what was wrong..

Address line conflicts -- found with NOP generator (thanks Lee) and logic probe.

5 of 8 2114 RAM bad -- Tony found with RAM test.
Bad CPU -- Tony found by swapping CPU.

Well, now it works again..

Thanks to all for encouragment, help and support, but most of all, thanks to Tony for his help and troubleshooting to find the main problems I'm not sure I would have got around to guessing bad CPU for awhile, since it seemed to be working.

Later,

AndyD