Page 3 of 5

Re: board testing

Posted: 04 May 2019, 00:08
by gillandro
the part that both boards are doing the same is that when the board is plugged in led5-8 stay fully lit

Re: board testing

Posted: 04 May 2019, 00:35
by marcelk
The LEDs should run their back-and-forth pattern without stopping, if that answers your original question.

Re: board testing

Posted: 04 May 2019, 00:55
by gillandro
yes it does thanks...the spread sheet prob doesnt need make this clear but i assume i measure the chip pins from the soldered side? or from the front?

Re: board testing

Posted: 04 May 2019, 01:03
by marcelk
There are photos in the spreadsheet.

Re: board testing

Posted: 04 May 2019, 01:09
by gillandro
thank you....i will test all the chips till i get the desired readings from each and see what happens.

Re: board testing

Posted: 04 May 2019, 01:28
by marcelk
The sheet is the best we can do to assist without using an oscilloscope. It also helps to inspect each joint carefully visually with a loupe, and re-address those that look a bit off. From our experience, the pins for the EPROM and RAM need the same kind of attention as ground pins, because the different thermal behaviour. In the hobby realm, it isn't a hard science, but with some care and patience it will always succeed.

Coincidently, there was a really nice article on Hackaday today on the hard science of soldering: https://hackaday.com/2019/05/03/get-to- ... ng-of-ics/
One of the commenters kindly links to the NASA pictorial guide that shows what it takes to bring solder joints to Mars and out of the solar system: https://workmanship.nasa.gov/lib/insp/2 ... meset.html (section 6 for through-hole soldering).

Re: board testing

Posted: 12 May 2019, 00:09
by gillandro
readings should be in the case of pin one of u1 0.700v? and pins 39 and 40 of eprom common?

Re: board testing

Posted: 12 May 2019, 01:34
by gillandro
i found u3 and u4 having 0.435v on a pin where it should have read 0.700v or so

Re: board testing

Posted: 12 May 2019, 18:41
by marcelk
If that's just pin 3,4,5,6 for U3 and U4 (the two low PC counters) I think that's still ok. These inputs are connected directly to the data bus. I just checked and I have one board where these measure ~500 while the same pins on U5 and U6 (the high PC counters) give me ~800.

It just occurred to me that the diode mode pin check is blind for unintended shorts between adjacent pins. So this is a thing you can check separately by measuring resistance between pins. You mentioned that with some cleaning things improved to the point where the LEDs were briefly running. When LEDs are running, everything is ok, because these LEDs are driven by the software. This temporary improvement can be the effect of residual solder flux removal like The 8-Bit Guy showed in his review of the PE6502 computer kit.

[Edit: attached an updated sheet where I mark, in green, all pins connected to the data bus. My guess is that we can see a lower value there because all these are also connected to the RAM chip. Perhaps I had a different RAM chip when we originally made the table.]

Re: board testing

Posted: 12 May 2019, 19:49
by gillandro
so according to the new chart those chips are good then because i was pulling readings exactly like those on those chips so far....thanks