[vcf-midatlantic] Schematic work needed

alan at alanlee.org alan at alanlee.org
Wed Aug 29 11:44:36 EDT 2018


It's two 8 bit opposing latches and some address decoding glue.  If one knew the port address, you could create the schematic blind and be 98% correct.
Sent from my Verizon LG Smartphone
------ Original message------From: Kyle Owen via vcf-midatlanticDate: Wed, Aug 29, 2018 11:17 AMTo: vcf-midatlantic;Cc: Kyle Owen;Subject:Re: [vcf-midatlantic] Schematic work needed
On Wed, Aug 29, 2018 at 10:05 AM RETRO Innovations via vcf-midatlantic <
vcf-midatlantic at lists.vcfed.org> wrote:

>
> Many of the signals go under the ICs.  Sometimes, you can hold to a
> light and see through the PCB and validate where they go, but other
> times, I have missed wires connecting to more than 1 pin of the IC under
> the IC.  Ohming, if applied verbosely for each signal around the IC
> compared to each IC pin, should find all of them, but I don't think I
> would promise 100% perfection without spinning a PCB and testing the
> completed unit.
>
> Maybe everyone else is better at ohming out boards and the hidden
> signals under ICs never bother anyone else, but I've gotten hit twice by
> them on two different efforts.
>

Yes. If you do not completely ohm out each pin/component, I could very much
imagine generating an incomplete schematic. The netlist generated by ohming
out the board should have every pin on every component by the time you're
done. If not, you've missed something. With some conditional formatting,
it's possible to highlight possible errors like duplicate entries, too many
pins per component, etc.

Hidden traces are no problem *if* you are not relying on visuals, but
rather the ohmmeter.

And if you're using a Fluke, don't forget to hold down the yellow button
when turning it on, lest you have it automatically power off while you
attempt to ohm out the board. Ask me how I know... :)

Kyle


More information about the vcf-midatlantic mailing list