[Coco] Zap 0.0.4 Beta has been uploaded to my drop box

[Gene Heskett]

On Tuesday 31 March 2015 21:17:01 K. Pruitt wrote:
> There seems to be a blind spot covering portions of at least 3 of the
> house codes when using the RS232 pak.
>
> At first I thought it was a logic error in my code. Turns out it
> wasn't. Though I introduced a few trying to fix this problem.
>
> I swapped out the CM17A and the TM751 with others and still I have
> these blind spots, so it's not my X10 modules. And I don't have these
> blind spots with my PC's serial port using the same modules.
>
> This could be an RS232 pak issue. It should have sufficient power and
> it operates most of the house codes, both upper and lower banks of
> unit codes. I've eliminated all other potential sources of the problem
> (that I can think of) and yet I still have these blind spots.  I don't
> have a second RS232 pak to test it out on.  Though I don't expect it
> would give different results.
>
> If any one has any ideas, please toss them at me.
>
> I will continue to refine my program, though working on something that
> only does a portion, even a large portion, of what it is intended to
> do annoys the hell out of me.

The buffers used to translate the 6551's ttl signals up to the +-12 volt 
levels to drive the port to rs-232 specs,  have been rather a nuisance 
to maintain, anyplace they are used, not just with the rs-232 packs, in 
fact I've had less troubles with them there than in most of the other 
places I have found in my life as a broadcast engineer/CET.  Those are 
the mc1488 and mc1489 chips in the rs-232 pack.  The firecracker I 
believe, actually gets its power from the logic levels its being fed 
with and has to make very frugal of that limited power resource.

One of the two is the receiver buffer, converting the incoming line 
signals back to ttl levels for the 6551, and that should not be 
critical, but the transmit buffer is the one I'd replace on general 
principles, it may have lost part of its output stage from an EMP 
(nearby lightning strike, atom bomb...) and can no longer deliver full 
power in one or both directions from ground.  At this late date I do not 
recall which of the two is which, but in the absense of a calibrated 
oscilloscope to confirm its output levels, I would clip the chip out, 
clean up the pin  holes, put in a socket and a fresh chip.

Our rs-232 pack also has another achilles heel, that little tin cased 
Aztec 5 volt to +-12 volt invertor supply.  Mine went away nearly 20 
years back so my pack has been patched to get its +-12 volt power from 
the MPI.  Those 2 voltages should also be checked while its powered up 
and if either of those voltages is under 10 volts, you will probably 
find the tin case is way too hot.  There are electrolytic capacitors 
sealed into it, and if they are failed or in the process (heat is rifle 
bullet deadly to them), it will get hot.  Best bet, which will stop the 
potential for the smoke escapeing is to just unsolder it & remove it, 
bin it, then add some wire wrapping wire to connect from the pins at the 
packs card edge connector to get the +-12 volts it needs from the MPI, 
which can supply it without a struggle.  A cheap digital meter, hot 40 
watt or so soldering iron, wet sponge for wiping iron tip clean, a 
solder sucker, and a small roll of ROSIN cored solder should get you 
going.  And of course a schematic of the rs-232 pack, which ought to be 
available several places on the web.

Good luck & report back with results please.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>