[GreenKeys] Model 15 + C-64 Baud rate question

[drlegendre .]

@JIm & All

I don't have to use the C-64. That was my choice as I have a semi-permanent
C-64 system running, I'm generally familiar with the system, and it seemed
(at first) to be the simplest way to handle this. As for old PCs, I do have
an original IBM PC and an XT around here, but neither are in-service at the
moment, nor do I have the space to set one up right now. Also unsure if
either has a serial card.

But if it comes down to it, that might be a way to go, at least for a
while.

Now as for other old but small hardware, I have Tandy 100/102 machines,
Epson HX-20, Epson PX-8 Geneva and... hey, there's a Zenith SUPERSport
laptop that I think is an XT clone. I'll bet it has a serial port - do you
think +that+ might be able to do it? There's also a NEC UltraLite that runs
a NEC V20 (or V30?) CPU, that might also be of use if it has serial (and is
documented).

I've also heard that you can drive mdl. 15 with Arduino stuff, even the
first versions.. I have ZERO exp. with that hardware, how much trouble
would it be?

Thanks for all the help, gents!!

On Sat, Jul 4, 2015 at 4:54 PM, Jim Haynes <jhhaynes at earthlink.net> wrote:

> Putting a delay between characters will not help, because the individual
> bits are still shorter than the machine expects.
>
> The selector cams sample the state of the selector armature in a narrow
> slice of time.  You want these samples to take place in the middle of
> the signal pulses, so that if the boundaries between pulses are off
> for any reason the selector has the best chance to get a correct
> selection.  All this timing is derived from the START pulse when the
> selector magnet goes spacing and allows the selector cam clutch to
> start rotating.  The range finder moves the relationship between
> the sampling pulses from the cams and the beginning of the START pulse.
>
> The process of adjusting the range finder between its limits while
> receiving a signal is intended to set the sampling time as near the
> middle of the bits as possible.  The errors you get at the extremes
> are caused by the sampling time being so early or so late that the
> selector armature is moving at the time of sampling.
>
> If the motor is running slow (relative to the incoming signal) the
> transitions between bits will steadily get ahead of the optimum time
> for selection.  You may be able to get by with this by advancing the
> sampling times to be early.  If you are lucky you don't get it so
> far advanced that it samples at the beginning of the #1 pulse, nor so
> retarded that it samples at the end of the #5 pulse.
>
> If you don't have to use the C-64, if you can just use an old PC, the
> old serial ports can be set to run at 45.45 baud.  But if you must use
> the C-64 then it would seem like getting the amateur radio program for
> it would be the easiest way to go.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/greenkeys/attachments/20150704/05dbc868/attachment.html>