On Tue, 9 Sep 1997 08:38:28 -0700 ERIC SCHLAEPFER
<eric.schlaepfer@AUTODESK.COM> writes:
>     Hello All,
>
>     A while back there was a thread about connecting a computer
>moniter to
>     a PIC controller. Well, I was intrigued, so I decided to see if
>this
>     could be done.
>

>     The final version displays some numbers(albeit static) on the
>screen
>     in a surprisingly crisp and noise-free manner. The resolution is
>about
>     25x370.

I'm in the very preliminary stages of building a PIC circuit as part of
my amateur TV transmitter (which is also in the very preliminary stages).
 The PIC will lock to the incoming signal and periodically overlay
station ID on the picture.  I'm looking at using a bitmap of 64 x 8
pixels stored in the EEPROM of a 16F84, stretching the pixels vertically
by repeating each line 4 times or so.  Using a PIC clock of 12 MHz, and 2
instruction cycles per pixel, the 64 pixels should fit in the viewable
area of the screen.  Also the PIC would generate an "early" horizontal
pulse to boost the voltage to the RF power amplifier though this could be
done rather readily with analog circuitry if need be.

The PIC clock needs to synchronize to the incoming video, I was planning
to do that by using software to compare the sync pulses with the state of
the program, and output pulses to a varicap diode that adjusts the
frequency of the PIC oscillator.  Anyone tried this before?  Just
synchronizing the software will let the text jitter (maybe slowly, maybe
quickly) one pixel or so back and forth.  Also a custom crystal would be
required.  So I'd really like to synchronize the oscillator itself.

(As a note, most TV's that display stuff onscreen have two oscillators in
the TV controller microprocessor, one is a crystal that runs all the time
and the other is LC.  The LC oscillator is started exactly in phase for
each line by a horizontal pulse and clocks the pixels out for that line.
Doesn't seem too practical to use this kind of design with a PIC though
maybe something along these lines could be done with the USART or SSP)

>
>     I had to calculate the timings based on the fact that the '84 at
>4mHz
>     runs at about one microsecond per instruction,(two for jumps) and
>that
>     the MDA scan rates are 18.432 kHz horizontal, 50Hz vertical. 4mHz
>is
>     just about the bare minimum to display legible numerals on the
>     display; at least with these refresh rates.
>
>     If anyone is interested in the circuit diagram and program source
>     code, feel free to email me and I will send them to you.
>
>     Eric Schlaepfer
>
>     P.S. Off Topic: does anyone know what could be wrong with my
>broken
>     moniter? I think it is a fuse because the power light doesn't
>come on
>     when I flip the switch. Also, when the thing broke there was no
>smoke
>     or that nasty electrical burning smell.

Probably you applied an incorrect (too low frequency or wrong duty cycle)
horizontal drive signal.  This causes various inductors to saturate and
make too much current flow in the horizontal output transistor, burning
it out.  They nearly always fail shorted, thus causing the fuse to blow
and the pilot light to go out.  I suppose it's conceiveable that the
excess current could just blow the fuse and the transistor is OK, but it
doesn't seem likely.  Many MDA monitors (including the original IBM one)
are susceptible to damage this way because they apply the horizontal
pulses from the video cable directly to the horizontal output circuit.

Every other type of monitor uses a PLL oscillator which is certain to
drive the horizontal output at an acceptable frequency.  If the supplied
horizontal pulses are out of lock range or not present at all, the
picture won't lock but no damage will occur.

>