Crystal load?

[tuinkabouter]

I learned a little more about this circuit.

It's a proprietary video card card from a piece of German offset printing
equipment.

I used a sig. gen. in place of the crystal. The circuit generated a video
signal without any data (just sync pulses). The video card is separate from
the system processor, so being on the bench the card has no data to display.

Viewed on a scope, varying the 8.867 MHz frequency doesn't change the video
signal at all.

I'm guessing (with my limited understanding of how video works) that the
8.867 runs the dot clock, basically how fast the dots are shoved out of the
data bus into the video generator IC.

So, how critical *is* this frequency?

You can still buy it in germany.
Standardquarz, Grundton, 8,867238 MHz 0.18 euro (0.24 us dollar)


https://secure.reichelt.de/Crystals...8AAAIAAA-vdIc96bf9a2d5cb50299c3e3c87380e0d4b7

Minimum order 10 euro.
Only drawback is the transport cost (

 

[rickman]

Have you ever worked with video? It doesn't sound like it.

Here is a photo of a mono monitor being driven by a CGA card, which
is less than 16 hz off at the H sweep:

<http://i.ebayimg.com/t/Mono-Video-C...Aw/$T2eC16RHJF0E9nmFSIbDBQ-CLvDguw~~60_12.JPG>

It is outside the capture range of mono monitors. If it wasn't
critical, they would have used a low tolerance ceramic resonator instead
of a crystal. Being off 10 Hz at 3.579,545 MHz is out of spec. With the
typical 14.318,180 MHz crystal used for broadcast video gear, that is a
max of +/- 40 Hz. or less than 2.8 PPM. I've seen more than one mono
monitor die because it was used on a CGA board, and some went up in
flames. The flyback is designed for a specific frequency, and quickly
loses efficiency as the Q drops.

3,579,545 Hz = color
3,583,185 Hz = mono
---------
3,640 Hz = difference in dot clocks.

I have repaired monitors for over 40 years. I was a broadcast
engineer at three TV stations. You keep spouting vauge crap. Should.
Maybe...

I notice the machine the monitor is sitting on looks like it has a tape
drive backup in it. What is this a 486 some 20 years old? Yes, the
earlier monitors were difficult beasts that were very sensitive and
could even be damaged by running them off frequency. But they haven't
made that sort of thing for a long, long time. Far too long to still be
in use. If the OP is actually driving one of those monitors, I would be
surprised. In that case, the crystal might need to be ±1000 ppm
accurate. But for any modern monitor it will adapt to whatever you send
it nearly without limit. Have you worked with a monitor that isn't 40
years old?

My point is that for $5 the guy can get some crystals that he can try.
Otherwise he seems to be having a ton of difficulty finding anything he
can use. I believe he hooked the board up to a monitor to test running
it from a signal generator. Did the monitor burn up then? Why don't
you try getting a little more real rather than busting my chops when you
don't know what type of monitor he is driving?

I can't believe you have worked for so long and actually believe the
crystal driving a CGA board was ±3 ppm accurate. That's an order of
magnitude more expensive oscillator than what they actually used,
perhaps two!

 

[rickman]

Have you used a 8275 CRT controller in the last 30 years? (Think!
That was used with the 8085 8-bit processor)

'Modern' monitors are not NTSC. They are digital, and the current
types are HDMI. Therefore, all timing is generated by the onboard
display controller. I still repair monitors, including LCD types. Not
many, and not often but I do know how they work.




It's in an old Geramn made printer he's trying to reserect. Why
would it have a modern monitor. You need your 'chops' busted, in real
life. He is working on an old, retired system. You keep tossing out
crap that has nothing to do with his problem. He has been pointed to
multiple sources of the proper crystal.

Yes, an old piece of equipment that is in tatters and may or may not use
the original monitor. That is my point. Until the OP responds, you
simply don't know what monitor is being used. If he wants to test the
equipment with a $1 crystal rather than a $50 crystal that will take two
months to get, that is his choice.

Damn, you're stupid. CGA video used the 14.318180 system clock in the
PC. It was divided by 4 to provide the required burst. Early PCs had a
trimmer capacitor to set the exact freqency.

Have you ever seen a CGA only video card with a crystal? Once again:
COLOR BURST. SEVEN CYCLES. PHASE LOCKED TO BURST ONCE PER SCAN LINE.
Colorburst crystals were made by the millions, and for decades. They
only had to be able to be pulled to the burst, and remain stable for one
scan line. The two only have to be close enough to be pulled into
phase. I have looked at a lot of the burst crystals in free running
mode, and they were within 5 Hz. Is that good enough for you? Really,
you need to stop making a fool of yourself and look at the real
circuits.

I'm not going to argue with you about this. The crystals used in PCs
and nearly all consumer electronics are only accurate to a few 10's of
ppm and typically are only stable over temperature to roughly the same
range. You can tune the crystal to an exact frequency with a trimmer
and it will be off by 10's of ppm by the end of the day or before the
equipment has warmed up (or conversely after). Did PCs warm up the
oscillator before turning on the monitor?

The point is that talking about depending on crystals being ±3 ppm
accurate is not valid unless it is a more expensive piece of equipment
that can justify the cost of a temperature compensated oscillator. I
don't care about how many crystals you have looked at, holding a ±3 ppm
spec on an oscillator is not something you do with a $0.50 crystal
oscillator. CGA never had to rely on the timing being anywhere near
that accurate, period; or any other display I have ever worked with.
Heck, that is outside the spec of many frequency counters from HP and
elsewhere! The model HP5383A timebase is specified for 3 ppm/month
aging, 2.5 ppm between 0°C and 40°C and 0.5 ppm due to power 10% line
variation. This is a piece of equipment with an accuracy limited by the
crystal oscillator, so I think they would put a lot more effort into
that feature than PC makers would in their reference clocks.

Your claims that CGA monitors (or any others) need to be driven to a
timing accuracy of ±3 ppm is not supported by the facts.

Go be rude with someone else!

 

[josephkk]

It's a monochrome system (even more forgiving?).

So, it's series resonance I want?

This is the one in Ireland:

<http://www.donberg.co.uk/catalogue/passive_components/quartz_crystals/8..86723
8mhz.html>

Should this one work in my circuit?

The only reason I was going with the manufacturer is that they can provide
some expertise re. the application, whereas the Irish are only sellers.

I'm no engineer so cannot determine for myself the appropriateness of this
crystal for my application (hence my posting here).

Thanks,
Dave

It is my opinion that you can easily just use the crystal from the Irish
or German vendor. For some reason that particular frequency is very
standard in Europe and is mass produced by the millions. Very likely to
be just fine. The high volume production lowers the price. I doubt that
the actual frequency is any better than 50 ppm though, and more than 1000
ppm of frequency error may or may not cause problems.

?-)