What happens if...

[Nicholas Bodley]

.... if you take a modern CRT monitor, a couple of function generators,
connect them, and vary the horizontal and vertical sync (drive?)
frequencies continuously (within specified ranges)?

So far, I've seen only specific resolutions and corresponding scan
frequencies (often only implied).

If the monitor "doesn't mind", what happens if a frequency varies fairly
fast? (Surely, a change in resolution is a step function, perhaps with
garbage until it settles down...)

Seems to me that CRT monitor scan circuits probably don't contain any
high-Q resonant circuits or equivalents, and that when multifreq.
monitors were first designed, perhaps the approach was to "broadband"
their scan circuits.

I assume there are something like scan-amplitude-feedback circuits to
keep the raster size within reason.

Maybe I should sub. to sci.electronics.design, and post this query
there.

Regards, and TIA!
--
Nicabod =+= Waltham, Mass.

 

[Samuel M. Goldwasser]

Nicholas Bodley <n_bod_ley@speakeasy.net> writes:

... if you take a modern CRT monitor, a couple of function generators,
connect them, and vary the horizontal and vertical sync (drive?)
frequencies continuously (within specified ranges)?

So far, I've seen only specific resolutions and corresponding scan
frequencies (often only implied).

If the monitor "doesn't mind", what happens if a frequency varies fairly
fast? (Surely, a change in resolution is a step function, perhaps with
garbage until it settles down...)

Seems to me that CRT monitor scan circuits probably don't contain any
high-Q resonant circuits or equivalents, and that when multifreq.
monitors were first designed, perhaps the approach was to "broadband"
their scan circuits.

I assume there are something like scan-amplitude-feedback circuits to
keep the raster size within reason.

Maybe I should sub. to sci.electronics.design, and post this query
there.

They rarely care about V - anything within the spec'd range should work.

For H, some monitors have had more than one range with gaps between them.

It's not clear what you are trying to do but you will probabl just
have to test it with your candidate monitor.

--
sam | Sci.Electronics.Repair FAQ: http://www.repairfaq.org/
Repair | Main Table of Contents: http://www.repairfaq.org/REPAIR/
+Lasers | Sam's Laser FAQ: http://www.repairfaq.org/sam/lasersam.htm
| Mirror Sites: http://www.repairfaq.org/REPAIR/F_mirror.html

Important: Anything sent to the email address in the message header above is
ignored unless my full name AND either lasers or electronics is included in the
subject line. Or, you can contact me via the Feedback Form in the FAQs.

 

[William Sommerwerck]

Driving a CRT monitor outside its design range might damage it. I would
stick within the spec'd ranges and frequency combinations before
experimenting outside them.

 

[JB]

"Nicholas Bodley" <n_bod_ley@speakeasy.net> wrote in message
newsan.2009.07.19.06.52.52@speakeasy.net...

... if you take a modern CRT monitor, a couple of function generators,
connect them, and vary the horizontal and vertical sync (drive?)
frequencies continuously (within specified ranges)?

So far, I've seen only specific resolutions and corresponding scan
frequencies (often only implied).

If the monitor "doesn't mind", what happens if a frequency varies fairly
fast? (Surely, a change in resolution is a step function, perhaps with
garbage until it settles down...)

Seems to me that CRT monitor scan circuits probably don't contain any
high-Q resonant circuits or equivalents, and that when multifreq.
monitors were first designed, perhaps the approach was to "broadband"
their scan circuits.

I assume there are something like scan-amplitude-feedback circuits to
keep the raster size within reason.

Maybe I should sub. to sci.electronics.design, and post this query
there.

Regards, and TIA!
--
Nicabod =+= Waltham, Mass.

If you bypass the blanking and sync circuits and apply your external drive
as deflection, you get this:
http://en.wikipedia.org/wiki/Lissajous_curve

Of course this assumes the gun(s) is biased on all the time for sufficient
brightness of the spot.

Note that there must be some frequency relationship between the horizontal
and vertical inputs in order to have a stabile scan or your eye won't be
able to follow it's path. It will continually shift with the difference in
phase between the two inputs.

 

[Eraklik]

On Sun, 19 Jul 2009 09:01:56 -0400, Samuel M. Goldwasser wrote:

They rarely care about V - anything within the spec'd range should work.

That's what I expected.

For H, some monitors have had more than one range with gaps between
them.

Had also suspected that; I think I once encountered a monitor with a
relay; I could hear it click when changing resolutions. Considering how
horizontal scan circuits work (sorry to say, I never completely learned),
those for TV surely do not seem to be broadband at all. Not sure I'd call
them "tuned", but designed to run only over a narrow range of
frequencies, it seems.

Fwiw, the Megascan (?) monochrome monitor (4096 x 3840, or something
close, ran H scan at 240 kHz, as I recall. Yoke was amazing -- seemed to
be like a "stator yoke", but windings extended out about 9 inches, like
petals of a flower; cooling, maybe?) Res was 300 dpi, just about sure.
Astonishing. No jaggies visible at all.

It's not clear what you are trying to do but you will probably just have
to test it with your candidate monitor.

Mainly, just wanting to learn more; replies greatly appreciated.

Thanks much, Sam!

--
[e]

 

[Nicholas Bodley]

Sam,

Sorry about this; couldn't e-mail you although I wanted to include you.
Message was posted to the group, though. Essentially, an ack. and thanks,
with a bit of interesting stuff. Remembered your filter only after
posting, and totally understand. Shouldn't have done an all-nighter!

Gory details:
Only fairly recently did I switch to Linux only (after accidentally
deleting my partition table (MBR) and losing Windows). Pan newsreader
doesn't link to Opera, which I use for e-mail.

Thanks, again. You've been of tremendous help to countless people over
the years.

--
Nicabod =+= Waltham, Mass.

 

[Nicholas Bodley]

On Sun, 19 Jul 2009 06:06:03 -0700, William Sommerwerck wrote:

Driving a CRT monitor outside its design range might damage it. I would
stick within the spec'd ranges and frequency combinations before
experimenting outside them. I wouldn't push my luck, though!

Indeed so; thank you! However, it seems that many, possibly all modern
CRT monitors detect whether the requested scan rates are within range,
and refuse to display (and self-protect) if outside of range. A crash a
while back made my computer put out something out of range, and I got a
red warning on the screen. Rather startling!

One reason I asked for info about the DEC VRC-21WA was hope of getting
scan rate limits specifically for it. (I really have tried!) For now, I'm
using the limits for the one that it seems to be similar to.

Main point of my query was largely curiosity about what happens if you
vary scan rates continuously within acceptable range. (I'm not likely to
want to display at some quite-oddball resolution.)

I see that my newsreader is replying by e-mail, but that doesn't work;
sorry. I use Opera e-mail, but Pan doesn't "know" that.

Best regards,
--
[e]



--
Nicabod =+= Waltham, Mass.

 

[Samuel M. Goldwasser]

Eraklik <eraklik@yahoo.ca> writes:

On Sun, 19 Jul 2009 09:01:56 -0400, Samuel M. Goldwasser wrote:


They rarely care about V - anything within the spec'd range should work.

That's what I expected.

For H, some monitors have had more than one range with gaps between
them.

Had also suspected that; I think I once encountered a monitor with a
relay; I could hear it click when changing resolutions. Considering how
horizontal scan circuits work (sorry to say, I never completely learned),
those for TV surely do not seem to be broadband at all. Not sure I'd call
them "tuned", but designed to run only over a narrow range of
frequencies, it seems.

Fwiw, the Megascan (?) monochrome monitor (4096 x 3840, or something
close, ran H scan at 240 kHz, as I recall. Yoke was amazing -- seemed to
be like a "stator yoke", but windings extended out about 9 inches, like
petals of a flower; cooling, maybe?) Res was 300 dpi, just about sure.
Astonishing. No jaggies visible at all.

It's not clear what you are trying to do but you will probably just have
to test it with your candidate monitor.

Mainly, just wanting to learn more; replies greatly appreciated.

Right, and as at least one other poster noted, and the S.E.R FAQ emphasizes:
Attempting to run at an unsupposrted scan range may result in smoke.

--
sam | Sci.Electronics.Repair FAQ: http://www.repairfaq.org/
Repair | Main Table of Contents: http://www.repairfaq.org/REPAIR/
+Lasers | Sam's Laser FAQ: http://www.repairfaq.org/sam/lasersam.htm
| Mirror Sites: http://www.repairfaq.org/REPAIR/F_mirror.html

Important: Anything sent to the email address in the message header above is
ignored unless my full name AND either lasers or electronics is included in the
subject line. Or, you can contact me via the Feedback Form in the FAQs.

 

[Nicholas Bodley]

On Sun, 19 Jul 2009 16:21:22 +0000, JB wrote:

If you bypass the blanking and sync circuits and apply your external
drive as deflection, you get this:
http://en.wikipedia.org/wiki/Lissajous_curve

Of course this assumes the gun(s) is biased on all the time for
sufficient brightness of the spot.

We're in the realm of vector graphics, but vectors are straight;
Lissajous figures are nearly always curved. "Stroke writing" seems more
appropriate -- think paths in PostScript. This is *very* analog!

The idea is valid, but the horizontal scan circuit is not an amplifier.
The sawtooth current is created within the output stage, not created at
low level and amplified by the output stage. It's in the class of what
were called "pulse circuits" decades ago.

Early in the days of TV, when magnetic deflection became mandatory*, a
simple brute-force power amplifier for high-frequency scanning would have
been impractical, I'm essentially sure. Huge amount of power, maybe kW,
likely, because a low-inductance yoke needs lots of current, while a
higher-inductance yoke needs quite a high voltage.
*The HP electrostatic wide-angle deflection yoke (I'm not kidding! CRT
bulb similar to that of a mag. defl. CRT) had not been invented yet.

Horizontal scan circuits for mag. defl. TVs and monitors are quite-clever
circuits that, I've read, recover energy (a bit like regen. braking in a
vehicle, most likely). Unless I'm badly confusen, the horizontal output
tube or transistor is essentially a pulse switch, and the whole output
stage has the character of a resonant circuit.

Shame on me -- I never learned, in any detail, how TV horizontal sweeps
work, and would welcome a link to a good explanation. I do know that the
"damper diode" is an essential part of the circuit.

If you wanted to modify a monitor to work as an X-Y display, the
horizontal yoke windings would, very likely, have too low an inductance
(and the vertical, perhaps too high!). There have been mag. deflection
large-screen X-Y displays, but they were never fast; that's why H-P
developed that large-screen short-tube X-Y display. Even in that display,
large beam swings at its fastest capability required booster transistors,
and iirc one couldn't have fast, repeated, full-screen deflection at its
upper frequency limit. Small swings, yes.

Nothing wrong with the CRT, itself, afaik, for use as an X-Y analog
display, although color convergence would be a ball. (It has been done,
in an arcade game, no less. I was painfully tempted to buy a surplus
color X-Y display, years ago.)

If you were to use an existing CRT for an X-Y display, you'd want a
different yoke, with (probably) the same inductance for both axes.

Then, you'd have the fun of designing analog power amplifiers; probably,
adapting an audio power amplifier would be a way to go, but a practical
deflection yoke might have many times more inductance than any
loudspeaker. A first-rate EE should be able to do it. However, don't
expect much swing in the ultrasonic range, and forget 100s of kHz.

There was a low-cost video game (Vectrex?) that used an X-Y analog
display with magnetic deflection; small CRT, maybe 9 inch. It could put
up small rasters.

--
Nicabod =+= Waltham, Mass.
who could have bought a surplus Charactron CRT, years ago.
Neck was about a yard long, or so it seemed.