[Coco] FM-77AV video sync problems

Gene Heskett gene.heskett at verizon.net
Wed Apr 12 01:40:24 EDT 2006


On Tuesday 11 April 2006 22:23, George's Coco Address wrote:
>Gene Heskett,
>
>  Since you are(as far as I'm concerned) the guru on raster scan
> television, I have a question..
>
>  I am confused about the power line frequency and how it relates to
>television.

Generally speaking, all the tv systems in use on a local basis where 
local is the country, will have vertical rates that match quite 
closely, the powerline frequency so that any residual hum will be very 
close to stationary rather than rolling thru the video due to the speed 
miss-match.

>All the TVs I've worked on convert the incoming AC power 
> to DC,
Yes.
> switch it to AC(almost), goes through a transformer at a high 
> frequency(efficiency), and then converts it back to DC.

Today this is done for all the voltages in a tv, but in yesteryear, only 
the CRT's anode voltage of several kilovolts was generated like that.

> All our cocos convert the incoming AC power to DC, albeit a lower
> voltage. If  a matching computer and monitor are used on a power
> frequency(50 or 60HZ), how can they have problems with sync, when
> there is no reference to the AC power frequency coming in?

Basicly the tv standard emitted by the coco must match that which the 
monitor is optimized for.

>  I can understand power supply problems when operated at incorrect
>frequencies, such as a transformer that is supposed to work at 60HZ
> vs. 50HZ.

This isn't a huge problem in most cases.  Yes, a 60hz rated tranny will 
run a little warmer on 50hz, but not enough to endanger its health and 
well being.  Conversely the 50hz tranny running on 60hz might run a 
couple of degrees cooler at the end of the day.

Where it gets to a right sticky wicket quickly is in the monitors 
horizontal circuitry.  In order to get decent efficiencies at vga and 
up speeds or 31 kilohertz and above, the amount and formula for the 
ferrites is optimized so that at the low end of 31khz, the full range 
of the magnetic material is being used, and its rather close to being 
saturated magneticly when the beam is at the right edge of the screen 
and the current, which was switched on at about the middle of the 
screen as the beam is swept (free wheeling diodes supply the currents 
for the left half of the screen, a form of energy recovery if you will) 
is at that point quite close to saturating the magnetics.

Now, imaging slowing this vga built monitors sweep frequency down to 
ntsc speeds of just under 16 kilohertz.  The current now has 2x the 
time to build up in an inductively limited rate of rise fashion, so it 
continues to rise into the region where the magnetic cores become 
saturated, at which point they drop out of the formula, leaving only 
the inductance of the coils as if they were wound on air to limit the 
currents rate of rise.  Under normal conditions there might be 2-4 amps 
of current circulating in the circuit, with the psu making up maybe .5 
amp for the circuits losses.

But when the cores effect on the inductance drops out, the current rise 
can go up to an additional 5 amps per microsecond or more.  Do this for 
the final 20 microseconds of each sweep and the average currents have 
risen to the failure point of the junctions in the transistor and it 
fails to turn off due to way too much stored charge in the substrate 
even after the drive has been removed.  A transistor than can turn off 
a 5 amp current in .3 microseconds now takes 5 microseconds to shut off 
50 amps, during which time the junctions are being subjected to the 
usual formulas for resistive power dissipation.  This effect alone will 
raise the heating in the transistors innards by several orders.  

So the time until the transistor shorts totally is usually measured in 
milliseconds, and of course that tends to take a lot of other stuff in 
the monitors power supply chain with it, letting the smoke that makes 
this all run out & it doesn't work anymore.  Also scattering small 
pieces of epoxy B around as the transistors usually will explode, 
although all you'll hear is a bit of a pop.

Generally speaking, its often not practical to try and repair such a 
blown up monitor, particularly if the core of the scan transformer was 
actually driven above its curie point in the process of the failure.

Most ferrites have a relatively low curie point, where thats the point 
in its temperature curve that its magnetic effects becomes 
non-magnetic.  Many modern ferrites have curie points well below the 
boiling point of water!  And if its subjected to a magnetic field when 
that hot, it seems to freeze the properties forever in the non-magnetic 
camp even after its cooled down.  I've been bit by that effect several 
times over the last 55 years.

And thats probably more than you ever wanted to know about the 
subject. :)

And I hope that helps to explain why I'm 100% against trying to make a 
monitor run at less than its rated hsync rate in its spec panel.  Its 
virtually guaranteed to let the smoke out & maybe break the mirror.

>George

-- 
Cheers, Gene
People having trouble with vz bouncing email to me should add the word
'online' between the 'verizon', and the dot which bypasses vz's
stupid bounce rules.  I do use spamassassin too. :-)
Yahoo.com and AOL/TW attorneys please note, additions to the above
message by Gene Heskett are:
Copyright 2006 by Maurice Eugene Heskett, all rights reserved.



More information about the Coco mailing list