[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Will my graphics card support this?

Gene Heskett wrote:
On Sunday 31 December 2006 17:19, Ric Moore wrote:
On Sun, 2006-12-31 at 20:39 +0800, Chong Yu Meng wrote:
On Fri, 2006-12-29 at 09:15 -0800, alan wrote:
There is a modeline generation utility, but I cannot remember the
name of it.  (My brain is still booting this morning.)
Be careful as hell with xvidtune. It'll do whatever you merrily tell it
to do. I blew the flyback out of a beautiful multisync back when a 15"
model cost some large. I drove it just a teeny tiny itsy bitsy too much.
ZING! <puff of smoke> That was a CRT monitor, BTW. I destroyed it
dinking with it and xvidtune, cause I didn't know exactly what I was
doing. That was one sad day. Ric

Your list resident C.E.T. speaks:

You didn't drive it too hard Ric, you drove it too slow, specifically the horizontal rate. The design of an h-sweep circuit is a rather large balancing act, with the inductance of the sweep transformer often at the ragged edge of a saturated core when running at the usual lower end of a vga monitors range of 31 kilohertz.

The reason I mention saturated core is that at the lower end of the range, there is more time for the current through its primary winding to build up that must be turned off by the sweep transistor at the right edge of the screen. When the core is not saturated, the inductance controls the rate of current rise and it might be 1.5 amps at that particular fraction of a microsecond, but if slowed down too much, (30khz is pushing your luck, 29khz maybe a few seconds, 28khz generally means toast in about a second) the rate of rise of the current will approach and hit the core saturation point, it can't hold any more magnetic field regardless of how much current flows, so the surplus field is ejected into the air around the core AND the inductance that controls the current disappears. The currents can then rise from the 1 to 2 amp range it operates normally at, to 10 or 20 amps in the next microsecond. The transistor cannot shut those current levels off at the right edge of the screen and survive for long, sometimes only milliseconds before it shorts out internally, and a lot of other parts are blown/burned in the process before the main shutdowns can operate, be they fuses or mechanical circuit breakers.

On the other end of the range, I've operated for many years an old NEC 5FG that's rated to go as high as 67khz, on a driver that merrily runs it at 79khz while doing a 1600x1200 screen The width and brightness perhaps may suffer, but its otherwise sharp and in perfect convergence today. The NEC 5FG had buckets of width overdrive, unlike this 19" Starlogic which is all run out at this same resolution and not filling the screen by about 1/8" on each side.

Anyway, that's an explanation, hopefully in understandable terms, of the probable cause as to why it made toast on you.

There are also several other failure mechanisms, usually related to partially failed electrolytic capacitors going un-noticed or ignored until the blowup, but those failures are going to happen even if its driven completely within its range ratings. Those are related to heat and old age, and the quality of those capacitors in the first place.

Thanks for the info Gene. The explanation seemed clear to me and quite detailed. Not a Certified Electronics Technician though.


Every young man should have a hobby: learning how to handle money is
the best one.
		-- Jack Hurley

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]