On Saturday 17 February 2007, Tim wrote: >On Wed, 2007-02-14 at 11:48 -0500, Gene Heskett wrote: >> At the other extreme, 50 kilohertz and above, you will eventually run >> into the display width limitation the power supply's maximum voltage >> will impose, but this is generally harmless since the currents go down >> too and generally things will actually run cooler. > >Though scanning voltages go up. Yes, but it's the psu's regulator feeding the scanning circuitry that goes up in an attempt to maintain width in the face of ever shorter charging times at the higher rates. The currents however do not rise as you have pictured in your mind, but remain stable so the EHT developed for the crt anode generally remains pretty stable too. However, resonance effects are also at work in a real world circuit, and these generally are reduced at the higher frequencies, causing a generalized lowering of the EHT developed. The end result, carried to extremes if the circuit is pushed to say 25% above its nameplate maximum hsync range, is not only a narrower scan, but a dimmer image due to falling EHT. As long as the switching times of the scan transistor are decent, there will be no harm. But I'm not above pushing a monitor 3-6 khz past its ratings if it can still give me a full width pix and adequate brightness. This monitor I'm looking at is a case in point, if I reset the upper hsync range to its nameplate value, then 1600x1200 is not available. Its been running about a year that way, 24/7/365. The only case I'm aware of, where temps in the scan transistor went up to the self-destruct point involved a 17" Viewsonic whose scan transistor had an off time of nearly 2 microseconds. It is during the switching time of the transistor that the majority of junction heating is generated. Replacing that el-cheapo transistor with one rated with a 100ns off time lowered the temp of the heat sink by at least 75F. That circuit also had other design problems however, and after replacing nearly $50 in parts for the 3rd time, and it ran about 3 weeks, I gave up on that particular piece of junkyard electronics. And yes, I am a Certified Electronic Technician, my certificate is registered as NB-118. >And if EHT is generated by the >horizontal oscillators, that goes up as well. No it doesn't. Its the actual magnetic field collapse of 'x' current that generates the EHT, and the magnitude of 'x' is quite well tied to the actual width. No modern monitor has an actual coil to control the ratio of current through the yoke to the current through the EHT transformer, it is all done by a switching regulator circuit that adjusts the supply voltage fed to the scan circuit that is the 'width' control. Running at 31.5 khz vga speeds, this voltage can be less than 30 volts, crank it up to 75 khz, and it might be 390 volts, but the currents remain pretty stable, changing with the width settings only. >I've seen sets go "bang!" >when the horizontal frequency went too high, This might have been possible 35 years ago in an all tube type tv, but I have never personally encountered it myself, and my experience in the tv service shops goes back to 1949. Note that up to this point, we are discussing over range. Under range, below vga's '31.5 khz' speeds will generally be cause for immediate alarm, if not instant breakage of the mirrors, letting all the smoke out. Here, the increased time for the currents to rise can run into the core saturation point in the EHT transformer. At that point the core effectively disappears from the inductance formula, and the rate of rise of the current through the scan circuitry rises very quickly to 5-10x normal despite the regulators attempt to turn the voltage down to 10 volts or so. As this also, due to charge carrier lifetime in the semiconductor, slows the shutdown transition period, heating of the junction to the self-destruct point is very rapidly accomplished. Often in milliseconds. The circuit is destroyed so fast that you never get a chance to see the telltale scan compression at the right edge of the screen as the tube hasn't warmed up enough to give you a picture. The moral is of course to never, ever, run a crt monitor slower than its nameplate minimum h-rate. Personally I'd stay 5 khz above that minimum just to stay safe when production tolerances rear their ugly head. All of this talk, please understand, applies ONLY to crt circuits, the lcd circuitry in a lappies screen should always be run at its native pixel resolution, anything different will generally lead to an obviously degraded image, if the lcd will even accept it. [...] -- Cheers, Gene "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Yahoo.com and AOL/TW attorneys please note, additions to the above message by Gene Heskett are: Copyright 2007 by Maurice Eugene Heskett, all rights reserved.
