amgtree wrote:Hello I know it's a while since my last message on this thread, but I've had this idea lately about modifying the magnetic yoke of a crt monitor to increase the vertical and horizontal frequencies.
I'm aware of the dangers that this involves with the voltage of a crt and I will take precautions before doing anything. However what I'm not aware on is the bottlenecks of these devices. If anyone could point in the right direction such as a book or page that would enlighten me on this it would be great.
While I might know more than the average person I am still greatly ignorant on the exact technology that is going on so I don't really know if this possible/practical.
If this belongs on it's own thread let me know.
In theory, it is possible.
It'll likely be much easier to modify the vertical deflector than the horizontal deflector, since the vertical is much lower frequency than the horizontal. That way, you can at least reduce vertical resolution to get higher vertical refresh rates. So 1600x1200 progressive ~85Hz could in theory do 640x480 interlaced at ~425 Hz at the same horizontal frequency (no horizontal deflector modification), assuming the blanking interval to picture ratio is roughly the same.
Vector CRTs can do quite high refresh rates (both horizontally and vertically) if you're drawing only a few vectors (e.g. just a single square). However, in raster CRTs, the vertical deflector may be underdesigned relative to the horizontal deflector (due to the lower frequencies involved) -- and it might even be at the yoke level (less beefy coil of sorts). But odds are that the coil for vertical/horizontal is the same, and you just have to upgrade the scanning circuits (not familiar enough with this) to permit higher vertical Hz.
Another technique before modification -- to get higher refresh rates (e.g. 220Hz interlaced) -- with 110Hz progressive capability -- is to proportionally scale your VBI (Blanking interval) bigger.
This make sure it takes the same amount of temporal time between refresh cycles. This might bump the refresh rate down a bit, like 210Hz or 205Hz, but a larger blanking interval means more time for the vertical deflector to reset to the top. Vertical height of picture might shrink because of limited vertical scanning velocity headroom.
For example, with Vertical Totals (VT) in a Custom Resolution Utility.... If you use VT1125 at 1080p/60Hz (45-line blanking interval), you might need VT1170 at 1080p/120Hz (90-line blanking interval) which might need to become VT1260 at 1080i/240Hz. The scaling is not exact as the horizontal scanrate changes with larger blanking intervals at the same refresh rate. So your jump might only be 110Hz->205Hz rather than 110Hz->220Hz.
Assuming your scanning circuit doesn't play policeman and prevents it from happening, but in theory, half-height ~205Hz interlaced (black areas at top/bottom + proportionally larger blanking interval) with large blanking intervals, doesn't push your vertical deflector any harder than full-height 110Hz progressive. It ends up being the same vertical scanning velocity, and the same reset velocity (larger VBI at higher Hz = same physical time on clock = same time to allow electron gun to reset to begin the new refresh cycle).
A lot of this is beyond my scope, obviously, as I don't have enough knowledge to modify a CRT. But I have a good understanding of custom resolutions, horizontal scanrate, vertical scanrate, blanking intervals, horizontal/vertical totals, etc. All of that stuff was originally invented for CRT. So I at least consequently, how to maintain VBI time at different refresh rates (e.g. practically double VBI size at double refresh rate, in order to essentially maintain VBI time on the clock -- this can help satisfy law-of-physics constraints like electron gun momentum). So I have a good sense of what will be limitations electronically, and what will be limitations physically.
----
First step though, try increasing blanking interval at higher refresh rates.
1. Load the highest resolution 110Hz mode you have.
2. Record the "Horizontal scan rate" or "Horizontal refresh rate". That's your maximum your display supports.
3. Create your working 170Hz interlaced mode
4. Try increasing blanking interval a little bit (proportionally increase Front Porch / Sync / Back Porch). Numbers will vary from setup to setup. But if it was a 5:30:5 pixel ratio (FrontPorch:Sync:BackPorch) at a properly working 110Hz, try a 10:60:10 pixel ratio at 170Hz. That's your "enlarged VBI" to allow your electron gun more time to reinitialize for the next refresh cycle. So doubling time of VBI at double refresh rate == same physical clock time. (Note: For progressive->interlaced switch, a 10:60:10 actually be 5:30:5 due to the interlacing). Doubling the VBI size is probably overkill at only 110Hz->170Hz, but this will allow you to begin try testing for higher refresh rates.
5. Look at your new "Horizontal scan rate" setting in your CRU (e.g. ToastyX or NVIDIA or AMD). The number is probably still lower at 170Hz interlaced with slightly bigger VBI, than 110Hz progressive at original VBI.
6. Now increase your refresh rate slowly, in 5 Hz increments, testing the new refresh rate in between.
7. Repeat Step 6 until blanks out, back off 5 Hz, and that's your new max refresh rate.
Vertical picture height may shrink, but that's normal -- that simply means the vertical deflector isn't working harder at higher refresh rate. (If it doesn't, picture height remaining the same -- then your vertical deflector IS working harder at higher refresh rate -- CRT automatically trying to compensate vertical shrink with more power to vertical deflector).
With this VBI headroom trick, you might find that your 170Hz interlaced limit is actually closer to, say, 190Hz or 200Hz interlaced -- assuming your electronics circuits (e.g. OSD, interlocks, safeguards, etc) doesn't play policeman and prevents this from happening by automatically blanking the display on unrecognized timings.
Do make sure you understand fully, how & why, the above works, before you begin physically modifying the CRT.
[If it turns out you're actually working the vertical deflector harder (non-shrunk picture + more heat coming from CRT) -- at 170Hz than 110Hz -- and you want to risk more power to an existing vertical deflector -- try adding a cooling fan to the CRT case; to keep things a little bit more cooled down -- in order to prevent overheating and premature failure. No guarantee.]
----
Beyond this -- if you want to begin modifying your CRT -- it is out of the scope of my knowledge abilities. Also, it's very dangerous voltage in there (enough voltage is left, in the capacitors, to kill a person -- even after the CRT has been unplugged for an hour). Homebrew LCD monitor modding is safe in comparision to CRT modding, so you won't be able to find much information. If you dare to do it, be armed with the necessary safety equipment, high-voltage-capable multimeter, safe capacitor discharge device (resistor+wires+probes, essentially), get a TV/monitor repair manual first. Study all CRT TV repair safety tips -- don't skip studying high voltage safety. There's something like 27,000 volts in some CRTs/circuits! You have been warned...