Main > Monitor/Video Forum
HDMI to Old CRT
Zebidee:
About that pencil note on your schematic - they wrote "RED | BLUE | GREEN | WHITE". White against BLK.
This made me think about your friend worrying about contrast levels and pull-ups.
Blanc and blanco etc, in many languages, mean white. In most video signals, there is a luma (composite, svideo, component all have it) which is, more or less, brightness and everything else except colour. This luma is carried on the same signal line as the sync. But not so for RGB
In RGB, all that luma data is rolled into the colour signals. The sync is just the sync The blanking is just a switching signal and nothing else, not "white". When the voltage is between X and Y, OSD RGB is switched on.
In terms of pull-ups: most times you'd input external RGB and not use the OSD; but leaving the OSD's RGB lines connected is important for pulling up the voltage. Between the OSD, the jungle IC and RGB inputs, the resistors form a voltage divider matrix, which ideally delivers 0.7vpp to those 104 capacitors and RGB pins.
abstract3000:
Ok so the TV was altered to the suggestions made above but still no dice :( here is the current schematic of how it's set up
This is what I got:
I have made the suggested modifications and am even more convinced the issue is the jungle is looking for a TTL video signal. If I run a CGA TTL signal in we get a perfect picture. Keeping all the parameter the same and simply changing the signal to analog you get nothing. Signal is being generated from a Sencore CM2125.
Adding the potentiometer on the blanking line does give me control of the blanking threshold and changing the resistor values I think made the picture crisper but still will not work with an analog signal.[ /i]
As you can see we just put a db15 plug on the TV and running VGA to vga cable from PC directly to TV.
I'm going to try and play with the CRT emudriver software some more tomorrow.
Zebidee:
OK, that does look about right to me.
Make sure that the PC with CRT_emulator is outputing 15khz signal properly. Try testing it with another 15khz display/TV/monitor, if possible.
If no other display to test on, and if you have access to one, hook up an oscilloscope to check out the sync.
abstract3000:
Ok so checked the Software and Still no dice, it is in the emulation mode giving out the 15.7, but as mentioned the Sencore CM2125 Machine has the settings 15.8 with the 640x 480 set signal sending the screen. When switching the machine to digital we get a Screen, but when we turn off the digital for analog it disappears. This has us convinced the signal is being locked out due to being Digital RGB, at this link https://crtdatabase.com/modding/rgb-via-mux you can see he states the following criteria is a "Must" for the RGB Mux Method:
* If your jungle does not have RGB inputs you cannot use this method.
* If your jungle has digital RGB inputs you cannot use this method.
we did in fact test the signal on it's way to the Jungle with an oscilloscope and this is what we got:
Digital input gave the following:
.978 (5.28pp)
Analog Gave the following:
.128 (2.3pp)
All throughout the SHUMPS Thread it is mentioned that if the Jungle is Digital RGB it's a no go. So this had us seriously discouraged. Though after reading for a few hours through everything I stumbled upon a post that discussed the Sanyo LA76843N Chip which you linked earlier "IX3354CE" chip is really a "X3354CE" chip which is really an "LA76843N" <- That was quoted from MarkOZLAD.
So the Thread I stumbled upon discussed how Voltage needed to be injected into the RGB Lines for this Chip to get the Signal. This explains in the first attempt of the mod, why when handling that little circuit board I had with both the auto on circuit and the RGB mod, when messing around with it I managed to get an image displayed as I must have been accidently introducing the 5V from the On/Off circuit to the RGB lines. You can see The picture here on the screen:
So It appeas from his workup on Stack Exchange here: https://electronics.stackexchange.com/questions/568045/single-supply-op-amp-non-inverting-level-shifter-for-ntsc-signal-processing-from/568154?noredirect=1#
He uses the THS7315 OP amp and drives it up to 2.5Vpp, though my friend suggests it might get better with 3.5Vpp. Not sure if you have any idea on this work around, or most likely being a one off. The issue is all the people involved in this model of TV MarkOZLAD, rx7turbo233, Syntax, and benyamin39 have not logged into the forums over there for months or in a couple cases years. The main thread has ran stagnent with the tail end just being numerous people asking for help with no response from those that once lead it. I understand the times change and people move on, but it would be great if I could at least get ahold of the Ben guy to see what the final results were.
I tries responding to his thread quoting it "Crossing my fingers" it triggers an Email and he checks, but luckily it appears to be allot of detailed info he left behind that at least we have something to work with.
Zebidee:
--- Quote from: abstract3000 on October 31, 2023, 11:45:57 pm ---Ok so checked the Software and Still no dice, it is in the emulation mode giving out the 15.7, but as mentioned the Sencore CM2125 Machine has the settings 15.8 with the 640x 480 set signal sending the screen. When switching the machine to digital we get a Screen, but when we turn off the digital for analog it disappears. This has us convinced the signal is being locked out due to being Digital RGB, at this link https://crtdatabase.com/modding/rgb-via-mux you can see he states the following criteria is a "Must" for the RGB Mux Method:
* If your jungle does not have RGB inputs you cannot use this method.
* If your jungle has digital RGB inputs you cannot use this method.
--- End quote ---
I get this now. The TV can show analog fine, but the digital OSD RGB inputs won't read analog.
--- Quote ---we did in fact test the signal on it's way to the Jungle with an oscilloscope and this is what we got:
...
Analog Gave the following:
.128 (2.3pp)
--- End quote ---
I don't understand this, as you say you are measuring sync before it gets to the jungle IC. Are you measuring the analog sync at (or close to) the composite/AV input jack? Analog sync should still have the square waveforms, much like the digital input picture you showed. If there is no sync from the VGA card, might explain some things.
Check the sync both before and after any modifications (resistors etc).
--- Quote ---So the Thread I stumbled upon discussed how Voltage needed to be injected into the RGB Lines for this Chip to get the Signal. This explains in the first attempt of the mod, why when handling that little circuit board I had with both the auto on circuit and the RGB mod, when messing around with it I managed to get an image displayed as I must have been accidently introducing the 5V from the On/Off circuit to the RGB lines. You can see The picture here on the screen:
So It appeas from his workup on Stack Exchange here: https://electronics.stackexchange.com/questions/568045/single-supply-op-amp-non-inverting-level-shifter-for-ntsc-signal-processing-from/568154?noredirect=1#
He uses the THS7315 OP amp and drives it up to 2.5Vpp, though my friend suggests it might get better with 3.5Vpp.
--- End quote ---
The blanking signal might have multiple states? Like, it might be one voltage level for digital, another for analog. When your finger touches the 5v for blanking, it may be pulling the voltage down (like a voltage divider, some of it being grounded through you), triggering analog. It's just a theory, but it is quite common for different blanking voltages to trigger different things. Could be worth putting that pot back on and going through the range while inputting analog.
You could try a video amplifier from Ultimarc. They are relatively cheap, a bit more powerful, and I think they supply a VGA cable with it.
What puzzles me about this is that if this amping idea works, I'd expect to see at least a faint image without it. Maybe there is a threshold.
--- Quote ---Not sure if you have any idea on this work around, or most likely being a one off. The issue is all the people involved in this model of TV MarkOZLAD, rx7turbo233, Syntax, and benyamin39 have not logged into the forums over there for months or in a couple cases years. The main thread has ran stagnent with the tail end just being numerous people asking for help with no response from those that once lead it. I understand the times change and people move on, but it would be great if I could at least get ahold of the Ben guy to see what the final results were.
I tries responding to his thread quoting it "Crossing my fingers" it triggers an Email and he checks, but luckily it appears to be allot of detailed info he left behind that at least we have something to work with.
--- End quote ---
Those guys, especially MarkOzlad and Syntax, would probably know what is possible. They both post on the FB CRT groups. Mark is one of my friends and I've seen him active recently, at least on the Aussie CRT group. I'll PM some details to you so you can contact him.