Diagnosing the DXC-M3A

[kf4rca]

Sorry, no IO's here. I think the CE's ran off with those.
The tubes on the left are from an Ikegami 730 camera and the tubes on the right are from a Sony BVP-3. As I recall, they had good beam but were changed out due to spots.
Here's an IO on Ebay:
https://www.ebay.com/itm/26630473807...Bk9SR6CVxaSbYg

[sdsw4]

This is overdue, a capture I made using some vhs to dvd thing (while I work on an rgbs capture method).

https://mega.nz/file/afggWSoS#wpG5cO...mM8dVGhpgG9Hp8

H.264 encoded, only things done post process was deinterlace. Other than that I set it to no.3 filter (cloudy day).

Since this though I have adjusted the tube focus to correct the corners. I'll wait for another cloudy day to record another take since I don't trust myself with tubes and broad daylight.

[ARC Tech-109]

The tubes are far more forgiving of bright scenes, start out with the ND filter at f16 and open the iris from there. What will damage them is a sharp contrast of dark & light parts of the scene as the auto beam control will not compensate fast enough, use the zebra function in the viewfinder to keep the image within the safe limits.

The image above looks great, the colors will definitely "pop" under real sunlight.

[sdsw4]

So in my quest to further mess with the camera, my latest gag reel is working with the RGB+S. I wanted to produce a signal akin to VGA as capture cards for VGA are a dime a dozen.
In this case since I don't know whether the CCU gets the sync from green or composite, I opted to use an LM1981 chip to extract the vertical and horizontal signals from the composite out.

And this worked, to a degree.


The monitor happily locked to the signal, and the frequencies are correct for NTSC.
The problem is, the image produced is extremely dim and almost purple.


In person, this is much much dimmer, almost unnoticeable.
While I get there is going to be some nasty noise due to my current construction, I am having trouble finding out why the signal is just not good. Is there something I am missing with the RGB? I am thinking either I need to amplify the RGB lines or I am misinformed on RGB+S vs a VGA signal.

[Alex KL-1]

I wonder that if the monitor is able to correctly display the 480i format (some will, some don't), even showing it on OSD...

[sdsw4]

I connected it to my old ultrasharp (that has composite in), same thing.

EDIT:
So I looked over how the CCU-M5 handles the RGB, and it appears it does indeed do amplification to the RGB sent by the camera.
What blows is I only have some uber 25w power amplifiers and not the tiny ones in the schematic. Oh well, what's the worse I do, fry my monitor?

[ARC Tech-109]

I'm not following what you're trying to do here with the amplifiers. Are you attempting to take this directly from the camera or are you using the CCU outputs? This is very important as the CCU RGB signals are conditioned to drive another 75 ohm source while the camera RGB at the connector is only a low-level of a few hundred millivolts. Another thing to consider is the monitor auto-switching over to a TTL input which is a positive going 5V at that scan rate.
Have you looked at the composite signal directly from the camera? You stated connecting it to your Ultrasharp and got the same result but it's unclear if you were in RGB or composite at this point. Got an old school CRT based monitor available to look at the composite right out of the camera?

[sdsw4]

Yeah with the amplifiers, I'm trying to go directly from the camera, and as you said it's really low level.
The ultrasharp gave the same result with RGB as that other monitor, but the composite from the camera works fine.

EDIT:
So following your advice, I'll take the sync negative that the 1981 outputs and send it through a 74lso4n inverter I found in a tube in my junk drawer.
If only I did more research, I would have powered the 1981 on 5v since I have to work with some amplifiers.
And sadly the only CRT I have is a green screen.

[ARC Tech-109]

I'd strongly recommend finding a CCU to experiment with, if nothing else it would buffer the camera from an accident and give you the correct voltage levels.

[sdsw4]

Yeah it looks like before I get too far down the rabbit hole of spending money on ICs and probably still not figuring it out, I decided to just buy the CCU-M3. Since I already have to carry stuff in the backpack (battery and recording computer), the CCU shouldn't be much more. And I can still power it from the 12v battery. I wanted to buy the CCU-M5 as a supplement, but I don't think there's much use for having that second CCU except if I want to use it on a newer camera, which I'll go there when I go there.

So the least I did was blow $25 on the opamp(which I might have a use for later) and some precision pots.
I still need the sync grabber for the H/V so that wasn't wasted. I just hope I haven't done damage in my experimenting. The green and red output still the ~.2 negative volts but the blue somehow puts out 3v positive. Didn't check that before experimenting though.

[ARC Tech-109]

Do you have the proper camera to CCU cable? Very important as it tells the camera its feeding a CCU. What I'm trying to figure out is what exactly are you trying to do here... using a computer to do some type of RGB digital capture feeding an SVGA port?

[sdsw4]

Pretty much the last statement.
It's akin to what was done here:
https://youtube.com/watch?v=ubIAsmq69L0
Sadly I have to use Windows with this method as both the card used and the other one I'm considering (startech) don't work 100% on linux.
VGA was what I selected as the connector for convenience rather than route it through SCART which I have nothing that can work with that.

This is more or less the cable I have:
https://www.bhphotovideo.com/c/produ...to_14_pin.html
Except about 6' rather than 164'.

[ARC Tech-109]

I'm still wondering why you're going this route to capture in the first place. Plenty of uncompressed capture mediums out there that are far more NTSC friendly such as the digital Betacam format and the Sony DVW-250 portable DigiBeta. Granted its a 4:2:2 with a 10 bit luma sampling frequency of 13.5 MHz and this is going to far exceed the actual resolution capabilities of the M3 on its best day in RGB mode. You will however need to find a way to convert the RGB into either an SDI digital or Y R-Y B-Y to feed the camera connector as very little in the NTSC realm works with the raw RGB signalling

[sdsw4]

Well the first goal was originally to produce a compact way of getting the RGB out of the camera. Which from there a lot of general consumer stuff could be used to process the signal and capture based on cost vs quality.
After realizing my EE knowledge is waaay to little to figure that out I went for the CCU approach. Most of the electronics can be fit in the battery compartment, so after I put it all together then probably I can see with those alternate capture methods.
(Some of which I haven't known about yet, so that's why I'm here!)
But so far, it looks like I'll be hauling that CCU one way or another.

(I really like taking the hard way...)