Yeah, I posted about this a while ago, but just now discovered the
Raspberry Pi's DPI module. The timing chart [
http://jb-electronics.de/tmp/timing.jpg ] does not mention inverted
pulses, though. Do you have any intuition what could go wrong if you use
3.3V p-p signals instead? Could this expl
Here is the manual: http://jb-electronics.de/tmp/TR-60S1A-manual.pdf
To add to the mystery, I should say this: I managed to get the display
to work with a composite signal and a GS1881 sync separator to extract
the hsync and vsync pulses. I know that the timing is a bit off, but by
tweaking th
I get the feeling I have a very similar crt in my collection, but I also
think I mentioned this a while back on here, have you posted already about
this? I think mine is either Toshiba or Panasonic and was a unit often used
in T&M kit, Tektronix protocol analysers come to mind? I had mine runnin
Jens,
The CRT monitor may not be adjusted properly. The horizontal oscillator
probably has a frequency control. This what we called Horizontal Hold on TV
sets.
Do you have any documentation such as a schematic, service manual or
similar for it?
On Fri, Feb 25, 2022, 9:34 PM jb-electronics
wrot
PS: Also the video data's timing is spot on. Silly question: The Pi runs
at 3.3V, the display expects TTL (5V). Could this lead to these
distortions? I have not yet added a proper level shifter.
On 2022-02-25 11:34 p.m., jb-electronics wrote:
Hey David, I had a few minutes spare time tonight an
Hey David, I had a few minutes spare time tonight and I verified that
both the hsync and vsync signals appear at exactly the right frequency
(19.2kHz and 60Hz, respectively) and have the correct duration. Yet, the
image is still garbage. How can that be, any ideas?
Jens
On 2022-02-21 7:59 p.
Jens,
Connect your oscilloscope to the H Drive signal and verify that it is 19.2
kHz. If your oscilloscope cannot do this job, then do yourself a favor and
buy a better one. It will be worth it.
I say this because I did a lot of video work in the 1980s, and a Tektronix
465 scope or its equivalent w
Nick, nice decoding of the numbers there. I can set the pixel clock to
110ns per pixel (9.09 MHz), but since this is a free parameter in the
Raspberry Pi it should also work with the 20MHz that I originally
specified, right?
Best wishes
Jens
On 2022-02-20 10:52 p.m., Nick Sargeant wrote:
Comp
David,
Thanks for your pointers! In the Raspberry Pi's DPI configuration I set
the amount of horizontal pixels to 844px, which at a pixel clock of
20MHz should result in 19.2kHz if I am not mistaken. Sorry, I forgot to
mention this part.
Regardless of how I think I configured it, you are of
Comparing the timings in the picture and going for common denominators, you
can see that the front porch and back porch are (4 units of 660 ns) long,
hsync is (7 units of 660ns) long, video itself is (64 units of 660ns) long.
So I would offer that there is a 64 character display, where each cha
Jens,
The horizontal line timing is incorrect. This is why the image is diagonal.
The numbers that you provided do not specify the horizontal line total time
or frequency.
The monitor requires 52 microseconds or 19.2 kHz.
You need to find a way to specify this, perhaps it is the character count
or
11 matches
Mail list logo
