Help me find a cheap solution for capturing pixel perfect real-time video from LCD interface (Tektronix TDS3xxx).
Receiving an image via ethernet is too slow (max 4 frames/sec) and the oscilloscope control is blocked while the frame is being prepared.
It would be ideal if the device received parallel data and output video via
USB UVC (USB Video device Class, Works like a regular web camera).
Signal parameters:
RGB (pixel bus) PixelClock 25,175 MHz, V-Sync 59,94 Hz, H-Sync 31,468 kHz 640×480, only 4 bit per color,
3.3 V CMOS Logic Levels (not LVDS).
You
MIGHT be able to use LVDS to HDMI converter module with dual LVDS connectors.
Might is the key word. Because there is also might not work.
Basically, you connect lvds cable to this board, through an adapter, then through another adapter connect screen to this board, and you get HDMI out.
Doesn't the TDS3000 have VGA output??
Doesn't the TDS3000 have VGA output??
There is a VGA output only if you use the TDS3GV option, and it is very expensive. Theoretically, i can design a board with the same ADV7120 chip, and then connect a VGAtoHDMI converter and capture HDMI with a capture card. But this is too redundant.
https://www.eevblog.com/forum/testgear/reverse-engineering-tektronix-tds3gv-module-for-tds3000-series-oscilloscopes/msg1086337/#msg1086337You MIGHT be able to use LVDS to HDMI converter module with dual LVDS connectors.
In my case, there is no TMDS (Panel Link LVDS) interface. The display is connected via 3.3 V CMOS Logic Level parallel interface.
PS
TFT Module side connector (Header):IL-310-T31PB-VF - Japan Aviation Electronics Industry Limited (JAE)
DF9-31P-1V - Hirose Electric Co Ltd (HRS)
Mainboard side connector (Receptacle):IL-310-T31S-VF - Japan Aviation Electronics Industry Limited (JAE)
DF9-31S-1V - Hirose Electric Co Ltd (HRS)
DF9M-31S-1R - Hirose Electric Co Ltd (HRS)
Favorite solution so far:
github projectBut I’m not sure that I can change it to suit my configuration. Plus this project does not work under Windows OS
Since you have access to a 'DOT-CLOCK', any 3.3v logic DVI transmitter IC will work with a little MCU to initialize the DVI transmitter's I2C registers. Each of the scope's RGB 6 bits should be wired to the DVI transmitter's bits 7 through 2 and bits 1 through 0 should be tied to GND. Don't forget to wire the 'DE'...
Example DVI transmitter:
https://www.ti.com/product/TFP410Note that you may place a HDMI plug on the output of this transmitter for modern convenience.
will require some verilog but .. https://wiki.sipeed.com/hardware/en/tang/Tang-Nano-9K/Nano-9K.html
I would be only too happy to use your option. But I don't know Verilog. I definitely won’t be able to implement Chroma Subsampling 4:2:0, MJPEG encoding, isochronous endpoint and USB PHY on an FPGA.
Example DVI transmitter IC: https://www.ti.com/product/TFP410
Note that you may place a HDMI plug on the output of this transmitter for modern convenience.
There is a high probability that I will choose your option for implementation. It is TFP410 that is easy to obtain and does not require little MCU to initialize the DVI transmitter's I2C registers.
I wanted to avoid developing a printed circuit board, but apparently I will have to do.
Webcam isn’t an option?
No, filming the screen with a camera is a proven option. Low picture quality, long preparation time to set the desired shooting angle. And the camera physically interferes with working with the oscilloscope.
I have seen something similar in Chinese internet segment, they were using either RaspberryPI or OrangePI for that task.
But it was couple of years ago and I forgot the link
I have seen something similar in Chinese internet segment, they were using either RaspberryPI or OrangePI for that task.
But it was couple of years ago and I forgot the link
I have seen this project based on Raspberry Pi Zero + CPLD Xilinx:
https://github.com/hoglet67/RGBtoHDMI/wiki