QL pico VGA, Need troubleshooting help

[bola_dor]

I come from.
viewtopic.php?p=50691#p50691
trying not to get in the way of that thread

I built my own board using a conventional Raspeberri PI pico, the long plate.

and I got this kind of result,

please click to see in motion

i have checked the connections, the resistor values at both the input and output as well as the voltage divider arrangement at the inputs. i also made a shield of the whole thing with aluminium foil connected to gnd. nothing worked.


Today I did all the measurements again with the oscilloscope using the blue channel, for that I ran a basic line using the composite output.
this is the screen whithout the composite output attached to it https://www.dropbox.com/s/nqw6n1s66br9w ... 3.mp4?dl=0
and this is how it looks when I plug the TV to the composite out https://www.dropbox.com/s/rnwsk1hz5drnq ... 2.mp4?dl=0

this is how the blue channel looks at the QL output

an this is at the VGA output

input is noisy but TTL levels at 15,6Khz
output is 0.6v I think as spected and 48Khz

I doublechecked resistor values and resistors, in fact pico inputs are about 3.3v

the fact that attaching the composite output changes the noise pattern sugests me that something is wrong with the Ql grounding,
Its a UK ISS 7 Ql, board is original and using the original PSU.
this is how it looks like at my 1989 NEC, and why I used a DB9 at the QL output ant then to the pico.

[ones' complement]

The junction of all your 680ohm resistors don't appear to be at a ground (OV) pin on Pico. Either way you insert the Pico they appear to be connected to GP0 or GP16 ? Could we see underside of strip board to confirm the apparent wiring from the topside?

[Maskenlos]

As mentioned earlier, could you kindly share Osci-Picture of the input pins of Raspberry board?
Means GP3, GP5 and GP7 towards ground.

Thanks!

[Peter]

If you have a second RPi Pico module, please check if it has the same issue.
If you don't, please buy one, because that is cheaper than the time needed for debugging a possible module fault.
Is the oscilloscope ground floating, or could it implement a ground loop?

[bola_dor]

The junction of all your 680ohm resistors don't appear to be at a ground (OV) pin on Pico. Either way you insert the Pico they appear to be connected to GP0 or GP16 ? Could we see underside of strip board to confirm the apparent wiring from the topside?

Hi, they are connected to the GND pin at the input DB9, (same as QL video output pin 2)
Should I conect them to the Pico gnd?
The grey/red wire underside

[Maskenlos]

I guess you are missing the ground link. See schematics from Peter:

Note: the pin numbers of the Pico are different here.

Best regards,

Stephan

[Peter]

Hi, they are connected to the GND pin at the input DB9, (same as QL video output pin 2)
Should I conect them to the Pico gnd?

All grounds should have solid connection. Ideally by a ground plane, which is why a PCB is better.

[ones' complement]

The ground input from QL (red/gray wire) appears to only go to junction of 680ohm resistors. The ground should be common to all three devices QL-PICO-VGAout. Perhaps I've missed where you may have linked other D connector ground pin from QL to Pico, but I would also insert link wire from resistor junction to a ground pin on Pico.

If you look at wiring diagram on https://github.com/holmatic/video_if_ql_vga you will see QL ground input is also connected to Pico pin 8 (Gnd).

[Peter]

If you look at wiring diagram on https://github.com/holmatic/video_if_ql_vga you will see QL ground input is also connected to Pico pin 8 (Gnd).

But a bit hard to see in the drawing. Maybe bola_dor overlooked it?

[bola_dor]

OK, I followed exactly this wiring but using a DB9 as input as I usually use a DIN->Db9 adapter to my OLD CRT.
I 'll ad a wire Linking all three GND (all coming from different PSU, that scares me a little) and try again.
I don't have another pico at hand but I'll get one or two. They are cheap enough to have them...