405 lines from Raspberry Pi 5

[ppppenguin]

Frank, thanks for reminding me how that works.

To get the audio right you really do need a scope with sufficient bandwidth to see the modulated carrier.

[Mike Watterson]

And faster Pi5 storage
https://www.theregister.com/2024/05/15/r...pi_m2_hat/

[OldTVEnthusiast]

I’d completely missed the requirement for DC bias, which explains the low sound despite there being more than adequate output from the DAC. So what is happening, is the modulator AGC reducing signal level to an expected sync level and the DC bias lifts the signal out of the expected sync area?
Anyway I’ll breadboard a DC bias stage before going for a version 2. I do have the scope but not a spectrum analyser.

[ppppenguin]

Simply that the video input and expects a particular voltage for zero carrier and the directly coupled audio needs bias to give that. I think!

[`FRANK.C]

HI Mike
In the 2 threads I linked to in post #100 you may find an explanation.
As far as I remember the DC bias disables the sync clamp within the chip and sets the unmodulated carrier level.

Frank

[ppppenguin]

After I posted I was wondering about the DC restoration (not sure if it's simple DC restoration or a keyed clamp) and how it was disabled to make the video input work with audio. What i don't remember is why everyone used the video input for audio. The chip can do AM sound so why not the audio input? I'm sure there was a good reason.

[Mike Watterson]

That's why two chips. The Sound input is using NTSC/SECAM/PAL offsets, not the lower System A.
Though there is a version of the chip where you can feed in a subcarrier of any frequency, (like Nicam), but you'd need a separate 3.5MHz AM LO.

[ppppenguin]

I know that's why you need two chips. But since they are programmed separately why can't the one used for sound just be programmed to the correct frequency? Or does it need to have a vision carrier present in order to do sound at all?

[`FRANK.C]

Hi Jeffrey
Yes that is the reason. For sound you need the vision carrier as well on those chips. But the sound carrier can be disabled.

Frank

[Mike Watterson]

Though the chips have two "carriers", one is created by modulating a subcarrier and then mixing it with a 2nd RF carrier same frequency as video carrier. So with the chip that has a direct aux in to by-pass the subcarrier AM or FM, you have to supply your own 6.5 MHZ digital subcarrier or 3.5 MHz AM. Presumably the block diagram doesn't show all the details of the RF end.

No doubt it's done this way so there is only one high frequency input divide by N. The UHF VCO seems to run at twice UHF carrier. The VHF being simply the UHF frequency divided by 2, 4, 8 or 16, hence squarewave out. I imagine close to a squarewave at UHF too, but there seems to be a UHF low pass filter.

The mixer for sound sub-carrier is likely a balanced kind, maybe even more complex to cancel the other side band.