Golborne Vintage Radio

Had a lightbulb moment this morning. Realised that I wasn't using the correct settings on the interpolator and effectively wasn't intepolating at all. This photo doesn't do it justice. There's no moire and it's sharper than that on the screen.

A few things to check and I can do the final programming into EPROM (yes, EPROM, it's that old).

The only significant modification I might do is to add a facility to genlock the output to colour black. Not a huge job but I won't bother unless it's really needed.

The Scanner Truck / OB unit isn't going to have the old BBC experimental Colour 405 with the variant NTSC?
Yes, I understand even with mono 625 you might want genlock.

Looks good. It's sort of amazing that Laptop LCDs are good enough to be able to say so. I remember the ghastly early colour LCDs and earlier even worse mono LCDs that made you wish for an orange plasma screen.

The photo was taken awkwardly trying to avoid odd reflections on the monitor. Then I reduced the file size from 1.4MB to under 100kB. The 1.4MB version has much less moire. Although GVR will aloow some pretty big files I try to be kind to the storage requirements.

We're never going to originate on experimental 405 line NTSC so at least I don't have to contemplate a decoder for it. I wouldn't be too surprised if the SAA7118 copes with decoding NTSC405, even though it makes a bad job of V/F sync. But I'm never going to bother to find out.

Great result Jeffrey!

Jac

Great result Jeffrey.
It looks really good and it is incredible that you could achieve that with the hardware you already had

Frank

Thanks, Frank.

The only hardware changes were removing the 24.576MHz xtal and fitting a single wire link.

Today I got some time to get back to it and now have 405 to 405 working.
Pictures below from two different sources. Test card C from a free view box via Hedghog and my PM5544 style 405 test card.
Next the interpolater and syncs for 625

Frank

Well done. How are you extracting H and V sync from the 405 input?

Hi Jeffery
The embedded H syncs are good so I am using the SAV signal from them and also using the odd/even signal from the LM1881.
The odd/even signal I use to produce a frame reset signal. When it changes from '0' to '1' a flag is raised. On the next SAV after the flag is raised the frame reset signal is produced and the flag is reset.
Everything else is done with counters.
The lock is good and solid. It does need the frame store to work as the video clock from the TVP5150 is too jittery to drive a PLL reliably.

Frank

Very much the same as I found with the SAA7118. The H sync fromt he decoder is good but the field/frame sync is not. Since the TVP5150 doesn't have the same flexibility as the SAA7118 it looks like the external sync sep is essential.

I can't comment on the clock quality from the SAA7118 - mine is feeding a framestore. I suspect that the clock jitter is rather better when the input is 625 and especially when it's colour. Colour burst is a much better signal than sync pulses for deriving a low jitter clock.