Golborne Vintage Radio

Hi.
Curiosity was making me think that if the 1ZH37B easily oscillates at 41 to 45mhz as it does would it work with baseband video? The valve is a doddle to get it to oscillate and using a crystal marked as 45.033mhz it burst into life at 18v HT.
I quickly cobbled up a test circuit with no RF filtering or tuning and fed the output to my Sony TV9-90. I initially fed the baseband video to the second control grid of the valve directly, I was rewarded with a negative picture and on the scope the envelope was badly crushed. A little attenuation resulted in excellent results, a little leaning toward to much HF though addition of an RC filter sorted that out.

The video shows the results but my uploads are not HD and it doesnt do the result justice. Now to include proper RF filtering and a sound stage.
https://youtu.be/r6j98UlMWFU

A lot simpler than Franks's and my attempts at fully digital modulators!

Thanks Jeffrey.
These dual grid Rod Pentodes are really so versatile and easy to work with.
Once I've done more filtering I'll post up the circuit etc.

Is there any form of DC restoration? Obviously if the video is DC coupled all the way from source to the grid of the valve the black level should stay constant.

Hi.
At the moment there is DC coupling from the Aurora although I don't know if the baseband from the Aurora has a cap on the output? There is a little sync crushing, reducing the signal level a little clears the hooking on the top of the picture.

Hi Trevor
Incredible results from such a low component count.
I don't know about the Aurora but the Hedghog video is DC coupled. From memory into 75R the sync tips sits at around 300 mV.

Frank

Very good work.

Adrian

The Aurora output is also DC coupled. I think the sync tips also sit at about +300mV. This will drift slightly with temperature as the output stage is a simple emitter follower.

Another idea showing the versitility of these valves. Good work Trevor!

Tracy

Hi.
Today I removed the choke on the anode and made a tunable RF transformer for the output. This greatly increased the signal output and improved stability. The lash up on the G2 has now been improved which now has improved the starting reliability of the oscillator. The G2 choke is now 470uh which gives optimum results. I then tested this on the Sony TV9-90 and aslo on a Philips TVette, there was still some hooking on the top of the picture. Looking at the envelope it showed signs of sync crushing so I tried out returning the ground of the video source to filament + this cleared up the hooking and the sync was just about perfect, moving the ground to the battery side of the filament dropping resistor improved it with some reserve. This has given G1a a little positive bias, as the grid is very high impedance with a 1m bias resistor to filament - no harm will happen to the source, an AC coupled video source though could be an issue so I'll investigate if this is going to be an issue.
The results are excellent, I've also tested this on my Pye V4 and the TV24C, Signal strength is actually way higher than the Aurora or Hedghog, I needed 48db of attenuation on the TVette before a hint of snow.
I'll update with a video shot earlier later on.