Cogging on Pye D16T

[ppppenguin]

There have been various reports of cogging on Pye D16T (and the electrically identical B16T) when used with Aurora SCRF converters. This is most visible on the builit-in testcard C. This must be due to the sync separator disliking some aspect of the signal, most likely the front porch.

The Aurora SCRF produces a pretty much immaculate waveform but there are some slight variations between versions. I haven't had a chance to measure a v1 unit but got slightly differing results from versions 2, 3 and 4. The current hardware build is v4. The horizontal parameters are actually set by the firmware which also differs for each hardware version.

Notably my v2 unit has a very long front porch, 2.85us and that didn't give cogging on the Vintage Wireles Museum's D16T. The correct front porch is 1.7us, with a tolerance range of 1.5us to 1.9us. I've seen mention of front porch ranging from 1.5us to 2us. Current v4 units (firmware build 4.2) have a 1.85us front porch when used with live video. With the factory programmed testcard it's 2.55us suggesting that the TCC used for programming has timing errors. I will try reprogramming my v4 unit with my lab 625 line TCC. I've measured this and it's spot on accurate for timing.

There is more work to be done. One day when I get round to running up my B16T I'll investigate in more detail.

[Geordie McBoyne]

Hi Jeffrey,
             an example of picture cogging on the screen of the GEC BT310.   The problem was caused by the 0.5microfard grid coupling capacitor having gone leaky.
in fact the design of the Pye B16 sync separator is similar to the GEC BT310 and BT1156. The pentode sync separator valve conducts on the negative going sync tips resulting in positive sync pulses being developed across the anode load resistor.
The test card image generated by an Aurora.

Geordie McBoyne.

[ppppenguin]

Thanks for the illustration of cogging. The cogging on the Museum's D16T was never anything like as bad as that, perhaps about the width of a vertical white line in TCC.

[Geordie McBoyne]

Correction to my last post.   I stated that the sync separator valve conducts on the negative going tips of sync pulses.   This is of course incorrect. The sync separator valve conducts on the positive going tips of the negative going video waveform and the syncs at the anode of the sync separator are negative going.
The attachments show the negative going video present at the anode of the video amplifier.
The sync pulses are somewhat compressed. Note also peak white to black level before the sync front porch period is not instantaneous.

Geordie McBoyne.

[Murphyv310]

Hi.
I've been mentioning this for years TBH on the UKVRRR, VRAT and now here.
This causes foldover on the LHS of the screen on some flywheel sets and cogging on direct sync sets, I have three Aurora's of different vintage and all give different results regarding this problem.
The RT Russell test card generator is the only one with the correct timings and doesn't have the issue.

[ppppenguin]

Can you let me know the measured front porch, sync pulse and back porch widths for the RR generator.

[ppppenguin]

In the recent thread about the Pineapple converter I came across a BBC EID document from 1967. This is very late in the day for 405 as the changeover to 625 was happening and the 405 specification would have been settled for some time.

Front porch 1.7us. This is what we believe is the right value to use

H sync width 8-10us. We think that 9-9.3us is right and I don't think it's too critical

H blanking width: 18.5us. This is what we believe is the right value to use

[Geordie McBoyne]

Hi Jeffrey,
18 to 18.5microseconds for the line blanking period is correct. I understand the original active video was 85 microseconds and later on after 405 line TV was result of conversion from 625 the active video become 79 to 80 microseconds.
Very old manuals define the active video as being a fractional value as 17/20.

Geordie McBoyne.

[ppppenguin]

625 TV has active line of 52us or 81.25% of total line. With a simple conversion to 405 this would give active line of 81.1us with blanking of 18.7us.

The old 17/20 fraction gives active line of 84.8us, 15us blanking.

These things took a while to settle down. Initially I'm not sure they were even regarded as too critical in the early days.