Getting a frame pulse early !

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Steve Anderson, Dave Moll, Andrew Davie

Getting a frame pulse early !

Postby Viewmaster » Sun Nov 18, 2012 7:57 pm

I chuck this idea into the ring and see if it rebounds. (I'm sure it will!)
I haven't seen it mentioned before, although it probably has.

As I read it, the standard method of recovering the missing frame pulse is, as has been said, 'The missing the bus' method......... setting a multi vib to be retriggered at each line and it only finishing its cycle when the line pulse is missing at line 32, and so giving the frame pulse.

Suppose this recovered frame pulse triggers yet another multi vib whose timing was one whole frame, 80ms, but whose timing could be adjusted a few ms, or parts of a ms, either way, by a low value pot in the RC timing cct.

Using this o/p from the 80ms vib. as the new frame pulse, one would not miss the bus if the timing of the 80ms multi vib were set very slightly less.

So, one is using a previously recovered frame pulse to give one the present pulse, as it were, and its timing is adjustable plus or minus to suit. I cannot see a problem with that, but I daresay that there is. :cry:
“One small step for a man,"......because he has Arthritis.
Albert.
User avatar
Viewmaster
Frankenstein was my uncle.
 
Posts: 1288
Joined: Fri Apr 06, 2007 4:50 am
Location: UK Midlands

Postby dominicbeesley » Mon Nov 19, 2012 4:11 am

Perfectly sensible way of doing it...

I started doing a flywheel sync unit for mine on a similar principle. The idea being to try and put clean syncs back on a (very)noisy signal. It is buried somewhere under fifty other half finished projects!

A 4046 and a handfull of components should do it plus a monostable to give you a frame pulse rather than a full half-frame.

D
User avatar
dominicbeesley
Anyone have a spare straightjacket?
 
Posts: 683
Joined: Sat Sep 13, 2008 11:32 pm
Location: Hebden Bridge

Postby Viewmaster » Mon Nov 19, 2012 6:30 pm

Hi Dominic. I think that in my suggestion, the main problem is that each mono vib has innacuracies in the timing due to RC changes with tempereature etc.

Peter Smith used a very 50us stable IC ( 4538) but using two of them
(2.6ms + 80ms) willl double timing error so the final frame pulse may not be stable enough, time wise.

Maybe, a way around this would be to just use ONE 80ms (+-) multi vib. Then when it is triggered by the last line pulse before the missing one, mask it off from any further line pulses until it is allowed to finish its cycle to give the frame pulse.
Then repeat its triggereing with the line pulse at the start of a new frame. Masking maybe done with another vib in sort of parallel, as it were. But there again, a long time 80ms vib might be subject to error where the shorter 2,6ms is not so much
..........I do wish I knew what I was talking about. :lol:

Flywheeel sync, had a brief one line mention in the handbook, but I don't recall any ccts anywhere suitable for NBTV.
“One small step for a man,"......because he has Arthritis.
Albert.
User avatar
Viewmaster
Frankenstein was my uncle.
 
Posts: 1288
Joined: Fri Apr 06, 2007 4:50 am
Location: UK Midlands

Postby Steve Anderson » Mon Nov 19, 2012 8:44 pm

Viewmaster wrote:the main problem is that each mono vib has innacuracies in the timing due to RC changes with tempereature etc..


Albert I think you're worrying about timing drift with temperature a little too much. If a a metallized Polyester film capacitor (PET) and a metal-film resistor were used for the timing components the change in pulse length is likely to be a lot less than 1% over normal room temperatures.(10-30 Celcius).

The low resistance (as a percentage) 'framing' pot would also allow adjustment for this.However this method does require a reasonably close to 400Hz line-rate.

More important is supply stability, especially when using the CD/HEF series of monostables, the supply would definitely need regulating...but more often than not they are these days. The 78xx type being perfectly adequate for this in conjunction with good supply bypassing/decoupling.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3950
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby Viewmaster » Mon Nov 19, 2012 11:59 pm

Steve Anderson wrote:
Viewmaster wrote:the main problem is that each mono vib has innacuracies in the timing due to RC changes with tempereature etc..


Albert I think you're worrying about timing drift with temperature a little too much. If a a metallized Polyester film capacitor (PET) and a metal-film resistor were used for the timing components the change in pulse length is likely to be a lot less than 1% over normal room temperatures.(10-30 Celcius).

The low resistance (as a percentage) 'framing' pot would also allow adjustment for this.However this method does require a reasonably close to 400Hz line-rate.


Yes on thinking about some more, I now see your point about the reqirement to be close to the 400Hz rate.

But if the sync rate was not exactly 400 but stable, then a slight adjust of the fine pot might still give a reliable frame pulse, without 'missing the bus' :lol:
It would be interesting to see on the Niptrix the picture change as the pot were slightly adjusted either way. Rolling sideways I would guess.

OK also on your mention of the best R and C to use.....I was wondering about changing my existings to these anyway if problems arose with my existing cct setup.

Gee whiz, Steve, I am so happy that you and others, (so far !!) haven't grounded my idea for a 82.5 ms multi vib with +- frame timing adjustment.
“One small step for a man,"......because he has Arthritis.
Albert.
User avatar
Viewmaster
Frankenstein was my uncle.
 
Posts: 1288
Joined: Fri Apr 06, 2007 4:50 am
Location: UK Midlands

Postby Steve Anderson » Tue Nov 20, 2012 12:36 am

It's a valid method of frame sync detection with the proviso that the 80-odd ms mono can be adjusted.

Some time ago I thought of the idea below. Instead of using a long-duration mono a counter is employed which when it reaches 31 fires of the delay mono which in turn fires off the frame sync mono.

The 'Missing the bus' mono triggers a further one which resets the counter to zero during line 1. This will allow for much wider variations in the line-speed and no concerns about temperature stability.

Timings are nominal.

Steve A.
Attachments
Counter Method 1.gif
Counter Method 1.gif (3.54 KiB) Viewed 5348 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3950
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby Steve Anderson » Tue Nov 20, 2012 1:54 am

I've noticed that 8-input AND/NAND gates are getting a bit rare these days so a suggested alternative is below...note following monostable input is changed too...

Steve A.
Attachments
Counter Method 2.gif
Counter Method 2.gif (3.36 KiB) Viewed 5346 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3950
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby Viewmaster » Tue Nov 20, 2012 7:17 pm

Thanks for the cct, Steve.
What with your cct, Peter Smith's and my own little idea of masking a
82.5ms +- vib with another one I should certainly be able to,
'catch the bus.' :)
“One small step for a man,"......because he has Arthritis.
Albert.
User avatar
Viewmaster
Frankenstein was my uncle.
 
Posts: 1288
Joined: Fri Apr 06, 2007 4:50 am
Location: UK Midlands

Postby Viewmaster » Tue Nov 20, 2012 7:52 pm

Steve, on further looking at your cct I notice that the 4024, 31 counter is reset by the first mono timed at 3.75ms with a short pulse from the second mono (100us)
Forgive me but I cannot understand this.

Shouldn't the 4024 counter be allowed to count 31 line pulses, which is 77.5ms, before being reset?

I know I probably have this wrong but cannot see how. :cry:
“One small step for a man,"......because he has Arthritis.
Albert.
User avatar
Viewmaster
Frankenstein was my uncle.
 
Posts: 1288
Joined: Fri Apr 06, 2007 4:50 am
Location: UK Midlands

Postby Steve Anderson » Tue Nov 20, 2012 8:43 pm

Viewmaster wrote:Shouldn't the 4024 counter be allowed to count 31 line pulses, which is 77.5ms, before being reset?


There are only 31 incoming sync pulses per frame (if there were 32 the AND/NAND gates could go) The last one signifies the end of line 31, the start of 32. I hope the sketch below makes things clearer...remember 0 is a valid binary count, unlike humans who start at 1.

It is entirely possible I've made a complete cods of this...It was an idea some time back and stored in my head which has many bad sectors.

Steve A.
Attachments
Counter Method 3.gif
Counter Method 3.gif (8.61 KiB) Viewed 5314 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3950
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby Viewmaster » Tue Nov 20, 2012 11:21 pm

Steve Anderson wrote:It was an idea some time back and stored in my head which has many bad sectors.
Steve A.


My head is more like a floppy :) , but I shall try to fathom
your explanation, and sketch, Steve.
Thanks for going to all that trouble in drawing it up.
“One small step for a man,"......because he has Arthritis.
Albert.
User avatar
Viewmaster
Frankenstein was my uncle.
 
Posts: 1288
Joined: Fri Apr 06, 2007 4:50 am
Location: UK Midlands

Postby M3DVQ » Wed Nov 21, 2012 12:28 am

Steve Anderson wrote:remember 0 is a valid binary count, unlike humans who start at 1.


Pfft, most humans do :wink:
M3DVQ
Just nod and pretend you understand me
 
Posts: 328
Joined: Sat Feb 10, 2007 10:52 am
Location: Lincolnshire

Postby Steve Anderson » Fri Nov 23, 2012 2:17 pm

In the background Albert and I have 'tweaked' this counter method a bit. For the NIPTRIX it requires a continuous 400Hz line sync, this would also apply to CRT based displays too (sorry Gary).

A simple change to the AND gating and the result is shown in 'Counter Method 4' below.

To be able to separately tweak both the generated frame sync and the inserted line sync the circuit in 'Counter Method 5' provides this.

The inveterate pot-twiddler will have plenty to do during those cold winter evenings.

Steve A.
Attachments
Counter Method 4.gif
Counter Method 4.gif (3.61 KiB) Viewed 5249 times
Counter Method 5.gif
Counter Method 5.gif (5.25 KiB) Viewed 5249 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3950
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby Viewmaster » Fri Nov 23, 2012 6:18 pm

Yes, and thanks for that Steve.

Not only does that give me plenty of leeway to get the timings right but I can now experiment to see how the picture is changed.

I have line and frame adjust , just like the old 405 line TV's.........now all I want are focus, width and height controls. :)......
........and Sylvia Peters :shock:
“One small step for a man,"......because he has Arthritis.
Albert.
User avatar
Viewmaster
Frankenstein was my uncle.
 
Posts: 1288
Joined: Fri Apr 06, 2007 4:50 am
Location: UK Midlands

Postby gary » Fri Nov 23, 2012 7:52 pm

Viewmaster wrote:Y.....and Sylvia Peters :shock:
Attachments
We_Bring_You_Live_Pictures_--_On_This_Historic_Day2.avi
(6.26 MiB) Downloaded 335 times
We_Bring_You_Live_Pictures_--_On_This_Historic_Day.wav
(5.92 MiB) Downloaded 338 times
Perfecting an NBTV system is like trying to slam a revolving door...
User avatar
gary
"Fester! Don't do that to 'Thing'"
 
Posts: 2303
Joined: Sat Jan 27, 2007 11:29 am
Location: Bundanoon, Australia

Next

Return to Mechanical NBTV

Who is online

Users browsing this forum: No registered users and 5 guests