Baird Television Limited

[peter scott]

Stuck in lock-down I decided to add another page to my website describing the BTL's system as used in the Alexandra Palace trial.

Peter

[ppppenguin]

BTL's film scanner worked very well, rather better than M-EMI's telecine which was basically an Emitron camera looking at a projector. Flying spot scanning, using a CRT as the light source, was used by Cintel ( a successor to BTL) to make some of the best telecines ever.

[peter scott]

Indeed, and no tilt and bend to worry about.

Peter

[ppppenguin]

Near enough perfect geometry too, limited only by the lenses.

Intermediate film may seem like a totally daft technology but it lived on for decades in specialist applications. Though not in real time. It was a substitute for video recording in situations where that wasn't feasible. Notably on space missions where pictures would be shot on film, processed (wet chemistry in space!) and then scanned slowly to send back to earth.

It's astonishing that it was more practical to send a photo lab into space than develop a space-worthy video recorder. But that's how it was done. It's easy to forget how hard video recording was. Also how limited and heavy video cameras were if you wanted high quality imaging. To add to the problems you didn't have the wideband links to earth to send images in real time and sometimes you didn't have line of sight anyway.

I think mechanical scanning is still used for specialist cameras (infra red?) .

Of course there's till a body of enthusiasts such as our own Steve Ostler (Panrock) who do interesting experiments with mechanical scanning.
http://www.nbtv.wyenet.co.uk/index.htm

[Panrock]

You've certainly been gainfully employed in lockdown, Peter!  Another useful resource on your site.

…. at which point, Steve wheels out - again - the blurb on reproducing the Spotlight System he prepared for that Windfall Films programme a few years ago. Still, it's got a lot in common with the Film system and includes data on the 240-line system.  

Whoops. .zip files don't seem to be attachable. I'll try the files as separates.

(TEXT FOLLOWS)

Hi Terry,

A note as promised, with some information (as far as I can glean it) about the 240-line system and the Baird Spotlight Studio.  I hope it may help a little when your team goes to Cambridge.


Overall physical structure and size

⦁ The floor space required is shown on Fig 1 attached. This is based on Simon Vaughan's original diagram, with scaling provided by Martin Allen.
⦁ The barrier between the scanner and booth looks like it may be of steel. Probably advisable!
⦁ The ceiling panel (glimpsed on another photo)  looks to be of board
See also Fig 2 and Fig 3, based on Simon's archive photos, with further annotations by me. These are my best guesses only. They could well be wrong.


Motor

⦁ High speed, high power motor, governed at 6000 rpm.


Lamp

⦁ Carbon arc now obsolete, produces high level of UV. A Xenon Arc may have to replace it. See Fig 4 (ex Wikipedia).
⦁ The high pressure filling makes Xenon Arcs safety-critical to handle.  Such lamps require specialised startup and high power current-source power supplies. Some form of forced cooling might be required.


Photocells

⦁ Huge surface area photomultiplier tubes appear to have been used. Photomultipliers were very new at the time. These would be blue sensitive. They will require high voltage, low current power supplies, potential divider chains and effective ambient light cut-out protection. If this fails, they can be ruined.


Electronics

Valve-based signal amplification and sync processing would have been used.


The 240-line standard

⦁ Please see Fig 5.  The pertinent facts appear to be as follows:

240-lines.
No. of active lines: 220.
Aspect ratio: 4:3.
Frame frequency: 25 Hz.
Line frequency: 6000 Hz.
Frame period: 40.00 mS
Line period: 166.67 uS
Frame sync pulse length: 2.00 mS (12 lines)
Frame back porch length: 1.33 mS (8 lines)
Line sync pulse length: 13.33 uS (8% of line period)
Line back porch length:  3.33 uS (2% of line period)


The Nipkow disc and spiral shutter

⦁ I've worked out some provisional specs for a disc. There's no need to worry about these right now, but they could come in handy later. They will certainly need double checking before the design is worked up in a CAD package. Basically, this is for a 4 x 60-hole spirals disc with an inner ring of holes for line syncs and a separate hole for frame syncs. Light through the holes would presumably trigger monostables with suitable time constants.
⦁ The picture would soon become blurry and jagged if the positional tolerances of the Nipkow holes wasn't perfect. Karen Orton has developed a software correction method.
⦁ Because of the high rim speed, operation in a vacuum would be mandatory. Balance would be critical. Once drilled with Nipkow spirals, the intrinsic balance would be compromised. A separate balancing arrangement would likely to be required.
⦁ The disc would need to be as thin as possible to avoid vignetting on the picture. Such a thin disc would be fragile and could shatter in use.
⦁ In this conception, a spiral-shaped broad slot on another disc in front selects the appropriate 60-line Nipkow spiral. This front disc would rotate at 1500 rpm and be 1:4 geared to the 'main' Nipkow disc. The gear tooth profile would need to ensure silent operation.

Overall disc diameter 915.00  mm (3 ft).
Outer guard band: 7.50 mm.
Number of Nipkow hole spirals: 4 (concentric).
Number of holes per Nipkow spiral: 60.
Sync hole rings: 2  (60 holes for line and single hole for frame - placed to inside of disc).
Max radius of outer Nipkow hole spiral: 450.00 mm.
Min radius of inner Nipkow hole spiral: 434.29 mm.
Mean radius of all four Nipkow hole spirals: 442.15 mm.
Radius of line sync hole ring: 425.00 mm.
Radius of frame sync hole: 410.00 mm
Line segment outermost width (including sync & back porch): 47.12 mm.
Line segment innermost width (including sync & back porch): 45.95 mm.
Line segment outermost width (less sync & back porch): 42.41 mm.
Line segment innermost width (less sync & back porch): 41.35 mm.
Line segment mean width (including sync & back porch): 46.54 mm.
Line segment mean width (less sync & back porch): 41.88 mm.
Total (240-line) raster height: 31.41 mm
.
Thus the picture, off-disc, would measure approximately 42 x 31 mm.
.
Keystoning: 7%.
Aspect ratio of each quarter section (not including sync & back porch): 16:3
Nipkow/sync hole diameter 0.13 mm.
Rim speed at 6000 rpm: 287.46 m/s = 1035 Km/h = 643 mph.
A g-force check will be required to establish whether the stresses in the disc are manageable.

 
Shrouding/structure/pump

⦁ There will be some heavy engineering involved with this project. Efficient vacuum generation and sealing will be needed, and the manufacture of a massively strong bell housing around the disc.


Optics

⦁ A projection lens would be needed, focussed on the disc surface
⦁ Projection magnification would need to be about  x14  to give a 600mm wide scanned area.. The same factor would determine the relationship of distances in front of and behind the lens. The lens would need to be of sufficient aperture to gather the entire light cone diverging out of the Nipkow holes at its operating distance. Generally, the operating distances should be the maximum possible, to keep the diverging light cone narrow. This would also allow a thicker disc.


Ventilation
Safety
Insurance
Time needed!
Cost

⦁ The above are factors I cannot comment on. 
⦁ Finally, fig 6. shows a lady, believed to be Madeleine Carroll, on a Baird Test Card of the time.

Good luck!  :-)
Steve

[peter scott]

Wow! That's great stuff thanks Steve. Did I calculate correctly in suggesting the scanning disc curvature would be less than 3 line widths for the 240 line picture?

Peter

[peter scott]

Wow! That's great stuff thanks Steve. Did I calculate correctly in suggesting the scanning disc curvature would be less than 3 line widths for the 240 line picture?

I think my assumptions are wide of the mark. I only assigned 60 cm for the disc diameter and I assumed that there was no image reduction between the scanning disc and the film.

Peter

[Mark Hennessy]

Whoops. .zip files don't seem to be attachable.

Hi Steve,

ZIP files should be OK. I've just double-checked, and it all looks good as far as I can see. The only limitation is the size, which is currently set to a rather random-looking 5.86MB.

No idea why it's not a nice, round number, but there are other types with odd numbers too, so there's probably some historical reason for that. I should have a think about that at some point in the future...

Was your ZIP file was within the size limit? If not, could you try a smaller one, just to be sure there's nothing wrong with the settings for your account.

Cheers,

Mark

[ppppenguin]

Steve, I remember that production only too well. I don't know how much of your advice they took.

Hugh Hunt, with a team of his post-grad students did the Baird system re-creation: http://www2.eng.cam.ac.uk/~hemh1/tv.htm They only worked at 120 lines. 240 lines would have needed a disc spinning in a vacuum. They used modern photomultipliers and electronics. ISTR the PMT output was digitised and fed directly to a PC. They asked my advice about getting the PC output into the TV production. I suggested an inexpensive VGA to TV converter box; somebody went to buy one. Then they could display on their own PC monitor and send 625/50 to the production team at the same time. It was then bumped up to HD.

I was genuinely surprised at how good the images were, especially when they were moving. Stills were less impressive but that's to be expected.

I was more involved on the M-EMI side, along with Paul Marshall. But that's a whole other story.

[peter scott]

I just came across a little section from Google Books describing telecine development at GEC:

https://tinyurl.com/yboqp532

Peter