Page 1 of 2

Valve Baird Televisor circuits for Marcus et. al.

PostPosted: Mon May 05, 2008 1:06 pm
by chris_vk3aml
Hi folks!

As many people seemed interested in suitable valve circuits for NBTV, I thought it would be of interest to republish a convenient circuit of this type, originally shown in "Ideas Sheet 6" way back in 1973. The valves are common and uncritical - 12AT7 or 12AX7 preamp; a 6CB6 or similar pentode voltage amp; and a 6L6, an 807 or a 6V6 beam pentode in the output. These are all relatively common valve types, at least they are in Australia - don't know about France.

The output stage could be modified for LEDs by placing a power resistor of about 6K in the anode of the output stage, placing a 500 ohm power resistor in the cathode circuit of the output to earth, and bypassing the anode to earth with a 1uF capacitor. The LEDs could then be driven by being connected in series across the 500 ohm cathode resistor, with a suitable current-limiting resistor in series with the LED bank, as shown in the Austrian circuit previously posted.

The aperture correction circuit in this was a bit of an oddity of my own devising. The speaker transformer with the shorted low-impedance secondary MUST be of SMALL size - in fact it must be a BAD and cheap speaker transformer with limited frequency response. This device relied on the transformer's leakage inductance, which would virtually short the output for low and medium frequencies, but give a terrific, rising hf boost up towards 10 KHz - a near-ideal curve for aperture correction. There are better ways of doing this, of course, but this seemed to be "near enough".

The article by H J Barton Chapple is a good fin-de-30-line article written in 1935 by one of the best television writers then active. It's a comprehensive resume of 30-line construction practice and I hope this will be inspirational, particularly to younger readers building the devices from a historical perspective.

The Barton Chapple article originally appeared in Newne's "Television Today", volume 1, pages 19 to 28. I've reduced each page to .gif format to retain readability of the circuit diagrams. Each page occupies about 150K of bandwidth, and I HOPE that I haven't offended the moderator by placing them here, at the position of easiest public access. If you feel that these would be better in the "Patents and Articles" section, Andrew, feel free to move them across.

Your enquiries inspired this posting, Marcus, so I hope they'll help you with your reconstruction of Baird's gear with valves.

Hoping that this is of interest,

All the best,

Chris Long VK3AML.

PostPosted: Mon May 05, 2008 1:18 pm
by chris_vk3aml
Just a note - an erratum - on the disc televisor receiver circuit on page 23 (.gif number 5) of the Barton Chapple article, above.

This contains a circuit omission which I actually corrected and included in my previous reproduction of the circuit under the subject heading of "valve televisor circuits". It's a necessary 0.1uF capacitor from the junction of the neon lamp and synchronising coils to earth. If this is not included, the inductance of the sync coils will "choke out" the higher video frequencies and not permit the neon to reproduce any detail in the picture.

Yes, publications were just as careless 75 years ago as they are today (!)

Happy days,

Chris Long VK3AML.

PostPosted: Mon May 05, 2008 1:35 pm
by chris_vk3aml
A further erratum of my own - when the Ideas Sheet 6 circuit is modified to drive LEDs with the output stage as a "cathode folower", the cathode resistor should be UNBYPASSED! Remove the 100uF capacitor from across the cathode resistor in this configuration, otherwise the video signal is working into a virtual short circuit.

(Slightly embarrassed),

Chris Long VK3AML.

PostPosted: Mon May 05, 2008 5:13 pm
by Marcus
Thanks Chris,
That was a good read.
I will be using your circuit as you have done all the work for me! I have a whole stack of 12ax7 & 12at7 in my junk box and also managed to find a couple of 6CB6. The closest I have to a EL37 is a EL36, would that be a suitable replacement? Otherwise I have a few other pentodes including 6BY7.
I also vaguely remember pulling an output transformer from dodgy valve radio a few years back, I'm sure that is around somewhere.
I found a nice little circuit for a shortwave radio a couple of days ago too. I'm thinking of putting it to use in my project. I would use the section with the 12ax7 for reception and feed it into your amplifier.

PostPosted: Mon May 05, 2008 6:39 pm
by Viewmaster
Marcus, have you seen these photos of vintage valve NBTV built by Roger Dupouy?


PostPosted: Mon May 05, 2008 9:13 pm
by Marcus
Wow, thats nice.
It looks like he used 1930's tubes too. The problem is those can be hard to get and can be expensive. I was thinking of using an array of flat faced neon indicators used in nixie clock projects for the neon lamp.
I'm not sure how they will go but for the price I think I'll get some to try.

PostPosted: Tue May 06, 2008 2:50 am
by Viewmaster
Marcus wrote:Wow, thats nice.
It looks like he used 1930's tubes too. The problem is those can be hard to get and can be expensive.

Many bog standard 1930/40 valves are not too dear, especially if servicable used. Try searching on eBay from time to time.
If in UK Midlands Chas Miller holds monthly auctions in which old valves appear very frequently (mains and battery).

I would like to build a NBTV using bright emitters but my piggy bank is too small :lol:

PostPosted: Wed May 07, 2008 7:46 pm
by chris_vk3aml
The problem you face, if you are REALLY trying to make an "authentic" erzats replica of a Baird televisor is simply: HOW MUCH ARE YOU PREPARED TO PAY FOR IT? The valves I've specified will give you a good working model, and the beam pentode output stage, driven as a cathode follower, will probably draw enough anode current to drive LED's across the cathode resistor. These valves are typical of the 1950s, however, and by the strictest historical standards are not contemporary with the Baird transmissions of 1928 - 1935.

However, if you chose to use rarer, more expensive 1930s output tube like an American type '45 or type '50; or the British types used by Baird such as the DET1A, LS5 or LS5A you will run into several problems:

(1) much lower transconductance
(2) much greater Miller effect capacitance and potential high frequency rolloff.
(3) Lower dissipation ratings than the 807 or 6L6.
(4) Lower maximum standing current (generally around 50mA) which may not be enough to cathode-drive a bank of power LEDs to peak current, and which certainly will not be sufficient to drive a bank of Luxeons.

These could certainly drive neons but they may not have the reserve standing current to fully drive high-output LEDs. Incidentally, a bright emitter tube would also be inappropriate for use with most Baird equipment, as they were "on the way out" soon after Baird gave his first public demonstration in 1926. In the halcyon days of the BBC-Baird 30-line transmissions, September 1929 to September 1935, bright emitter receiving tubes were quite obsolete.

So then we're back to the original, main problem - how do we get something to either BE or LOOK LIKE an original television neon?

The original cathode-glow neon lamps specifically designed for television, with a flat nickel cathode plate about 3 cm square, are now, quite without exaggeration, as rare as proverbial "hen's teeth". They were not manufactured for more than a brief period (1927 - 1937 at the outside) and very few people bought them. In my 40 years of interest in this subject I have seen THREE in the whole of Australia, and I HAVE BEEN LOOKING! One was privately held, one was in the Sydney Powerhouse Museum's original 30-line Televisor, and one loose example was in the Powerhouse Museum's valve collection.

Melbourne's Science Museum of Victoria (now part of Museum Victoria) assembled a mechanical television exhibit in 1971-72. The veteran engineer Gilbert Miles (b1904 - d1981, former NBTVA President), in charge of Melbourne transmissions of Jenkins-style telecine over station 3UZ in 1929, teamed with the Museum's Curator David Turner to construct the combination Nipkow camera-monitor. This had two Nipkow discs assembled on a single shaft, driven by a double-ended washing machine motor - one disc as a camera-scanner and the other as a monitor. Scanning was from 35 mm transparency slides projected onto the "camera" disc.

Turner approached the neon sign manufacturers, Claude Neon, with the idea of making re-created flat plate neons for the exhibit. On the surface of it, this seemed like a simple proposition - or so they thought. As a 17 year old schoolboy, visiting the Museum's workshop on almost a weekly basis to watch these preparations, I saw how complicated the "simple" job of making a "television neon" could be.

Turner obtained a number of old VHF tetrode tubes, type QQVO6/40s, which had an envelope and plate mounting wires of approximately the right proportions to accommodate the 1 1/2 inch square nickel cathode plate. The valves, of all-glass construction, carefully had their bases cut off and were gutted of their original cathodes, heaters, plates and grids. The polished nickel cathode plate was then crimped to the former anode wires, and an "anode frame" or square of wire the same shape as the anode's periphery was set a short distance in front of the cathode plate and anchored to the former cathode, grid and heater wires/pins of the valve base.

The metal parts were then scrupulously cleaned, the parts were sent to Claude Neon's glassblowing shop, the glass base was worked back onto a seal with the original glass valve envelope and the tube was evacuated through the original exhaust pip. With some 200 volts applied to the neon lamp that they'd made in this way, the experimenters then slowly admitted enough neon to initiate a discharge, and then more gas to reduce the striking voltage to a minimum. That being done, the tube was sealed and Turner et al retired for some preliminary self-congratulation.

Somewhere here, I have a photo of Gil Miles examining the neon lamp before its failure.

The bad news was that while this tube initially worked exceptionally well, within an hour its light output was well down and its striking voltage had risen from 90 to 300 volts. To make matters worse, the cathode plate progressively became unevenly lit. The problem was that neither Turner nor the guys at Claude Neon had any means of rf-heating the metal parts while the tube was being exhausted. Without this precaution, gas (probably oxygen) occluded into the metal surfaces would slowly "poison" the discharge as the metal parts were warmed by the discharge's ionic bombardment. After a couple more tries, Miles and Turner gave up the approach. It would be OK if they had access to professional valve manufacturing gear, but there was nothing of that type then in Melbourne.

Instead, Miles eventually used much smaller indicator neons - of the type you suggested, Marcus - in association with a very short focal length condensor lens system to optically magnify their source area. This was not an ideal solution, as it greatly restricted the angle of view of the image. Only one person could view it at a time, and even then with his/her head carefully positioned - which many Museum visitors never managed to master. Nevertheless this display continued on exhibition at the Museum's Swanston Street site until about 1984, when after many neon and valve replacements it went into storage.

And there, I understand, the replica remains - sadly.

I really feel that the Museum did not keep faith with either Miles, Turner or myself in doing this, and in more recent years Museum Victoria has turned increasingly away from the history of technology. When the "official" 50th (actually 77th) anniversary of Australian TV occurred in 2006, the Miles replica was somewhat conveniently forgotten. How would it have looked for Australian TV networks sponsoring a commemorative exhibition if they had to acknowledge that they were NOT the first in the field? Am I cynical about all of this? Of course! I subsequently worked at Museum Victoria! SAY - NO - MORE !

Anyway the time may have come where there are a sufficient number of US to actually GET SOME TV NEONS MADE - possibly as a group purchase from a current valve manufacturer like Sovtek in Russia. I am sure that our Eurodollars, English pounds and American dollars would look good to them!

Just a thought, fellas! Any takers? Is this nuts, or does this have potential?

All the best,

Chris Long, VK3AML.

PostPosted: Wed May 07, 2008 8:28 pm
by Marcus
I'm quite happy with not using 1930s valves all I want to do is make a unit with not a trace of silicone anywhere in the circuit, strictly valve based.

For this reason I'm more than happy to use the valves that you have listed as I have all of them on hand (except the EL37, which i'll sub for an EL36 I have).

I have contacted the online store about those neons, I'm going to buy quite a few and make a decent array. Approaching Sovtek sounds like a good idea, WHY NOT? The worst they can do is tell us were crazy!

Baby steps for me at the moment though, I'm going to get a start on your schematic.
I have got the motor ready for a phonic wheel and salvaged a 240v coil from an old airpump to see if it will suit the purpose. I've decided to go for the octagon look with a built in SW receiver for NBTV signals and another for audio signals.

PostPosted: Thu May 08, 2008 12:55 am
by chris_vk3aml
Hi again Marcus,

I have just looked at your profile and I see - to my absolute amazement - that you're right here in Melbourne!

I could for transmit NBTV on the 160 metre amateur band - 1840 to 1870 KHz, but only low power - could easily build a linear amplifier to raise that to about 60 watts if you're not too far off. If you're not too far away I could try radiating 32 line signals for you - I am in Surrey Hills, directly East of the city, about 12 kms from the GPO. Are you far from me?

All the best,

Chris Long VK3AML

PostPosted: Thu May 08, 2008 2:10 pm
by Marcus

I'm in South East of Melbourne in Bonbeach. I used to give an estimate of distance, roughly 35km from you. What distance do you think you would get with 60w? I could try to use a regular am tuning coil with the right capacitor to get up into that band.

I ordered an aluminum nipkow disc of Vic today, it hurt the pocket a little but it will be the most expensive part of this exercise for me and as a beginner will save alot of headaches. Hopefully the 12v fan motor I have is capable of driving it otherwise I need to look at another motor or some kind of gearing.

Still on the shopping list I need;
- A power transformer ( I might just piggy back a couple I have)
- Some form of chassis
- a phonic wheel for the back of the motor
- a hub for the nipkow disc

Time to clear the workbench ready for this one!

PostPosted: Thu May 08, 2008 5:41 pm
by Klaas Robers

I think you don't need a phonic wheel. This is mechanically spoken a difficult thing to make. To begin with you don't need it at all. By carefully changing the speed of the motor you will be able to synchronise the picture. I have very disappointing experiences with the phonic wheel, originally from a Baird Televisor. It gives lots of humming sounds and the synchronisation action is very limited.

I would not use a fan-motor. These are nomally series-motors, of which the speed is varying with the load. In a fan the load increases heavily when the speed increases, so the speed will stabilise. For the Nipkow disc this is much less. You should prefer a permanent magnet DC motor, the type that you can find in cassette players, or somewhat heavier, in rechargeable drilling/screwdriving machines. Those motors have a stable constant speed at a constant voltage. So see that you get a well stabilised voltage to drive them and you will obtain a constant speed.

Somewhat later you can use the extra synchronisation holes in the Club Disc. Mount an opto sensor and hide the synchronisation circuits somewhere, so nobody can see them. I know that Denis Asseman made very fine replicas of dics receivers, where the IC electronics hided in the hollow bottom board. Of the tubes only the filaments were in function.

However, if you want to drive the LEDs with a tube, e.g. the EL84, then you shoud connect the LEDs in series, one long chain, and connect this in the anode circuit. LEDs should be current driven, so the anode is a much better place than the cathode. In my monitor I have 32 high brightness LEDs in one chain and a supply voltage of about 120 V. That is the reason that I use the high voltage transistor BUT11A (see description at the The max current of my LEDs is 50mA, a current that the EL84 can have as a mean current, where it is now the peak current. The voltage accross the orange LEDs is about 2 volts per LED.

PostPosted: Thu May 08, 2008 11:11 pm
by chris_vk3aml
Very true, Klaas,

The series AC/DC motor has very much poorer speed regulation than DC motors, and DC shunt wound motors in particular provide excellent speed regulation.

However, there were two good reasons in 1930s Britain to use a "universal" AC/DC series motor in Baird's commercial disc Televisor receivers. One of these reasons has been overcome by the more recent standardisation of mains supplies to AC, but another requirement remains - and this may have been a reason for your lack of success with the phonic wheel, Klaas.

This is because a series motor is susceptible to speed changes by a variation of motor load, and with a phonic wheel speed corrector this susceptibility provides the means for the phonic wheel actually taking control of the speed. With a DC shunt wound motor, there would be considerable resistance to an attached phonic wheel's ability to change the rotational speed. The net result would be that the phonic wheel would have to be driven with a lot of power, and it would, as you say, be 'noisy and unreliable'.

When one investigates the reasons for Mr Baird seemingly placing a substandard motor design in his commercial disc televisors, one detects a degree of engineering subtlety which only increases one's admiration for the man.

Like Klaas, I would suggest that you forget complex synchronisation schemes as a first step to getting pictures out of your televisor. A simple speed control is quite sufficient, and as Klaas suggests, the motor out of a cordless battery-powered drill or screwdriver would be an admirable source of motive power. One could also examine a car wrecker's yard for small DC motors, and I know that Daniel Gosson has used windscreen wiper motors with excellent results.

As this subject of motive power for scanning discs has arisen, I will post scanned copies of two articles on the subject under a new subject heading.

Hoping - again - that this helps,

All the best,

Chris Long VK3AML.

PostPosted: Fri May 09, 2008 1:02 am
by Marcus
Klass and Chris,

I do have an old 12v drill that was discarded by my work as they lost the charging dock. I hadn't thought of it until you mentioned it.
I knew I hoard stuff for a reason!

Doing away with sync for the moment probably isn't a bad idea. First I'll get the picture up and running and go from there.
If I can manually sync a picture I'm pretty sure I'll be happy with that in the long term anyway, it just adds to the novelty.

Later on I might give making an opto sensor with a phototube a try.

Thanks again for your tips.

PostPosted: Fri May 09, 2008 6:56 pm
by Marcus
I have pulled apart the 12v drill and the motor will have more than enough speed and torque to do the job.
Without a lathe, can anyone suggest a way that I can make a hub for the aluminum Nipkow disc I have coming?