Thermionic standards converter......?

nuvistor said
I thought the Murphy TPG11 would have been used in the factory were TV’s were made. I didn’t see one in quite a large service department with about 15 bench engineers so certainly not in smaller shops.

It was not a cheap device by any imagination but the number of them I have seen in preservation suggests that they did go to the better shops.

In the earlier examples of the Television Engineer's Pocket Book the chapter on test equipment contains the opinion of the author that "...anything not producing a waveform to BBC specifications should not be allowed on the premises!"

At the time, only two devices (AFAIK) did - the TPG11 and the 'Radar' unit whose model number temporarily escapes me at the moment. A later Taylor Instruments unit did too.

Till Eulenspiegel said
The TPG11 was an expensive TV pattern generator and not to be unkind the and hardly sort of instrument one find in the workshop of the 1950s telly shop.  

Till Eulenspiegel.  

On the contrary, the TPG-11 was purpose-made for Murphy television dealers to purchase for their workshop, so that television sets could be repaired on the bench out of the, then very limited, hours of transmission.  It was the latest in a series of bench instruments produced by Murphy for their dealers, and, Murphy being Murphy, they went to a lot of trouble to get the performance right where it mattered - there was a pre-war Murphy signal generator, for example, whose r.f. harmonic performance put many so-called standard signal generators to shame.

A Murphy dealership, and the factory support that came with it, was very highly sought-after before and just after the war, and Murphy Radio were very selective about the firms to whom a dealership was awarded.  No firm without a prominent presence on a principal shopping street would even be entertained.

A wider range of test patterns, and a correspondingly more complex generator, were required for use in the factory.  I have no idea what Murphy used, but I have seen a photograph of the equipment used by another prominent British manufacturer in their production test area.  It occupied four bays - and, this being just post-war, the standard bay was then 22" across, not 19.
 

nuvistor said

The first pattern generator I had was the RBM dual standard which we obtained for the start of colour in 1967.  

If this is the RBM TPG-55 (greyish-coloured unit with video and r.f. out on Belling-Lees in the south-eastern corner of the front panel, handle on the top, runs on a PP9 or an optional extra mains unit), there is a good deal of internal evidence to suggest that it was designed, if not manufactured, at the ex-Murphy factory in Welwyn Garden City.

It is an interesting generator in that, although it produces convergence grille on both standards and is clearly intended for adjusting delta-gun sets, it does not produce a colour pattern of any kind and isn't, therefore, a huge amount of use for fault-finding on decoders - the most complex and, to the service engineer, unfamiliar area of a colour set.

The other odd thing about it - although I am relying on memory now - is that I believe the sync waveform is non-standard on 625, as there are no equalising pulses.

The circuits, waveforms and user manual for the Rank Bush Murphy TPG-55 are in the data library, should you wish to peruse them.

occiput said

On the contrary, the TPG-11 was purpose-made for Murphy television dealers to purchase for their workshop, so that television sets could be repaired on the bench out of the, then very limited, hours of transmission.  It was the latest in a series of bench instruments produced by Murphy for their dealers, and, Murphy being Murphy, they went to a lot of trouble to get the performance right where it mattered - there was a pre-war Murphy signal generator, for example, whose r.f. harmonic performance put many so-called standard signal generators to shame.
   

In reverse order: would that be the M1 oscillator? I believe there was also an oscilloscope/wobbulator as well (P1?). A much later service oscillator was however made by Advance, a thinly-veiled repaint of the E2.

The TPG-11 was available to non- Murphy dealers through (IIRC) F. Livingston Hogg and in trade journals of the time, no price was indicated; presumably on the basis of "If you have to ask..."

The RBM generator was for setting up not for fault finding in the colour circuits, it did the job it was made for, don’t know about the equalising pulses but again did what was required. I had to wait a while and then had a Philips colour bar gen, probably around 1972 but that date is movable. Test card F was still transmitted for long periods and that provided many features for fault finding so I never took the colour bar gen out on the van.

I also got a single standard Labgear cross hatch/ grayscale gen for the van around that time.

 Occiput wrote: "On the contrary, the TPG-11 was purpose-made for Murphy television dealers to purchase for their workshop, so that television sets could be repaired on the bench out of the, then very limited, hours of transmission.  It was the latest in a series of bench instruments produced by Murphy for their dealers, and, Murphy being Murphy, they went to a lot of trouble to get the performance right where it mattered – there was a pre-war Murphy signal generator, for example, whose r.f. harmonic performance put many so-called standard signal generators to shame"

Back in the "old" days to become a Murphy was something really special to aspire to.   In the early days of my business  I had to make do with wholesaler brands such as Alba and Marconi. Nothing wrong with that though.  Did actually become a Murphy dealer in 1972 but by then the company was giving dealerships to just about anyone that had a shop.

Building a power pack for the TPG11.  There's the remains of a scrap Murphy A4 radio chassis in a corner of the workshop. Will use that as the basis of the PSU.   At least it keeps things Murphy.  Considering using a pair of EY84s for the full wave HT rectifier.

Till Eulenspiegel.

This may be of interest, it’s some details of the GPO colour bar generator used in 1958 for the Television Society Exhibition.

page 160 .

http://www.americanradiohistory.com/Archive-Wireless-World/50s/Wireless-World-1958-04.pdf

Cathovisor said

In reverse order: would that be the M1 oscillator? I believe there was also an oscilloscope/wobbulator as well (P1?). A much later service oscillator was however made by Advance, a thinly-veiled repaint of the E2.

The TPG-11 was available to non- Murphy dealers through (IIRC) F. Livingston Hogg and in trade journals of the time, no price was indicated; presumably on the basis of "If you have to ask..."  

It was indeed the M1 oscillator.  My recollection - and it's well over forty years ago, was that the frequency-setting dial was scaled 0 - 180 degrees, and presumably therefore the equipment came with a set of calibration charts.

One of the craziest jobs I have ever been involved in - and since it will be clear that I used to work for the BBC, you'll appreciate that the bar is set quite high - was to recalibrate an M1 as a favour for a drinking buddy of the MD of the firm where I did my undergraduate work experience.  He'd got one which had clearly become parted from its calibration charts.

RF wasn't really our thing, but we were doing a job for the MoD which involved having to convince them that our electrostatic field measuring sets were not going to interfere with the receivers of shipborne radar or comms. For this purpose, we'd managed to persuade them to lend us a virtually brand-new HP spectrum analyser and tracking generator. (You will understand that electrostatic discharges and explosives are not a very healthy mix, particularly in the confines of a naval gun-turret).

Rien de plus simple, as they say: feed the M1 output into the spectrum analyser, spot the peak on the display, turn the track-gen so that the peak due to its output was on top of the one due to the M1, read the frequency off the nixies on the track-gen, turn the dial of the M1 on 10 degrees, rinse and repeat, and on all the other ranges as well.  Plot the results on some sheets of graph paper, get the french curves out and draw a nice neat line through the points, letter and number the axes, Robert is your father's brother.  Run the results through the Xerox machine and squirrel the copies away so as to save some time in the event of the originals also becoming parted from the generator.  Payment was in the form of a bottle of red wine which was, as you might perhaps imagine, more than acceptable to an engineering undergraduate.

While the calibration was going on, I thought it might be a bit of a giggle to look at the spectral purity of the M1, which is how I know it was quite good.  The object of the exercise from my point of view was to learn to drive the spectrum analyser, in which I think I may have modestly succeeded.

I did, at one time, own one of the Murphy wobbulators which, since it had 4V valves in it, I've always assumed to be pre-war.  I think the oscillator circuit was similar to the local oscillator of the A28C in that a reactance valve was used.  I didn't do anything useful with it, and I think passed it on to a more deserving home in preparation for a house move.  I imagine that the original purpose was to make easier the task of lining up overcoupled i.f. transformers in things like A28s and A40s.

"This may be of interest, it’s some details of the GPO colour bar generator used in 1958 for the Television Society Exhibition."                                                                                                                                                           Hi Frank, noteworthy is the delay line and pickup coils to facilitate the timing functions. Clever.                  The Murphy TPG11.   Lots going on inside the unit.   Have counted twelve 6SN7GT double triodes and five 6F13 pentodes. Reckon I should sell that ECC34 for hundreds of pounds to a audiophool. Pure triode sound you know. Till Eulenspiegel.

The HT requirement for the TPG11 is only 125volts. The HT is supplied by one GZ32 full wave rectifier valve. We'll have to find out the HT current requirement.   Heater current demand is 13amps @ 6.3volts so we're going to need a hefty gauge of wire for the transformer heater winding.

From the Murphy TPG11 test report:   The TPG11 has found other uses than in servicing or the development of TV sets. It is not at all difficult to remove the pattern, keeping the sync and blanking pulses and then modulate by means of a Monoscope or similar tube, and so have a complete TV transmitter.  It is also used by the BBC for certain OB testing, and by other authorities for testing lines and the like. It has even found an outlet in storage tube technique. It is understood that a model for 625 line Continental standards is being produced and will soon be available in this country as well as export. 

Till Eulenspiegel.

Till Eulenspiegel said
.........Heater current demand is 13amps @ 6.3volts so we're going to need a hefty gauge of wire for the transformer heater winding.

Till Eulenspiegel.  

How about the wire from the primary of a microwave oven's EHT transformer? Or even, leave the requisite number of turns on the original core, keep that as the secondary, and wind a new primary to drive it? (possibly from what was the original EHT overwind?) Perhaps you could wind an HT secondary onto the same core at the same time? - I guess it depends upon how adventurous you could become?

Just a thought...

Hi Marion,

              I do have the heater transformer removed from a scrap 1953 Pilot TV chassis.   But I'll still need the HT transformer.    As soon as a power pack has been constructed a start on the restoration of the TPG11 will begin.

Thermionic Standards Converter:   Is this my most outrageous project ever? 

Till Eulenspiegel.

From Wikipedia:    https://en.wikipedia.org/wiki/Selby%E2%80%93Driffield_line

Hi Marion,   are the tracks still in place or at least the track beds?

Till Eulenspiegel.

Till Eulenspiegel said
Hi Marion,   are the tracks still in place or at least the track beds?

Till Eulenspiegel.  

The trackbed is amazingly, substantially complete, though some parts have been scrubbed out to field level. Some of the station houses and/or stations are now in private ownership, with some being converted into homes and extended over the trackbed. At least one station, Market Weighton, has been completely eradicated and replaced by a small block of flats.

A bit like the thermionic converter, it would be difficult, but not insurmountable - if the money was there.

All in all, about 20 miles of contiguous trackbed still exists, but the once proud steel bridge across the Derwent was lifted out in about 1985. Of course, with an act of parliament, everything could  be swept away to make a clear route, but that's not going to happen. Realistically, a short stretch of around 10 to 15 miles could be relaid tomorrow if the financial backing was there.

It would be possible to reopen some of it as a heritage line, and if one could live long enough, the whole line could, with landowners' cooperation, be reinstated even if it meant rerouteing some short sections, however, to put it on context, there is one railway project where the line is completely eradicated, and the project has about a hundred yards or so of track relaid - they believe that they will eventually regain about 50% to 80% of the original line. In comparison, the S&DR would be a cakewalk.

Just hopping onto the line at any point (here I arrived at 'High Field'), you can explore the entire route on Nick Catford's excellent website;

http://www.disused-stations.org.uk/h/highfield/index.shtml

In 1978, my brother, a friend and I 'drove ' the line on three vintage David Brown tractors, from Bubwith to Foggathorpe, we pushed on a little further but were stopped by a huge drain that had been dug across the line (now a culvert). We previously did a moped run from High Field to Menthorpe in 1973.

Marion

Hi Marion, wasn't there a policy of allowing buildings to be constructed on former track beds and thus preventing any possibility of any rail lines ever being put back into service? Getting back to the Murphy TPG11. This is a masterpiece of design, well worth further study and even more so to be restored back to working order again.   The attachment shows the mains lock chassis. Along with the ECC34 two 6D2 double-diodes are employed in an interesting discriminator circuit. When locked to the 50c/s mains the DC control voltage corrects the frequency of the 20,250c/s master oscillator. The master oscillator takes the form of a multivibrator employing a 6SN7GT double-triode. By some means or other the 625 frame sync pulse can be used to control the mains lock chassis. This will then ensure the 405 line syncs will be of the correct frequency and thus good interlace will be maintained.  Till Eulenspiegel.

I've redrawn the frequency controlling circuits of the 20,250c/s master oscillator.  By using the 50c's 6.3V heater supply as the reference the sine wave is supplied to a two diode discriminator network which is compared with the counted down 50c/s frame sync envelope. The 500micro-second pulse serves as the gate for the eight broad pulses.   Switch S2 allows the controlling circuit to be disabled.   The 20,250 c/s oscillator is then free running. Till Eulenspiegel.

 ... and I thought this was ambitious ... http://www.radar58.com/projects/tubeclock.html

Hi Doz,  That's a brilliant piece of engineering. My thermionic valve standards converter pales into insignificance compared the all vacuum tube and nixie clock.

The core part of the converter will have to employ RAM memory ICs but rest assured as much of the back up circuits will employ valves.   To keep it simple a six bit ADC will be used, the RCA/Harris CA3300, my first converter used one with success. For the DAC how about a resistor R2R network?  Lets keep the circuits as crude as possible! 405 clock rate will be 6.5Mhz. The clock oscillators can use valves, 10Mhz for 625 and 6.5Mhz for 405. Good figures to work with in a simple converter because the actual number samples for the active video period will be 512, that's 52microseconds 625 and 80microseconds 405.  Don't expect broadcast quality pictures for this converter project, it's just a piece of fun like the Nixie Clock. 

Till Eulenspiegel.

Hi Frank,   Thanks for sending the link to the Storage Engine website.   All we need is to do is write in a single 625 64uS line  and read out at the 98uS line rate.  Work on a simple line dropping principle, worry about interpolation of the dropped lines later.    Link to the Silicon Engine website.    Invention of the transistor in France by Herbert Matare:

http://www.computerhistory.org/siliconengine/the-european-transistor-invention/

Till Eulenspiegel.