Golborne Vintage Radio

The eventual plan is to build a valved old fashioned SSTV/NBTV monitor using this ex RADAR tube for the display.  It is also a learning exercise for me, in that I hope to understand more of the principles of electro-magnetic deflection. 

Although I believe, a LOPT will not be driving the scans for such a slow unit I also want to understand these better and how they work and are calculated etc.  So as time goes on expect lots of questions that may seem trivial, but please understand before taking up the hobby I was not a TV engineer.

The coil former was turned from black nylon, 42.35mm diameter and 52mm long where the wires will go.  I then parallel wound 7000 turns of two 0.16mm dia, enamelled wire, giving a max rating of around 12mA.

It takes 66mA to focus the spot at 2KV EHT and 82.5mA at 3.5KV EHT.  Scan coils and static deflection to follow at some point.

Youtube video.

Adrian

Basic idea about inductor is that the voltage is proportional to the rate of change of current. V = L * dI/dt if you are familiar with simple calculus.

In a normal TV the line timebase works, to a first approximation, by applying a constant voltage to the scan coils. The current then rises linearly until the spot is at the right edge of the screen. Energy 0.5*L* I^2 is stored in the coil. You then switch off the voltage and the current tries to drop to zero very fast. This generates a high voltage peak which is tamed by resonating the L with some C. This also causes flyback. You then turn on the constant voltage again and the energy stored in the coil is returned to the supply for the left part of the stroke.

This is a very crude explantion but it shows that for normal H scanning it's the L of the scan coils that's dominant.

For the field scan the resistance ofthe scan coils is dominant in the forward stroke. The timebase is rather like an audio amplifier. During flyback you have to apply alarger voltage to get it to happen because the L of the coils matters a lot more.

For SSTV the H scan is so slow that normal audio amp techniques will work pretty well. You'll still get a higher voltage peak during flyback.

Just back in from a small family emergency some 20 plus miles away, (Blocked Bog!).

The part I am trying to understand is the action of the primary winding of the LOPT, the circuits I have looked at show that winding or parts of it in parallel with the line scan coils, so it is the combined L or just the L of the scan coils, or am I on the wrong track.

Also

If I look at this circuit from the Fergy
[attachment=21060]
If I am following this correctly the supply for the line output can only go the V14 to point e when the valve/diode is conducting, then goes to the scan coils via point d and guessing that all the taps to C are acting a bit like an auto transformer.  f to g being the overwind.  I understand the heater winding for V13, but a to b?  To me there seems to be a lot of windings that I think would add lots of inductance?

There is an article I am trying to follow and understand called scanning and synchronisation, this started in Practical Television September 1957.

Think I may have to look further back in the archives.

Probably a bit late to get sense out of me tonight, I will adjourn to sleep.

Adrian

Although the details are more complex, the LOPT is there to match the output of the valve to the scan coils. Just like any transformer (where the core is not saturated) its own inductance is of relatively minor importance.

This sort of line output stage will only work over a fairly narrow range of frequencies. If designed for 405 or 625 it will not work at all at slow scan frequencies.

I have attached a good article on line timebases that use energy recovery. This is relevant to fast scan TV but barely at all to slow scan.

To give some idea of why energy recovery is essential for fast scan TV, on each stroke a certain amount of energy is stored in the scan coils. If this is not recovered the line timebase for a wide angle CRT would consume many 100s of watts which would have to be disspated somewhere in the set. I once calculated that a first generation colour TV has over 2kW of power circulating in the line timebase. But energy is recovered so efficiently that the line timebase consumes only 10s of watts from the mains.

For slow scan, there are far fewer strokes per second so much less power is needed. For a fairly narrow angle CRT like the 5FP7 the scan power is fairly low anyway. So the timebases are best treated more like audio amplifiers.

Thanks I will print and read the article.

Cheers

Adrian

If you want to have any scan rate, like a scope has, then the EHT circuit needs to be separate from the scanning. For up to maybe 2 kV or 3kV a diode multiplier is needed, otherwise use he LOPT in a standalone boost or Cuk inverter using a 16 kHz approx frequency. Maybe 20 kHz to 40 kHz will work.

A pair of high power IC audio amps can drive the scan coils.

Hello Mike;
I do appreciate that a LOPT will not be of anyuse for the SSTV monitor, but could be used for my DIY 405 TV. I used a separate HT unit in that with a Thyristor and small EC90., but trying to see if I can make use a LOPT initially as a EHT generator and then progress later to line scan.

What's that dirty word of IC's, has to be valves at this end, typically B7G's.

Adrian

OK, then EL95s to drive the scan coils.

And the LOPT can be a standalone valve based EHT generator.

A ringing TX like in projection TVs could be used for EHT, obviously much smaller, a multiplier would make the winding of the TX simple.

Hello Trevor.

The EHT unit I built for the home made TV came from one of the television circuits in PT and in the Television Circuits, there was also a RF style osc/EHT unit with tuned plate tuned grid single valve unit that I looked at, is that the type of thing you mention?

I believe I should have details of the projection EHT in another book and I will have a loot at that.

For SSTV using the 5FP7 I would be happy with a maximum of 4 to 5KV.
I hope to play coil winding again over weekend and have a go at a deflection coil set.

Adrian