CRT General Discussion.

[Steve Anderson]

...I've now got better though by no means perfect results...so must have had something wired up wrong somewhere. I now get 1mH for 3.8R series or 1/2mH for 1.9R for the H coils. Dom

Actually it doesn't look too bad at all! I assume the 'dotting' is your colour sub-carrier?

I've done some simulation using a very simple circuit and it appears that at 400Hz with a retrace time of 125us the 500uH inductance with 1.9 Ohms in series is simply too high. The inductive time constant comes out too long at around 1ms...which is why I say it looks very good as it is. Are these coils the ones that were originally the 50Hz vertical ones?

Reducing the inductance down to 100uH makes it all look a whole lot better.

It would be useful to know the values for the other set of coils too.

Steve A.

[AncientBrit]

Dominic,

Still very good results.

A few years ago I modified a "Leader" X-Y LF magnetically scanned display unit and I know the problems you are facing.

Very well done.

Regards,

Graham

[dominicbeesley]

Thanks Steve,

These are the original H-scan coils be reused, I suppose when they were being used along with a proper flyback transformer there would be plenty of oomph available for the flyback action? (I still really have a bit of a blind spot when it comes to how inductors work....I'm getting there slowly!)

Also they are now wired as 2 in parallel in series with the next two in parallel, in the original TV they were all wired in paralel I just couldn't get enough width doing it that way but would have reduced inductance?

The frame coils are wired in parallel and measure 3.8R and 3.65mH with the DMM.

Yes, the spots are the sub-carrier, hardly visible on a televisor due to the aperture but on this pin-sharp CRT everything shows up. I'll try adding a 15kHz trap at some point.... and possibly a comb filter (I need a 2.5ms delay line though, I'm thinking of a microphone and speaker 85cm apart?)

The photo actually gives a nicer impression than the real thing as the camera has thickened the lines a lot. In reality it all looks quite spidery and flickery!

Graham,

On that leader scope how were the frame coils wound and driven? Was it a valve device? I'm still struggling to get spot-wobble working. To get enough drive at high frequencies, I need to be at least 200kHz but preferably more, for this I need quite high drive levels to overcome the inductances....

Thanks

Dom

[Klaas Robers]

[quote="dominicbeesley"] I'm still struggling to get spot-wobble working. To get enough drive at high frequencies, I need to be at least 200kHz but preferably more, for this I need quite high drive levels to overcome the inductances....
[/quote]

Dom, that might be less a problem than you fear. The inductance works as a low pass when you start from the voltage across the coils. For a higher frequency you need a higher voltage. But as you need only very small deviations for the spot wobble, the voltages could be well within the limits.

But I have seen that in the 405 line times a separate tuned coil was used for the spot wobble. That is a thing that you also could use: tuning, i.e. resonance. Then the voltages for the series tuned circuit can be rather low.

[Steve Anderson]

Dom,

I'm now totally confused! The original horizontal (15.625kHz) coils have an inductance of.....?...and a resistance of....?

The original vertical (50Hz) coils have an inductance of...?...and a resistance of...?

Which ones are you using for what? And/or have you rotated the yoke on the neck?

My suggestion would be to use the 15.625kHz windings for the 400Hz NBTV line rate and the 50Hz windings for the 12.5HzNBTV frame rate.

What compounds the confusion for me (which is not hard), is the fact you appear to be using horizontal line scanning and vertical frame scanning.

Let's at least get a decent display first and think about spot-wobble downstream...

Steve A.

[dominicbeesley]

Sorry, my fault I have to confess I've confused myself a bit with the original duff readings...

For slow scanning I will stop using H/V labels as this is confusing because I tend to like wide aspect stuff (I'll mark then 400/12.5). The scanning is NBTVA on its right to left, top to bottom...All my material is recorded in wide rather than portait mode...

I have NOT rotated the yoke, as this seemed counterproductive as you note...

Original H coil is used for 400Hz, has an inductance of 1mH and 3.8R (when wired as 2+2 as it is now)
Original V coil is used for 12.5Hz, has an indutance 3.65mH and 3.8R
(original undisturbed wiring)

Yes spot wobble is last on the list (I remember now what the extra capacitor was for in the feedback loop - to stop NFB for high frequencies i.e. increase gain for anything over a few kHz...bad idea it just overloads everything more!)

Sorry for causing confusion...

Dom

[Steve Anderson]

Sorry, my fault I have to confess I've confused myself a bit with the original duff readings...I have NOT rotated the yoke, as this seemed counterproductive as you note...Sorry for causing confusion...Dom

No prob, I'm easily confused anyway! OK, all noted. For 15.625kHz on the original horizontal (and now still horizontal/line) deflection there must have been quite a pulse during the retrace time which was less than 10% of NBTV retrace time.(12us as opposed to some 100us+) I've just got home from a session with the lads so it'll be a day or three before I can have a stab at this as I'm off for a couple of days on 'real work' again. Still, it cranks up the air-miles...

Steve A.

[dominicbeesley]

No worries,thanks for looking, I'll probably not be getting much done until next week as I've got to go boozing in Glasgow....

Dom

[dominicbeesley]

Hello all,

I'm getting closer - the circuit is now a lot more stable, runs at about 12V/400mA for the whole telly (not a single burnt finger for days!) and is fairly linear....next steps - work out how to centre the image, flywheel premptive sync for frame and some form of flyback blanking, rotate the yoke to put the picture back straight!...

It's taken me a couple of weeks but now I've managed to get a flywheel line-sync circuit that doesn't have a tantrum when there's a missing sync! and I now finally understand how these work.

Anyway I'm sure I'll be back again soon with more silly questions!

Dom

[Steve Anderson]

Dom

You really could do with giving that screen a rub-down with a damp cloth! It'll improve the contrast!

Steve A.

[dominicbeesley]

Hi Steve,

That may look like muck but is in fact the reflection off the bench in front. The screen is actually surprisingly clean.

The contrast isn't bad at all now, the white looks overloaded in the photo but that is the fault of the phone camera not getting the exposure right and the reason why I had to leave the lights on - if not the phone ups its exposure time and overloads even more badly....

I'll endeavour to get better photos in future but that involves sneaking the digital SLR into the workshop past the boss...

Dom

[Steve Anderson]

Hi Steve, I'll endeavour to get better photos in future but that involves sneaking the digital SLR into the workshop past the boss...Dom

I'll not take the subject further...she obviously has control...no different here...note the grin...

Steve A.

[dominicbeesley]

Plied her with alcohol and sneaked past....here's the pics, not great though...I may have had a little booze myself!

Still not sure whats going on with the slightly wobbly right hand edge to the picture - as this is the start of a scan line I would have thought that problems either with speed variation of the sync unit or cross talk between frame and line would have had an equal effect at both ends...I'm sure I'll track it down and get it bob on at some point though!...anyway at least the flyback blanking sort of works!

Dom

[AncientBrit]

Dom,

That's pretty impressive, well done,

Regards,

Graham

[Klaas Robers]

Dom,

I think the wavy raster is because of the instability in the High Voltage. When the beam current is not constant the high voltage should regulate itself to constant. During the frame fly back blanking the HV is not loaded, so it can increase. That gives a smaller picture! Then during the first lines the load returned, so the HV goes down, making a wider picture and returns to normal due to HV feed back.

If this is the case the raster should become better if you go for less contrast. Then the beam current is lower and you get less variations in the load of the HV.