Types of CRT used in colour TV's.

[ctc15]

Hi All
I am restoring a BRC 2000 Marconi 4701 it seems shame that the components for the pincushion correction have not been fitted when it was made. The board had been made to take them.
I used a AT4041/37 Transductor and a coil type AT4040/50.
It is now working on 625/405 lines.
Keith

[Simon Gittins]

Originally Posted by InsertNameHere

were these decent? I’d imagine for £8 back in the 70s/80s, as the OP of them photos said, thats fairly expensive, so I’d imagine they were somewhat decent?

Thanks

I never used a solid state PL802, I remember that opinions differed as to how good they were compared to the valve. However, the idea was that these wouldn't wear out and the PL802 was always a relatively expensive valve so this could have been a cost effective solution.


There wouldn't have been any heat reduction, the large resistor was to maintain continuity of the series heater chain and would have dissipated as much heat as the valve (although it was slightly futher away from the CDA panel which may have helped).

[19Seventy7]

Hi,

sorry I didn't reply sooner, I didn't see you had replied until now.

It makes sense to replace with something more reliable. Do transistors not wear out?

Thanks

[Maarten]

Transistors don't wear out as a rule, but they can have faults that manifest themselves over time and circumstances so there can still be a time/usage related factor in defects.

Well known examples are the early leadfree tinned transistor cases (most Philips/Mullard germanium transistors in tin cases) that can suffer from internal tin whiskers that build up over time, Mullard lockfit transistors (early molded plastic cases that may not always be as sealed as they should be) and Japanese 'black legged' transistors from the 1970's where the silver plating on the legs tends to creep (visible whiskers between the legs are also reported).

Also, transistors that are run relatively hard, might suffer from some kind of wear and tear as heat speeds up any chemical and physical process and can help form hotspots or local migration of materials.

[19Seventy7]

Thank you for explaining

So theoretically, if someone were to want to keep originality with components, they could “repair” a transistor?

Thanks

[slidertogrid]

The solid state PL802 came about due to a shortage in the supply of PL802 valves certainly for a while anyway. When I was an apprentice the firm I worked for had loads of PYE hybrids out on rental I remember there being a real problem getting PL802s. All the valves we used were Pinnicle or Solus then suddenly we had Mullard PL802s! It was because they were the only ones the service manager could get. They didn't go on the valve rack though you were allowed a couple in your Kit and had to return the dud one to get a replacement.

The Ledco and DR developments Solid state CDA panels were made to replace the cooked up originals. we fitted stand off valve bases to original panels that hadn't gone too far but some were just had it!
The replacement panels came out a bit too late really though the sets were getting on a bit by then, there was an incentive to keep old sets going though they were 'decontrolled' and could be rented without a large advance rental so were popular with the less well off of customers...
The service manager was of the opinion that the ventilation slots in the later sets were not big enough which caused the sets to overheat large cabinet models didn't suffer like the 22" CT205. The hot summer of 1976 didn't help...

The bonus of fitting a solid state CDA meant that if the Tube was a bit 'gassy' the picture improved and didn't 'green band' at higher contrast levels.

The big dropper mounted on the top of the replacement CDA to replace the valve heaters soon fried the replacement panel though.. However by then the sets were old and virtually scrap anyway!
I liked the sets though, not that reliable but very easily fixed!


Rich

[1100 man]

I knew the solid state PL802 that I bought 36 years ago would come in useful one day! So here it is, the only useful thing it's ever done is to provide this photo! I've carted it about ever since!

I found the proper valve gave a sharper picture then the solid state version.

An unbranded PL802, from my CT205 is shown next to it.

Cheers
Nick

[1100 man]

Hi Aaeron,

Both valves & transistors have their good and bad points. As valves age, they tend to loose emission, that is, the cathode emits less electrons, so the power the valve can deliver reduces. The hardest worked valves in a TV are the Line output valve, efficiency diode, frame output valve and HT rectifier. The effects of reduced emission are reduced picture width or height.

The efficiency of transistors on the other hand, tends to stay fairly constant with age. The biggest problem with transistors, though, is they are much more fragile than valves. They can easily be 'killed' by voltage spikes or excess current. Faults in other parts of the circuit can destroy a transistor where a valve would survive perfectly well. Valves are much more rugged and will survive more abuse.

For this reason, many TV manufacturers still used valves in the power output stages long after all transistor sets had been introduced. Transistors in the early '70's were pushed to their limit and often failed. Whilst valves tended to offer greater reliability, they also generated much more heat and used much higher voltages. This tended to 'cook' the cheap circuit boards used in TV's, which then carbonised, became conductive and sometimes had spectacular burn ups!

All the best,
Nick

[1100 man]

Hi Aaeron,
As promised, I've taken some pictures of my delta gun tubed Pye CT205 showing the effect of some of the convergence adjustments.
You can now see how essential a cross hatch pattern is for these adjustments.

First picture shows a reasonably well converged picture with all three guns switched on. The picture is white.

Second picture:- I've turned off the blue gun so the picture is yellow.

Third picture:- I've mis-adjusted the green static convergence control. You can see the green & red rasters are not over the top of each other.

Forth picture:- Blue gun switched back on but blue static convergence control mis-adjusted so you can see the blue raster displaced from the yellow. Don't forget the yellow is made up of the red & green rasters on top of each other....

[1100 man]

It's essential that each colour gun only hits it's own colour phosphors on the screen. As part of the convergence adjustments, the colour 'purity' needs to be set up. This is done, in conjunction with the static convergence, by moving the scan coil assembly back & forward to achieve a uniform red raster.

The first picture shows a reasonably pure red raster.

The second picture shows the effect of moving the scan coils fully backwards. Only the red gun is active, but it's beam is now hitting different colour phosphors on different parts of the screen.

So far, we've only played with the controls mounted on the scan coil assembly on the tube neck. There's all the dynamic convergence controls to adjust yet and then the 5 greyscale tracking controls!

So you can begin to see that setting up a delta gun CRT is not a 5 minute job!

All the best
Nick.

[Andrew2]

We had some of the solid-state PL802's in, but we didn't like them much. The pictures seemed a little softer and had less 'sparkle' than those from a thermionic one.

[19Seventy7]

Originally Posted by 1100 man

Hi Aaeron,

Both valves & transistors have their good and bad points. As valves age, they tend to loose emission, that is, the cathode emits less electrons, so the power the valve can deliver reduces.

The efficiency of transistors on the other hand, tends to stay fairly constant with age. The biggest problem with transistors, though, is they are much more fragile than valves.

Transistors in the early '70's were pushed to their limit and often failed. Whilst valves tended to offer greater reliability

Thanks for all of the info!


By sounds of it, the only thing transistors were better at doing at the time was keeping constant output with age and not getting as hot?

As promised, I've taken some pictures of my delta gun tubed Pye CT205 showing the effect of some of the convergence adjustments.
You can now see how essential a cross hatch pattern is for these adjustments.

First picture shows a reasonably well converged picture with all three guns switched on. The picture is white.

Second picture:- I've turned off the blue gun so the picture is yellow.

Third picture:- I've mis-adjusted the green static convergence control. You can see the green & red rasters are not over the top of each other.

Forth picture:- Blue gun switched back on but blue static convergence control mis-adjusted so you can see the blue raster displaced from the yellow. Don't forget the yellow is made up of the red & green rasters on top of each other....

Thank you for showing me that!

I do really see the importance of a Cross hatch generator. I can't imagine how tricky it is (If possible) to sort out the convergence without one!

it's very interesting to see how they all work together, I'm starting to want a DG set aswell to "Tinker" with and see how all different settings work and do what.

How tricky is it to align the three coloured rasters? Are they more fiddly, in the sense of getting them as close as possible?

Are there any software systems, or even DVD's perhaps (VHS would be a lot easier for me, but I doubt that even more) that have the pattern generators? I'd like to have one to just have a look at on my TV and see how well it is converged.

It's essential that each colour gun only hits it's own colour phosphors on the screen. As part of the convergence adjustments, the colour 'purity' needs to be set up

The second picture shows the effect of moving the scan coils fully backwards. Only the red gun is active, but it's beam is now hitting different colour phosphors on different parts of the screen.

So far, we've only played with the controls mounted on the scan coil assembly on the tube neck. There's all the dynamic convergence controls to adjust yet and then the 5 greyscale tracking controls!

So you can begin to see that setting up a delta gun CRT is not a 5 minute job!

Isn't it called "fringing" when the supposedly pure raster isn't one solid colour, and perhaps goes lighter in certain spots? Is this the same when the red beam is also hitting the blue and green phosphors, as well as red?

I can really see that it isn't any 5 minute job!

Thank you so much for explaining that all, I have a much better understanding of DG sets now! I really appreciate your help!

Thank you!

[Maarten]

Some other obvious advantages to using transistors: efficiency (apart from not having a heater, the internal resistance of a transistor is much lower) and size. Much compacter designs could be realised in semiconductor technology.

There's DVD's with test patterns on it, so you could use your DVD player as a cheap alternative to a test pattern generator.

[19Seventy7]

Thanks for your help

I'll get too looking online for the DVDs now!

Thanks

[1100 man]

Originally Posted by InsertNameHere

By sounds of it, the only thing transistors were better at doing at the time was keeping constant output with age and not getting as hot?

That sums it up quite nicely!!

The other problem with transistors is keeping them cool! Whilst valves need to be heated to make their cathodes emit electrons, transistors generate heat -especially when working hard.
If a transistor gets too hot, it goes 'pop'! So transistors often need to be mounted on heatsinks to keep them cool enough to prevent failure.

The first picture shows three very common line output devices used in different generations of colour TV's.

On the left is a PL509 valve as used in the Pye CT205 pictured above. In the middle is a classic BU208 transistor as used in many '70's sets. Your Hitachi would use a similar type of transistor but of Japanese manufacture. On the right is a BU508 which is more likely to appear in an 80's set.

The second picture shows efficiency diodes- on the left a standard PY500 valve and it's semiconductor equivalent on the right.

Note that the heater power of both valves is about 13W, so that's 26Watts when both valves are used in a TV. That's a lot of heat to subject the circuit board to especially when it's made of resin bonded paper as used in all TV's!

Hope that might be of interest,

Cheers
Nick

[1100 man]

Are there any software systems, or even DVD's perhaps (VHS would be a lot easier for me, but I doubt that even more) that have the pattern generators? I'd like to have one to just have a look at on my TV and see how well it is converged.

There's a downloadable DVD of various testcards here.

http://www.oldtechnology.net/TestCardDisc/V0.371.iso

Obviously, you will need the requisite equipment to actually get it onto a DVD or know someone who can do it for you.

Cheers
Nick

[19Seventy7]

How do heatsinks work? I know what they do, but not how they do what they do. Do they just enlarge the surface area to help dissipate heat, or something similar?

Do valves just push in like a plug? Do they not need soldering?

When did valves finally stop being used in TVs? Mid 70s?

As you say, my Hitachi does have a transistor like that. I think there is only one, at least that i can remember. There may be a second one but im not 100% sure.

Were the BU508s used in anything other than TVs? I think iÂ’ve seen these but I doubt itÂ’d have been in a TV.

What is the maximum “wattage of heat” (if that makes sense) a PCB can withstand?

Thanks for your help!

[19Seventy7]

Originally Posted by 1100 man

There's a downloadable DVD of various testcards here.

http://www.oldtechnology.net/TestCardDisc/V0.371.iso

Obviously, you will need the requisite equipment to actually get it onto a DVD or know someone who can do it for you.

Cheers
Nick

Thank you for linking that!

My PC should be able to burn it onto a DVD. I just need to get hold of some blank DVDs now!

Thanks again

[1100 man]

Hi Aaeron,

A heatsink does exactly what you describe- it just enlarges the surface area and conducts the heat away from the transistor. Heat is one major problem with semiconductors, As the speed of microprocessors in computers has got faster over the years, getting rid of the heat has been an ever increasing problem. The main heatsink in a computer is often fan cooled and is massive!

Most valves just plug in to a 'valve base'. As they have quite a few pins, (9 is a common number), soldering would be impractical. The PL509 & PY500 pictured above both have 9 pin bases. Both these valves also have a 'top cap' ie, a push on connector on the top of the valve. This keeps the high voltage connection away from the pins.
There are a few valves which have wires and are soldered into the circuit. These tend to be diodes which don't have many connections.

Transistors, on the other hand, only have three connections- the base, collector & emitter. To make them work in circuits, you need more external components (resistors & capacitors) than with valves. Transistor circuits, therefore tend to use more components and be more complicated.

I don't know when the last use of valves in TV's was. My thorn 1500 (black& white) was made in 1978 and still uses 7 valves, the rest is transistorised. Mind you, this was a pretty old design by then. Of course, the CRT is essentially a valve attached to a screen, so really at least one valve was used until the demise of the CRT!

BU508's were specifically designed for the line output stages of TV's. They are designed to withstand very high voltages and so are not that useful for non TV applications.

PCB failure mechanisms are complex and were probably not fully understood by the manufacturers.
Professional grade PCB's are made of a glass fibre compound which is really good at withstanding heat & high voltages. For economy, domestic TV's used a resin bonded paper board. These probably performed OK when tested by the manufacturer. One of the big problems in the '60's, '70's & '80's, was that many people smoked and there were many households that had open coal fires.

The tarry residue from smoking coats everything in the back of a TV- everything goes a shiny brown colour. This tends to be conductive at the high voltages that valves use. Add to that the extra humidity due to poor room heating and after the board has been cooked by the valves, circuit board failures were very common.

Cheers
Nick

[19Seventy7]

Hi,

I did think valves were just “plug and play” but wasnt sure.
Sounds like theyÂ’d be an easy (and relatively common) repair/replacement in TVs.

In a sense, wouldnt transistors be just as “unreliable” as valves as there would be more components to fail? It may not be the transistor itself that fails, just a lead up to it?

IÂ’ll see if a google search will tell what the last valved set was (Other than the CRT) iÂ’ll report back with what i find.

I dont think I would have seen them then, i might be confusing them for something else.

Is heat the only thing to cause PCB failures? If not what else can cause failure?

What was done when a PCB did fail?

Thanks