1956 Pye colour TV set (completed) and early colour - Page 2 - UK Vintage Radio Repair and Restoration Discussion Forum

oldeurope · 4th Mar 2006, 12:18 pm

Yesterday I telephoned with Yagosaga. He toled me why not convert
a complete NTSC signal to 405. I made some thaughts and came to
the conclusion that this must be possible. With the cxl5506 the
clock frequency is 8,8MHz. So everything up to 4,4MHz will be converted.
If you tune the subcarrier frequency of the NTSC coder to 3,98671875MHz
you'll get 405 NTSC colour at the output. In the interpolator a cap can
be added to make sure that the Y component is interpolated only.

What do you think?

Kind regards
Darius

FERNSEH · 4th Mar 2006, 12:34 pm

According to the Black Book the frame blanking period of the original M-EMI system was 2000 micro-seconds= 20 lines per field, so it follows that 365 lines were active. As the aspect ratio was 5:4 we have 456 picture elements in the 84 micro-second video part of a line. It works out that the max video bandwidth will be 2.7Mhz.

As time went by some of the characteristics of the 405 system were changed. 4:3 aspect ratio was introduced in 1950 and the active video of a line reduced to 80 micro-seconds to accomodate the characteristics of the 625 system. I'm not sure when the frame blanking period was reduced to 1400 micro-seconds. Anyway, all these changes will have resulted in the max video bandwidth being increased to 3Mhz.

DFWB.

channel405 · 4th Mar 2006, 1:46 pm

Hi Darius,

As I remember, and of course could well be wrong, one of the differences between PAL and NTSC decoders is that there is no delay line in the NTSC one, which would be a problem if we were to have PAL 405 cos the delay line is 64us instead of 98 as per 405 line duration. I am right in thinking then if we want colour 405, best to stick with NTSC? Could use CCDs I suppose.

Tim

ppppenguin · 4th Mar 2006, 2:03 pm

The only colour system ever used on 405 was NTSC. PAL and SECAM were never even considered for 405.

I don't think direct conversion of 525 NTSC to 405 NTSC would work. The colour subcarrier frequency would be wildly wrong. If you wanted a largely analogue approach there are 2 methods.

1. Decode to RGB or YUV, retime to 405 with multiple CCDs and encode to 405 NTSC.

2. Separate Y and C, retime to 405, modify the PAL to NTSC and heterodyne it to the correct subcarrier, add 405 Y and C back together.

I'm sure both methods could work but I wouldn't be enthusiastic about them. David Robinson's approach is sound and proven. It might be possible to redesign the Xilinx code in Aurora as I mentioned earlier. The discussions about 405 from a PC are also relevant as the RGB output could be coded into NTSC.

Heatercathodeshort · 5th Mar 2006, 1:08 pm

Hello,
Certainly before the days of standards conversion the 405 system could and did provide a bandwidth of 3mc/s. If the signal was a good clean one as it certainly was in the Wimbledon area where I worked at the time, the full 3meg was reproduced on sets such as the PYE CTL58 series. In practise of course only a complete nutter would worry about the extra 300kc's or so but believe me some customers were very test card aware in those days and complained if all the 'bars' were not visible. Its difficult to calculate today how good the 405 service was. [In practise of course most sets were not up to reproducing them but many were] I recorded many tapes prior to the closedown of the service and its quite a shock today to see how good the end result was. I have a GEC BT1748 14" model in my collection, its in very good working order and engineers are amazed at the reproduction from this 50 year old model. I get very annoyed when even the BBC describe the system as 'Back in the flickering days of black and white'. A great deal of trouble and expense went into providing what was considered to be the best television service in the World and that included picture quality. Regards John.

ppppenguin · 5th Mar 2006, 1:43 pm

Quote:

Originally Posted by Heatercathodeshort

Certainly before the days of standards conversion the 405 system could and did provide a bandwidth of 3mc/s. ......... I get very annoyed when even the BBC describe the system as 'Back in the flickering days of black and white'. A great deal of trouble and expense went into providing what was considered to be the best television service in the World and that included picture quality.

the main reason the bandwidth was restricted in the later days was to avoid the colour subcarrier being converted as a nast crawling pattern. This could have been solved by a luminance comb filter but that was a fairly exotic bit of circuitry then.

The BBC and ITV did indeed take the trouble to broadcast the best possible pictures. It's a shame that horrors such as mean level AGC, AC coupled video and poor EHT regulation made such a mess for the viewer. That didn't get resolved until the advent of colour forced manufacturers to do something about these problems.

channel405 · 6th Mar 2006, 11:08 am

Quote:

Originally Posted by ppppenguin

The only colour system ever used on 405 was NTSC. PAL and SECAM were never even considered for 405.

Yes I know that, but there were many dual standard colour sets made and a few still exist. So I was wondering if the PAL decoder could be used on 405 lines instead of being passive. Apart from having to devise a way of encoding a PAL 405 (a small step from NTSC?) signal the only other technical difficulty I can think of is the delay line, if you get my drift.

FERNSEH · 6th Mar 2006, 12:12 pm

The approach is first find a dual standard CTV and modify the decoder to operate on 405 NTSC. The delay line can be removed. As for the sub-carrier oscillator, a 2.65Mhz crystal can be made or, as in the '56 Pye it can be a free running type locked to the bursts.
The colour killer will need to be redesigned and the band pass characteristics of the croma amplifier will need attention.
Modiying a 405/625 colour set has been considered by others.

DFWB

FERNSEH · 28th Jul 2006, 4:00 pm

I mentioned some time back that the 23KV EHT generator employed in the Pye colour set was similar to the system found in the 1955 RCA 21CT55.

For clarity, in the drawing of the circuit diagram I've substituted solid state diodes for the three DY86s.
The 17KV from the line output transformer overwind is augmented with the 6KV from the anodes of the two parallel PL36s.

Note the shunt stabiliser tube, the 6BD4.

DFWB.

ppppenguin · 28th Jul 2006, 4:11 pm

I ought to point out the mistake I have made in this thread. PAL and SECAM were both tried on 405 as described here:

https://www.vintage-radio.net/forum/...&highlight=cox

That 6BD4 is a nice chunky tube. Looks like an immediate ancestor of the PD500.

4th Mar 2006, 12:18 pm	#21
oldeurope Retired Dormant Member Join Date: Dec 2003 Location: Solingen, Germany Posts: 727	Re: 1956 Pye colour TV set. Yesterday I telephoned with Yagosaga. He toled me why not convert a complete NTSC signal to 405. I made some thaughts and came to the conclusion that this must be possible. With the cxl5506 the clock frequency is 8,8MHz. So everything up to 4,4MHz will be converted. If you tune the subcarrier frequency of the NTSC coder to 3,98671875MHz you'll get 405 NTSC colour at the output. In the interpolator a cap can be added to make sure that the Y component is interpolated only. What do you think? Kind regards Darius

4th Mar 2006, 12:34 pm	#22
FERNSEH Dekatron Join Date: Mar 2005 Location: Gateshead, Tyne and Wear, UK. Posts: 7,444	Re: 1956 Pye colour TV set. According to the Black Book the frame blanking period of the original M-EMI system was 2000 micro-seconds= 20 lines per field, so it follows that 365 lines were active. As the aspect ratio was 5:4 we have 456 picture elements in the 84 micro-second video part of a line. It works out that the max video bandwidth will be 2.7Mhz. As time went by some of the characteristics of the 405 system were changed. 4:3 aspect ratio was introduced in 1950 and the active video of a line reduced to 80 micro-seconds to accomodate the characteristics of the 625 system. I'm not sure when the frame blanking period was reduced to 1400 micro-seconds. Anyway, all these changes will have resulted in the max video bandwidth being increased to 3Mhz. DFWB.

4th Mar 2006, 1:46 pm	#23
channel405 Tetrode Join Date: Jun 2004 Location: CH3, Blaen Plwyf Transmitter, Aberystwyth, WALES Posts: 72	Re: 1956 Pye colour TV set. Hi Darius, As I remember, and of course could well be wrong, one of the differences between PAL and NTSC decoders is that there is no delay line in the NTSC one, which would be a problem if we were to have PAL 405 cos the delay line is 64us instead of 98 as per 405 line duration. I am right in thinking then if we want colour 405, best to stick with NTSC? Could use CCDs I suppose. Tim

4th Mar 2006, 2:03 pm	#24
ppppenguin Retired Dormant Member Join Date: Dec 2003 Location: North London, UK. Posts: 6,168	Re: 1956 Pye colour TV set. The only colour system ever used on 405 was NTSC. PAL and SECAM were never even considered for 405. I don't think direct conversion of 525 NTSC to 405 NTSC would work. The colour subcarrier frequency would be wildly wrong. If you wanted a largely analogue approach there are 2 methods. 1. Decode to RGB or YUV, retime to 405 with multiple CCDs and encode to 405 NTSC. 2. Separate Y and C, retime to 405, modify the PAL to NTSC and heterodyne it to the correct subcarrier, add 405 Y and C back together. I'm sure both methods could work but I wouldn't be enthusiastic about them. David Robinson's approach is sound and proven. It might be possible to redesign the Xilinx code in Aurora as I mentioned earlier. The discussions about 405 from a PC are also relevant as the RGB output could be coded into NTSC.

5th Mar 2006, 1:08 pm	#25
Heatercathodeshort Dekatron Join Date: Jun 2005 Location: Warnham, West Sussex. 10 miles south of DORKING. Posts: 9,147	Re: 1956 Pye colour TV set. Hello, Certainly before the days of standards conversion the 405 system could and did provide a bandwidth of 3mc/s. If the signal was a good clean one as it certainly was in the Wimbledon area where I worked at the time, the full 3meg was reproduced on sets such as the PYE CTL58 series. In practise of course only a complete nutter would worry about the extra 300kc's or so but believe me some customers were very test card aware in those days and complained if all the 'bars' were not visible. Its difficult to calculate today how good the 405 service was. [In practise of course most sets were not up to reproducing them but many were] I recorded many tapes prior to the closedown of the service and its quite a shock today to see how good the end result was. I have a GEC BT1748 14" model in my collection, its in very good working order and engineers are amazed at the reproduction from this 50 year old model. I get very annoyed when even the BBC describe the system as 'Back in the flickering days of black and white'. A great deal of trouble and expense went into providing what was considered to be the best television service in the World and that included picture quality. Regards John.

6th Mar 2006, 12:12 pm	#28
FERNSEH Dekatron Join Date: Mar 2005 Location: Gateshead, Tyne and Wear, UK. Posts: 7,444	Re: 1956 Pye colour TV set. The approach is first find a dual standard CTV and modify the decoder to operate on 405 NTSC. The delay line can be removed. As for the sub-carrier oscillator, a 2.65Mhz crystal can be made or, as in the '56 Pye it can be a free running type locked to the bursts. The colour killer will need to be redesigned and the band pass characteristics of the croma amplifier will need attention. Modiying a 405/625 colour set has been considered by others. DFWB

28th Jul 2006, 4:00 pm	#29
FERNSEH Dekatron Join Date: Mar 2005 Location: Gateshead, Tyne and Wear, UK. Posts: 7,444	Re: 1956 Pye colour TV set. I mentioned some time back that the 23KV EHT generator employed in the Pye colour set was similar to the system found in the 1955 RCA 21CT55. For clarity, in the drawing of the circuit diagram I've substituted solid state diodes for the three DY86s. The 17KV from the line output transformer overwind is augmented with the 6KV from the anodes of the two parallel PL36s. Note the shunt stabiliser tube, the 6BD4. DFWB. Attached Thumbnails

28th Jul 2006, 4:11 pm	#30
ppppenguin Retired Dormant Member Join Date: Dec 2003 Location: North London, UK. Posts: 6,168	Re: 1956 Pye colour TV set. I ought to point out the mistake I have made in this thread. PAL and SECAM were both tried on 405 as described here: https://www.vintage-radio.net/forum/...&highlight=cox That 6BD4 is a nice chunky tube. Looks like an immediate ancestor of the PD500.