On Chroma Dots and how to determine if they are present

[VintagePC]

So I'm curious to know - is there an easy way to tell whether what I'm seeing on my set could be chroma dots or just a different type of noise which appears similar?

I haven't really been able to find good examples of what they look like in a moving picture, just still images.

Can it be determined by examining the schematic/circuitry whether they can and/or will be visible?

Alternately, I have a VCR attached to the set with composite inputs. If I were to feed it the luminance only of an S-video signal and observe the output for the dot patterns, would that work or does the internal VCR circuitry (e.g. OSD generator) re-insert colour information into the signal?

I realise it's an obscure set of questions but I'm curious to learn and start a discussion on it

[Eric H]

I thought Chroma Dots were something that the PAL system had but not NTSC?

I don't remember the details but I read that there were some B&W video tapes of what were originally color Dr Who episodes using the dot pattern to determine the correct colors.

http://www.techmind.org/colrec/

[VintagePC]

Originally Posted by Eric H

I thought Chroma Dots were something that the PAL system had but not NTSC?

I don't remember the details but I read that there were some B&W video tapes of what were originally color Dr Who episodes using the dot pattern to determine the correct colors.

http://www.techmind.org/colrec/

NTSC definitely has it, it's visible on sets where you have more than 3.5MHz bandwidth, since the colour subcarrier is at that frequency. If I had a bandwidth (multiburst) generator I could answer my own question easily as to whether it is possible to see them on my set - alas, I don't, so I'm curious if there's a cheap way to determine this when there is the possibility of other noise sources (e.g. dot crawl)


Bob Andersen talks about it when running some of his sets after alignment and restoration as well.

[wa2ise]

Originally Posted by VintagePC

NTSC definitely has it, it's visible on sets where you have more than 3.5MHz bandwidth, since the colour subcarrier is at that frequency.

The chroma subcarrier has 227 1/2 cycles per horizontal line. Most B&W TV sets made after 1954 limit the luma bandwidth to suppress the chroma subcarrier. Usually done by the way the picture IF strip is adjusted. Older sets can show the entire 4.2MHz of luma bandwidth as there was no chroma subcarrier added in back then.

The chroma subcarrier looks like a fine checkerboard moving vertically in areas of the picture with saturated color. The size of this checkerboard pattern is such that you could have about 330 checkers in a horizontal line (two checkers per cycle of subcarrier).

[bandersen]

Here's a close up of chroma dots on a 1949 Admiral TV with a wide video IF bandwidth.

[walterbeers]

I've seen that effect on some older B&W TVs, looks like 4.5Mhz beat, but adjusting the 4.5 trap never seemed to help. Always thought it might be some beat interference from the color burst or chroma signals. I've noticed it more now since I have to use a digital converter to pick up broadcast on the older TVs. Interesting!!

[VintagePC]

Interesting. What I'm seeing does look like the image posted by Bob, yet my set is early 60s and judging by the expected IF response (http://vintagepc.site50.net/site/fle...ice/TV_009.png) it looks like it should have 3 Mhz of bandwidth at 50%... which indicates I probably am not seeing chroma dots. On the other hand, it's quite possible it's not aligned to factory specifications given I'm getting bleed from the video signal into the audio (buzzing when bright white text on the screen) but that's an audio trap/alignment issue. One of these days I'll probably splurge and get a VA62 and mod it for 21Mhz IF so I can do something about that.

I'll try to get a good picture of it on my set next time it's playing; I'll see if I can find an old tape or something to feed in a still image so it's easier to snap a pic of what I'm attempting to describe.

[old_tv_nut]

This thread is confusing. Is your set an early 60s black and white set, or was 60s a typo?
1) An early 60s set would not have a 21 MHz IF. What do you have?
2) A black and white set might roll off the high frequencies, but would not have a specific trap for the color subcarrier, so it could have chroma dot crawl to some degree even if it's not full 4 Mhz bandwidth.
3) you seem to differentiate chroma dots and dot crawl. They are the same thing. The only difference is that in color sets with a subcarrier trap in the luma channel, the dots in large areas of the same color were eliminated, but dots still appeared on the left and right edges of colored objects. In later sets with a comb filter, depending on the design, color dots could appear inside the top edge and just below the bottom edge of colored objects, and were called "hanging dots." Sophisticated comb filter circuits included logic to switch in a chroma trap at these transitions. Fast motion could also fool the comb filter into letting dots through in some cases.

[VintagePC]

To answer your questions:

1. It has a 21Mhz IF. I refer to it as an early 60s set because I recall finding that date code either somewhere on the chassis or in the accompanying documentation. I suppose it could be late 50s; I haven't found _any_ info on this set online barring a similar-looking but not entirely identical picture from a 50s sale brochure (the 21-41): http://www.tvhistory.tv/1950s-CAN-Fl...-Brochure3.JPG. Someone with an RCC Master Index might be able to pin down an actual year.

I'll have another look and see to confirm or deny. I wouldn't be surprised if there were still 21Mhz IF sets in Canada later than most of the US; we were pretty behind with FM adoption in tube radios too (not all that relevant, I guess) and the manufacturers might well have opted to save some re-design costs by extending the 21Mhz design a bit longer...

3. You're absolutely right; I don't think dot crawl is the term I was looking for. I was reading about other types of noise and how one of them could be mistaken for chroma dots, but the name escapes me. Dot crawl must have been the first thing that latched into my mind.

[lnx64]

I run all my black and white sets through the Luma line of s-video or from component when set at 480i.

[VintagePC]

Originally Posted by lnx64

I run all my black and white sets through the Luma line of s-video or from component when set at 480i.

I gather you have an RF modulator of some sort, or did you modify your set to allow for a non RF video signal?

[old_tv_nut]

Originally Posted by VintagePC

To answer your questions:
...I was reading about other types of noise and how one of them could be mistaken for chroma dots, but the name escapes me...

Thanks for the clarifications. 4.5 MHz sound carrier can look a bit like chroma, but it dances around and changes rapidly from vertical to diagonal to dots (possibly too fast to see) as the sound modulates the frequency. It is much more stationary during silence. This is suppressed by sound traps providing the fine tuning is correct. Even when alignment is perfect, if you tune toward the high end of the channel, you may be able to see the sound carrier as it comes out of he traps, provided the IF and video aren't too narrow to pass it.

[lnx64]

Originally Posted by VintagePC

I gather you have an RF modulator of some sort, or did you modify your set to allow for a non RF video signal?

Rf modulator.

[Electronic M]

IIRC Canada was slow to adopt color too so the IF bandwidth may not have been designed to eliminate a chroma dot issue the set likely would not have had to deal with when new.

[VintagePC]

Originally Posted by old_tv_nut

Thanks for the clarifications. 4.5 MHz sound carrier can look a bit like chroma, but it dances around and changes rapidly from vertical to diagonal to dots (possibly too fast to see) as the sound modulates the frequency. It is much more stationary during silence. This is suppressed by sound traps providing the fine tuning is correct. Even when alignment is perfect, if you tune toward the high end of the channel, you may be able to see the sound carrier as it comes out of he traps, provided the IF and video aren't too narrow to pass it.

You may also be right on the date. I was able to peg the turntable (Collaro RC456) to around 1956 or 57.

I re-read my notes from the restoration and I now recall the CRT seems to have been a rebuild/replacement; I suspect very much the date I found was the date code on the CRT label, which would account for the discrepancy, unless those RC456s were in use for a long period of time.