On Sun, 02 Mar 2008 13:47:07 -0500 Michael A. Terrell wrote: | snipped-for-privacy@ipal.net wrote: |> |> |> | Sigh. The reasons are VERY well laid out in older TV design |> |> | handbooks. Maybe a little reading will open your eyes? |> |>
|> |> Are you talking about before or after color? |> | |> | |> | Both. The library at Cincinnati Electronics had all the books from |> | the original Crosley engineering department, along with all of the IRE |> | and IEEE papers on Television, and covered every system that was |> | presented to the FCC, and ALL of the test results. |> |> So there are books that talk about why the particular frequency was chosen |> for the color subcarrier, before there was color? | | | Yes, ones printed during the development and the deployment of | color. They went into great detail about the problems expected, and the | changes needed to prevent them. | | |> |> I've read the books. I wonder if you ever did. |> | |> | |> | |> | Sigh. No. of course not, you dumb ass. No one but you has ever read |> | them. They were written just so you could show off to everyone. That's |> | why i have a nice collection in my personal library. I always spend |> | lots of money on books I don't read. |> |> At least you are being honest. I have only one book that deals with TV |> technology. The rest I have read from the library. | | | Once again, sarcasm goes right over your head.
Who's sarcasm is the more subtle, eh?
|> | 5 and or 10 MHz have been the in house reference for decades. The |> | first frequency counters acceptable for TV use were built with ovenized |> | oscillators that produced at least one of these frequencies. The most |> | common counter was the HP 5245L with the proper front end plug in. |> |> And? | | | And what? Either ask a question or shut up, troll.
How does what you said apply? Why don't you connect it to what was being talked about?
|> | Why is it people like you, who have NEVER done the work talk down to |> | those who have? |> |> What work? Have you _designed_ a complete TV encoding and transmission |> system from the ground up? Can you even do the Fourier transforms (among |> other things), needed to understand the signals and spectrum energy needed |> to make the design effective? | | | Phil, some people are intuitive, and can see how things work. Others | need a pencil to take a crap.
You also need some paper with that pencil.
|> |> I already said the FCC requires it be plus or minus 10 Hz. |> | |> | |> | You also said that it was +/- 10 Hz at 14.318180 MHz, when it is +/- |> | 40 Hz |> |> The FCC requirement of +/- 10 Hz is for the on-air subcarrier. Do the |> math to figure out what it needs to be for other frequencies you might |> derive the subcarrier from. | | | I did. You are the one who claimed it was still +/- 10 Hz at | 14.318180 MHz
Fine, whatever you say.
|> If you think I do that, then be specific and to the point. There is no |> need to make personal attacks. You have done that a lot, as have a small |> handful of others on Usenet. One of them even posts here a lot. | | | Did you ever think that they are doing because you are wrong?
Did you ever think that maybe you ought to just point out specifically what you think is wrong, when someone posts something you think is wrong, and include what you think is right? And do that post as a direct followup to the specific post that has what you think was wrong.
|> | You really have no clue, do you? In most burst circuits, more that |> | two cycles difference, and it will not be pulled to the subcarrier |> | frequency. Do you have a studio grade sync generator, a broadcast |> | quality waveform monitor, or a broadcast grade vectorscope? |> |> You seem to be the one with no clue. | | | So, you have never looked at what your design is capable of? That is | exactly what I expected
Why would I have broadcast grade studio equipment at home?
|> Two cycles difference of what? Or do you mean 2 Hz? Well, I have news |> for you ... an oscillator that would naturally oscillate at 2 Hz from |> the transmitted signal can be pulled to that signal. Sure, it will |> slip between burst pulses. But at 2 Hz difference, it's not that much. |> It would be about 0.04576 degrees of phase by the time the next burst |> comes along. You wouldn't even notice the color shift from left to |> right. | | | Ok. sure. yeah. | | | You have no clue, phil. The chroma would change from the left to the | right side of the screen if the burst oscillator isn't closer to the | expected frequency. You don't think so, but I've worked with several | video directors who could see that across the room. The burst is used | to fine tune the phasing, and the tint control is used to manually trim | it. It sets the center frequency, and if you were right, there would be | no way to set the tint.
Yes it would change. I never said it would not. But I did the calculations and the amount of change (0.04576 degrees of phase) is so small it would most likely not even make a one bit difference if the resultant color was digitized.
Maybe you are talking about an oscillator that is off way more than 2 Hz?
| What directions? A lot of components were obsolete, and very little | documentation had survived over the years. I had schematics, and parts | lists with RCA stock numbers, but RCA was out of the broadcast | business. It required thinking on your feet, and being able to redesign | some stages to work. Tell me where you would find a RF component that | hadn't been built in 20 years, and the old one was burnt beyond | recognition? What would you do if some dumb ass had brazed the custom | made brass fittings in the cooling circuit to the copper pipe, over some | bad solder. Without them, the transmitter was scrap.
And did you have to do math? Vector math? Trig? I wonder what part you would have failed at if you had been called on to do it.
| Small potatoes. Have you designed an FQPSK encoding system, and the | decoding system? Hell, I've designed and built test fixtures that were | more complex. NTSC encoders were done with a handful of tubes and a | delay line for the sync. have you ever designed video amps with a 3 dB | point at 40 MHz, and less than .5 dB ripple over the entire pass band?
And yet you don't know how many degrees of phase change take place between two sine waves only 2 Hz apart in frequency over the time a one video line?
| I suppose you'll be bragging abut designing buggy whips, too? The | large semiconductor manufacturers have obsoleted their NTSC chipsets, | because of HDTV. No one will be designing any new NTSC encoders. Have | you designed and built a low phase noise synthesizer to track deep space | probes? It operated in three adjacent segments to cover 370 MHz to 520 | MHz continuous. All the math in the world wouldn't predict all the | quirks in a design like that. A lifetime of experience does. Simple | things, like changing from an uncased disk capacitor to a SMD part of | better quality caused the phase noise to shoot through the roof. All | the math in the world wouldn't explain it, but being VERY familiar with | RF PC board design made it obvious. Something that the other engineers | overlook, because they had no hands on experience with that design. The | large Vias that had been used to mount the uncased ceramics had to be | filled, all the way to prevent them from being low value inductors in | the ground plane.
Sure, experience helps. But getting math wrong can still destroy any design project.
|> No I have not matched a set of 16 6146 tubes. This relates to understanding |> the color subcarrier how? | | | Phil, you don't understand much of anything. If that distributed | amplifier isn't balanced, the chroma doesn't get to the modulator. The | burst was the highest frequency passed by the video amplifier, and | mismatched tubes cause a loss of response at the higher frequencies, and | some phase shift.
So if those 16 tubes are not perfectly matched, no signal gets out at all? Maybe you need a whole different approach.
Why are you using 6146 tubes to pass chroma? Oh, wait, you didn't design this.
|> There's plenty of real work that involves not knowing anything about |> waveforms, signals, or even mathematics. It obviously shows from some |> of the errors you've made in posts that you think you are so great |> because you've have your hands on all this stuff. But you couldn't |> create a mathematical model for how it works. | | | So, you would spend month reinventing hundreds of wheels, rather than | use what you already know? I went through the paper and pencil math | phase at around 13 years old, but I'VE outgrown it. A mathematical | model doesn't do any work. A piece of working equipment does. Your | models don't deal wth real world issues, only what would be in a perfect | universe. We designed and built $80,000 telemetry receivers without all | your excessive math, and they worked so well that we couldn't keep up | with the orders for a year or more. How many multi-million dollar | contracts do you get from your anal retentive math?
Based on the math you've show you can do, I don't want to be anywhere near those hazards.