Discussion: does the number of IF tubes REALLY matter?

[miniman82]

I've often thought that the reason for the number of IF amplifier tubes in a given set was related to it's build quality, I.E. lower cost sets generally have less tubes in the strip. It is often pointed out that famous chassis such as the CTC-2/2B had 5 IF tubes, the reasoning being that more tubes equals better reception especially in weak areas. I've always accepted this to be true, since the logic makes sense- until last week.


I was getting some help from my brother ordering peaking coils for my 2B chassis (I hate Digikey's website), when I pointed out that it had 5 IF's. We agreed that it likely does have great reception, but we looked into it a little more. Turns out there's more to the story than first meets the eye, and I'll explain. Later chassis like the CTC-10 have less IF tubes, but there is a caveat: transconductance.

We started looking at tube data sheets for the tubes used in different chassis, when we realized that the early tubes had much less transconductance than ones used in later chassis. For example, the 6DC6 used in the CTC-2 has 5500 micro ohms, and there were 5 IF's. The CTC-10 has only 3 IF's, but the tubes employed have nearly triple the transconductance (like 6GM6 with 13,000 micro ohms).

I'm thinking that at the time chassis like CTC-2 were built, they were simply working with what parts they had available. Later on down the line, better tubes such as the 6GM6 became available. Because of their high transconductance, less of them were required because the gain was so much better. So less IF tubes should not automatically mean a 'worse' reception or poorer picture quality, I think it simply means that better parts to do the same job eventually became available and so they were used.

Thought?

[Zenith26kc20]

Interesting point, but the tubes have little to do with the pass band. The IF transformers (the more of em the better) would make the response more "tunnel visioned" (I know, bad analogy) as it would ignore (roll-off) signals outside the passband with more tuned circuits. Gain would be determined by AGC action on most strong signals.
I guess later IF transformers were improved compared to older ones but they would still require some form of stagger tuning and the same bandwidth to pass the required signals.
Interesting discussion

[Tomcomm]

Original Feb 3, 2010
Quote:
Perhaps WB IQ didn't work so well after the CTC2/B was due to the fact the later CTV had fewer IF stages than the original six! When they got down to three stages they couldn't get the flat response and precise uniform phase delay required for WB IQ without artifacts, so they stuck with NB R-Y B-Y? Rediscovering WB IQ really fascinates me and others. Sort of cultish, eh?......Tom
I checked all the RCAs from CTC4 to CTC20. the CTC4 had a tuner IF plus three chassis IF amps. the CTC5 had same 4 IFs. The CTC7 had no tuner IF and 3 chassis IF, 3 total. All the rest to CTC20 remained at 3 total IF amps verses 6 IF amps of the CTC2/Bs!. No wonder WB IQ was wasted on these last tube CTVs.......Tom

[ctc17]

Sensitivity would be in the front end/tuner. And that could have been part of it too, they didnt have the good rf amp tubes that early. (Did they have the nuvistor that early?)

More IF stages would add to the noise as well as decrease reliability.

Maybe its due to advances in coil winding?

Some of the later sets you could tell they were just cheaping out, like the ones with 1 IF tube. 3 is a good number, 5 is to many to align and troubleshoot.

I have a cheep0 AA5 radio that really a AA4, no IF tube, one IF can. It has really poor reception over the radios with the IF tube. It seems like the higher end transistor radios have more IF stages.

Some random thoughts.

[Penthode]

I would like to share a few of my thoughts.

The design of the CTC2 is interesting from the perspective of the state of the art in the early 50's. I suppose the extra wide IF response of the CTC2was to help ensure a relatively flat I IF response (3.58 MHz +0.5/ -1.5MHz) and Q IF response (3.58 MHz +0.5/ -0.5 MHz). However, the extra luma response was more or less lost due to the need to employ a 3.58MHz notch filter in the luma channel.

The balance between a marginal increase in quality vs cost of implementation weighed heavily against continuance of the I Q demodulator approach. Also it was found that a narrower video IF together with chroma channel which compensates for the sloping response at the video IF at 3.58 MHz made sense with the savings in manufacturing cost. Not until the introduction of proper 3 line comb filtering in the late 80's did the wide band demodulator approach fully make sense.

It is interesting to note that pre1953 Black and White sets with 10" screens or larger often had full 4MHz video IF. After the introduction of color, most Black and White sets IF bandwidth was limited to 2.5MHz to 3MHz. Wider IF bandwidths were pointless because all you would get is a heavy NTSC dot crawl. The heavy color subcarrier on wideband B&W sets would be rectified by the gama curve of the CRT and tend to darken the image hence distort the gama.

The second NTSC which was responsible for color knew that there was going to be a tradeoff: I recall reading that the NTSC thought that the loss in luma resolution would be more than offset by the addition of color. (I may of read this in Donald Fink's book on the NTSC?)

I agree that the increase in tube transconductance would yield the need for fewer tubes. The 6EH7/6EJ7 tubes later used in TV IF amplifiers, have a transconductance of about 15000 micromhos. Hence two tubes could be used where three once were. However the transconductance alone I do not feel is responsible for the change in design philosophy of color TV receivers.

Sensitivity was mainly determined by the front end S/N and to a lesser extent overall gain. Cascode amplifiers and later the neutralized nuvistor amplifier (both with triodes) generated less noise hence improved the S/N.

[old_tv_nut]

You have all hit on contributing elements, I think. The 5-IF designs allowed many tuned elements to shape the IF precisely. Too few stages and you do not reach the noise-limited fringe areas. The 2-stage Muntz and Motorola B&W sets proved that. It was also definitely a matter of economics, to produce sets that would provide service people could afford. The number of IFs also affected selectivity and ultimately how tightly stations could be packed geographically - a huge trade-off off spectrum efficiency vs. an affordable TV broadcast service.

Regarding the IQ demods, I have posted elsewhere that the IFs that had become standard practice by the 70s had trouble with quadrature distortion if you tried IQ demodulation - mostly due to effects of the sound trap and upper end. A technically viable (but economically and logistically impossible) solution would have been to retrofit all NTSC encoders with different Q filters that had traps at 920 kHz (baseband). This would eliminate the residual 2.66 Mhz and lower Q energy at IF, which is what caused the residual quadrature distortion. The NTSC fathers never thought of chopping off the lower Q sideband as sharply as the 4.5 MHz trap would chop off the upper Q sideband.

Besides fewer than 5 IF stages being necessary for gain, designers discovered that a haystack shaped IF would pre-peak the video at 2 Mhz, which is desired for apparent sharpness, and since this could be a phase-leading function producing pre-shoots (not customer adjustable), it would complement a simple R-C video peaking control that produces variable post-shoots. So, with the peaking (customer "sharpness" control) set to mid range, you would have a medium-peaked picture with symmetrical pre and post shoots. Different manufacturers made different decisions from time to time as to how much of this IF prepeaking was desirable. Eventually, Zenith and others developed analog video ICs that could make a symmetrical customer peaking adjustment, but still in conjunction with a haystack IF, which by then was being done with surface wave filters instead of discrete components.

[miniman82]

Very interesting stuff, Wayne. So you're saying that because IF stages had become more or less standardized by the 70's, there was really no point to even trying IQ demodulation? I suppose the economics of that style of demod led quickly to it's own destruction, especially if the IF stages had been changed for some other reason and could now no longer support it. Sort of a double whammy, huh? Now that I have had a chance to see both wideband and narrowband color sets, I can honestly say that my eye is not sensitive enough to tell the difference. To me, the 21-CT-55 looks just as good as my cheapo Philco 17MT80 if you ignore the Philco's 3.58MHz dot crawl.

I strongly suspect that as soon as I am able to get component input working on the 17MT80, it will rival the perfomance of many higher end units.

[Tomcomm]

Originally Posted by miniman82

........ Now that I have had a chance to see both wideband and narrowband color sets, I can honestly say that my eye is not sensitive enough to tell the difference. To me, the 21-CT-55 looks just as good as my cheapo Philco 17MT80 if you ignore the Philco's 3.58MHz dot crawl.

I strongly suspect that as soon as I am able to get component input working on the 17MT80, it will rival the perfomance of many higher end units.

I have been actively modifying my 21CT55 to process external component video since January 2010. The CTC2B chassis seemed the ideal candidate CT since it uses low-level matrixing and has potentially >4mhz bandwidth video CRT drivers. I have progressed to the level of working out all the required interfacing and have constructed one of three RGB channels feeding the three CTC2B CRT driver boards. I have all the notebook entries complete with waveforms to finish the dual composite / component switchover. I have posted my rational for this component video effort on at least three occasions:

10-26-2010 Pete D………. to answer your question: In the mean time, I will use the CTC2B chassis only as a hardware test bed and waveform only monitor for my experiments with external Component video inputs. Seems that the three components: Y, Pr,and Pb, are available on three of my DVD players. Y is the luminance with sync that has no 3.58mhz chroma messing it up so it can extend the baseband to 5mhz which equals over 400 lines resolution, where all our roundys cut off at 3.2mhz or 256 lines! I think I can extend my 21CT55's original final three stages of CRT Drivers to possibly 5mhz without chroma interference. It would be great to see the resolution chart wedge with its lines displayed to 400 lines without all the chroma artifacts we see now. Do any of you see a problem?

01-01-2011 My CTC2B chassis is not constrained by the "restoration" consideration and has never processed the deceased NTSC RF/IF programmed inputs. It has always performed as a pure composite video monitor, driven from DVD, Laser Disk, S-video VCR, C-Band Big-Dish and 4DTV Digicypher. At the time of its flyback meltdown, I was in the process of modifying it to accept Component Video which should extended its baseband luminance and chrominance to a full 5mhz or 400 lines vs the 3.3mhz or 264 lines of NTSC Composite Video. Now if only I can get a replacement flyback that would converge as good as my original CTC2B, I can proceed.

04-08-2011 Since my 21CT55 is not an original RCA restoration but a "laboratory test-bed", I saw no reason to copy RCA's external video input circuitry using early '50s technology. Instead, I implemented a solid-state video op-amp that provides a "perfect" DVD quality 480 line resolution, 75ohm , 5vpp signal into the 55's 1st video amp grid. It also supplies the same 75ohm video at 1vpp to my two comb-filtered Sony monitors and a Leader waveform monitor. Two external video sources can be selected with a toggle switch, each with individual level and offset controls. Presently I run a DVD player with the DVE test disk into one input and a second DVD player with program disks or OTA HD converter into the second input. Now that the 21CT55 's PQ is at least as good as it was before the tragic FBX meltdown I intend to resume my implementation of 480 line component video processing in addition to the present 280 line composite processing, hopefully.......Tom