New Modulator

Discussion in 'Amateur Radio' started by K4TQF, Oct 15, 2017.

  1. K4TQF

    K4TQF Member

    There has been much discussion on this forum and info gleaned from the archives of other forums about my RCA mod iron.(901765-501) Well, perusing ebay last week I found one in it's natural habitat. It is being shipped this week.
    Uses a pair of 813's.

    s-l1600 (2).jpg
    s-l1600 (3).jpg
    s-l1600 (1).jpg
  2. Radio-Tuber

    Radio-Tuber Member

    I LOVE that AM/CW handle! Simplicity: This is why I love old Mil Gear.
    I'll sacrifice a cold 807 as an offering to the shipping deities. Hope it gets to you intact, OM.

    73 DE JIM K6FWT
  3. W5HRO

    W5HRO Administrator

    Looks like you were told wrong by certain individuals. Below is the correct info.

    1500 VOLTS AT 300MA AND 115/l/60.

    Look for the 50055 speech amp unit :icon_thumbup:
  4. K4TQF

    K4TQF Member

    Brian: you must have special access to Google. A search of all of those model numbers turned up only a couple of pictures of the speech amp, nothing on the T8K series transmitters
  5. W5HRO

    W5HRO Administrator

    No, just look at page 14 in the attachment. I highlighted the one for CRV-50193.

    As soon as you posted that CRV-50193 number I was able to find it and it goes with the T3K transmitter and not the T8K.

    250W will fully modulate a 500W final too. Be careful of what you are told via the other boards. They are often just full of shit due to their big egos.

    Attached Files:

  6. K4TQF

    K4TQF Member

    The consensus was that RCA iron I have was used in a MCW only transmitter and the freq response was crap. Well, this modulator uses the identical RCA iron... go figure/
  7. W5HRO

    W5HRO Administrator

    You will also see on page 5 of that sheet...

    CRV-50055 (CRV-50064)
    FREQUENCY RANGE 300 TO 3000 KC (probably a misprint)

    Here's the CRV-50064. It looks like the CRV-50055 was the speech amp inside of it or just the bottom part.



    Design. - Navy.
    Audio power. - 3 watts in 500-ohm load.
    Power.- Supply 110/1/60. Required : 125 watts. Allowable variation in supply line voltage +- 10 %, frequency +- 5 %.
    Input impedance. - 600/200 ohms
    Output impedance. - 500 ohms.
    Overall frequency response.- 300 to 3,000 cycles with +- 2 db of 1,000 cps. value.
    Modulation capability. - Up to 85 percent
  8. K4TQF

    K4TQF Member

    I think that doc was OCR. Apparently the addition of the CRV-50193 & Speech amp make a TBK transmitter a TBM. There is some confusion with the info posted on the internet because Westinghouse also made a version of this transmitter and accessory modulator.
    That document you found is correct for the RCA version of the CRV-50193. I'll post some info on the mil-surplus list and see if those guys can come up with a manual. BAMA and the other usual sites don't show anything.

    As seen below the Westinghouse is nothing like the RCA 'cept for the speech amp. Prolly painted Navy gray during a depot rebuild in the 1950's.

    IMG_0079.jpg TBM modulator.jpg
  9. W5HRO

    W5HRO Administrator

    I went back and looked at you RCA graph again and if by some chance it is right then the attenuation at the high end is not too far below the "0" reference point.

    What I would do is just build a good speech amp with a very high end response to compensate, but for a bit higher than the 3 watts. Make it more like 5 to 10 watts to get a full 100% out of the modulator. The output Z will need to be 500-ohms though unless you modify the input of the modulator to accept a high Z input.
  10. W5HRO

    W5HRO Administrator

    P.S. To keep it simple if you don’t have a matching transformer for 500-ohms out is just use a pair of tubes in push-pull connected as cathode followers with the resistors to ground. That would basically work as a buffer to couple to the 500-ohm input of the modulator.

    The CRV-50055 used a pair of 2A3's to drive the 813’s in the CRV-50193 modulator. One thing you could do is use a pair of 6AS7’s which are like dual 2A3’s/6A3’s, but they have separate cathodes and are not directly heated which would be perfect. Just use the second half of each tube as the cathode follower buffers to the modulator :icon_thumbup:


    The 6AS7's are also low mu so you can resistance couple to their grids if you want without using any interstage transformer. Just bias everything up to get at least 5 watts out.

  11. W6MQI

    W6MQI Member

    Where did you get those pictures that guy looks familiar I think I've worked him on the west coast military collectors group he has some really nice Navy radios.
  12. W5HRO

    W5HRO Administrator
  13. K4TQF

    K4TQF Member

    Well that's not a big deal. I have the preamp in the "Norton" & all sorts of xfmrs to match impedance.
    unfortunately, I sold all my 2A3, W.E. VT10, 25's, WE 300B ( 1937 engraved base ) etc... to the audophools years ago. The Norton has a "standard lineup of 6SJ7, 6SL7 6J5 & 6N7 . I should have several drawers full of audio xfmrs.
  14. W5HRO

    W5HRO Administrator

    I wouldn't use any transformers, they suck and attenuate frequency response. Run the 6SJ7 as the speech amp, then into the 6J5, then that into the 6N7 as the dual inverter amp to the grids of a pair of 6AS7’s. You can find a NOS pair of 6AS7’s fairly cheap.

    If you used the second half of each 6AS7 as cathode follower buffers and did it right you might even be able to bypass right to the grids of the 813's for a bit higher Z than 500-ohms. If not the 500-ohms would be simple to achieve using 250-ohm 6AS7cathode resistors to ground for 500-ohms p-p. The whole point is trying to get the best high-end frequency response possible due to the way the RCA iron is. If you use matching transformers you will lose it.

    P.S. Check to see what voltage they were running the 813's at. If you used a single 4-400A at a low enough plate voltage then it would be around 6K which would probably work fine on the RF side.
  15. W5HRO

    W5HRO Administrator


    I took one of my old drawing and quickly modified it. This it what I was referring to. If that modulator has a transformer inside to take the 500-ohms to the 813 grids then just bypass it and do the below. You can use two 1.25K resistors with the negative bias fed thru it's middle where they are connected together which will simulate the secondary winding with it's center-tap . The below circuit uses NO transformers at all which will always work better.


    Anyway, just an idea. You already have the 6SJ7, the 6J5 and the 6N7. All you need are two 6AS7's with the second half of each tube connected as cathode followers.
  16. K4TQF

    K4TQF Member

    Looks good. I'll need to check my tube caddies and see if I have any 6AS7s. There are two octal sockets on that chassis that were formerly used for the output tubes. There is a Chicago output transformer on there now, but I still have the multi-tapped Kenyon that may have a 500 ohm output on it.

    Wait... I have a new pair of 6080's on the shelf...
  17. W5HRO

    W5HRO Administrator

    Yep, 6080's are pretty close or the same and should work too.

    I bet they are running the 813's around 1500V so you probably couldn’t use a single 4-400A for the RF final unless they are running the 813's at 2000V. If so then you could. The 4-400A at 2000V and enough current would be around 6K and would probably work. If not then just find the right tube to give you around a 5K to 6K RFPA load on the secondary. A 4-400A must be run at 2000V minimum.
  18. K4TQF

    K4TQF Member

    here's another look at the RCA iron without load... white noise in and analyzer on the output. I'm sure that bump around 60HZ is just junk from the test leads laying on the bench.

  19. W5HRO

    W5HRO Administrator

    What's it look like with a 5.5k load on each side? Is that the new unused one or the one currently in the modulator?
  20. K4TQF

    K4TQF Member

    This is the NOS RCA xfmr... I haven't dug into the modulator, other than cleaning switch contacts, etc....

    With 5600 ohm resistor... the response is almost inverse. -10 @ 2.5KHZ -20@ 20K. The light blue peak hold line is where I swept my oscillator from 20 to 20K. Pay no attention to the lower solid bars, those are just ambient noise from the built-in mic when I took the screenshot. I am measuring using the line audio input direct .
    Batteries died in both of my white noise generators and I cannot find my old 30 pin charging cables.

  21. W5HRO

    W5HRO Administrator

    Mike, what's it look like with 11K on the input and output then with 11k on the input and 5.5K on the output? There is a reason why I am asking.
  22. K4TQF

    K4TQF Member

    10K input 5.6K output:

    11K in 5600 out.JPG
  23. W5HRO

    W5HRO Administrator

    And the reason why I'm asking is because a pair of 813's in the modulator is not 5.5K at 1500V. They are more like 10K plate-to-plate. If slightly above 1500V then 11K would be right for the primary.

    What I thinking is like before, the 5.5K on the primary could be only half. It might be for just for one tube to the center-tap, unless they are running the 813's below 1500V. What plate voltage are they running those 813's at? If it is 1500V or higher then 11K would be correct for the primary and then the next question would be is the 5.5K on the secondary correct or is it 11K too?
  24. K4TQF

    K4TQF Member

    11K in 11K out: follow the peak only... the solid bars are Tim prattling on 3.885

    11K in 11K out.JPG
  25. K4TQF

    K4TQF Member

    Yeah... no book on the modulator , but I think they were running it at 1500V.