Best forum for Kenwood TL-922 and 922A ?

Discussion in 'Amateur Radio Amplifiers' started by KM5L, Jan 26, 2021.

ad: l-rl
ad: Left-3
ad: L-MFJ
ad: Left-2
ad: abrind-2
ad: l-BCInc
ad: Subscribe
  1. W1NJC

    W1NJC Ham Member QRZ Page

    Thanks for the quick reply. Are you referring to Tom's suggestion of removing the wire from the NC terminal of RL-2-1? In an unkeyed state, this would result in the cathodes being connected to the CT and to ~80V through 100k instead of directly connected to 80V (and the CT). I think the idea here is this removes the path from that 80V, protecting the transformer if there's a G-C short...makes sense especially in context of his description of the need, or lack thereof of "cut-off bias". That connection to 80V is supposed to ensure that the tubes are cut off, but I guess that's not needed. Do I have this right?
     
    Last edited: Mar 31, 2021
  2. W1QJ

    W1QJ Ham Member QRZ Page

    You have it sort of right. Right now there is a connection from the relay coil (110vdc) over to the NC contact of the relay then there is a 100K resistor between the NC contact and the common along with a connection to the CT. At the NO contact you have the B-. What you do is simply cut the jumper from the relay coil to the NC contact. Then remove the 100K resistor connection at the NC contact and move to down the NO contact with the B- that is already there. That completes the bias mod. What you have now is NO connection at the NC contact. The B- and one side of the 100k resistor is at NO and at common you have the CT and the other leg of the 100K. So basically the circuit is like this: IN RX the B- is connected to the CT through the 100K resistor, this creates a "self bias" voltage due to the Vdrop across the resistor that cuts off the tubes. In TX the resistor goes away and the B- is connected directly through the CT with the bias diode. One last thing, you will notice a red wire that is still at the NC contact, that was meant to connect to the 110vdc for the ALC. You will now have to move that red wire over to the relay coil where the 110vdc is. Now you are done completely.
    Questions?
     
  3. KD9IQO

    KD9IQO Ham Member QRZ Page

    W1NJC:

    A picture (schematic) is worth a thousand words.
     

    Attached Files:

  4. W1NJC

    W1NJC Ham Member QRZ Page

    KD9IQO: Thanks, I've seen that drawing before but I'm trying to understand how it applies to this discussion. It involves a lot more than what we're talking about here. What am I missing?

    W1QJ: I'm trying to follow what you are saying, and probably making it more difficult than it needs to be. I appreciate your patience. In the green circle below is an attempt to capture your description.
    Is this what I'm looking to do? I realize that they are utilizing the NC terminal to make a lot of the connections to the "high side" of those coils & ALC and that's not well depicted in the schematic, but is this electrically correct?

    upload_2021-4-1_21-20-54.png
     

    Attached Files:

  5. KD9IQO

    KD9IQO Ham Member QRZ Page

    W1NJC:

    Whenever an arc-over (glitch) occurs, the TL-922 bias Zener dies. Sometimes, the filament transformer dies as well. The filament transformer is no longer in production. Several months ago I witnessed a TL-922A filament transformer sell for over $300 on eBay. Your goal should be to protect the tubes and circuitry if (when) an arc-over occurs. I suggest you disconnect the cut-off bias circuitry and use self-biasing. The circuit board sold by W7RY ($20) will provide the following functions:

    (1) adjustable variable ZSAC (zero signal andoe current)
    (2) provide self-biasing for cutoff
    (3) protect your negative rail (diode protection)
    (4) provide LED drivers for meter lighting and status lights
    (5) provide ALC voltage
    (6) provide soft-keying for modern transceivers

    Once that is accomplished, add the gas discharge tubes to the tube sockets. These provide an immediate path-to-ground for the arc-over current and protects the transformer and meters. The GDTs actually bypass the meters and transformer once triggered.
    Then add the 50Ω resistor and the 0.8A high-voltage fuse (with blast shield) to the anodes.

    *The W7RY board is a blank board. You must purchase the component and assemble.

    **Hammond makes a suitable replacement filament transformer (265U10) should the unthinkable happen.
     
  6. G3YRO

    G3YRO Ham Member QRZ Page

    Personally, I have just removed the Zener, and put a string of 3A diodes in its place to provide the correct Standing Current for SSB. Then used the SSB/CW switch to short some of them out on CW, to reduce the Standing Current on CW.

    To protect the PSU and the Filament Transformer in the event of a flashover, I've just put a 1A fuse in the line that goes to the Transformer centre tap.

    Roger G3YRO
     
  7. W1QJ

    W1QJ Ham Member QRZ Page

    Yes the section inside the green circle is the mod. You can see the difference compared to the old way. The ALC needs to be resorted by moving the red wire over the pick up coils DC voltage. Indeed the W7RY is great but certainly more involved. Never a need for QSK unless you are a big CW guy. Basically a waste for a phone operator. I’ve install both W7RY and Keesler QSK in 922 amps.
     
  8. W1NJC

    W1NJC Ham Member QRZ Page

    Thanks guys. I'm considering my options including the W7RY board(s) and a HV vacuum relay. I'm not a big CW guy but better relays and timing is interesting to me. I've already made my own soft-start and keyline mods, but the W7RY board is pretty inexpensive and built to fit so I may implement it.

    73 and thanks again,
    Nick W1NJC
     
  9. W1NJC

    W1NJC Ham Member QRZ Page

    @W1QJ/@KD9IQO: I have the W7RY board, his input relay board, and a vacuum relay all just about ready and am getting close to installing them. Two things I plan to do differently are not replace the lamps with LEDs, and add a bias point for a CW/SSB switch. The lamps might be under powered or I may need to change some R values. I also plan on using a small relay for the TX/STBY lamps. The bias point is straight-forward; I cut a trace on the board and just need to experiment with the pot to determine the correct fixed value to add in.
    So a couple questions for you guys who have installed the mod on a TL-922 before:

    1) Have you ever kept the stock lamps and how did that go?
    2) Were you able to mount the board without cutting the filament transformer bolt or at least not removing it entirely? It looks like this is possible if I bend the leads of Q7 back on themselves...but it's still close.

    I was about to take the amp of our service and get it on the bench but I figured I'd ask on here for some tips first.

    Thanks in advance,
    Nick
     
  10. G3YRO

    G3YRO Ham Member QRZ Page

    On my TL-922 I set the standing current at 100mA for CW, and 180mA for SSB. (obviously with full HT on both modes)

    Roger G3YRO
     

Share This Page