Results 1 to 21 of 21

Thread: Wacom WP-735-1 Vhf disassembly. HOW?

  1. #1
    PRC148 No Longer Registered

    Default Wacom WP-735-1 Vhf disassembly. HOW?

    I have a set of Wacom 6 cavity filters from the 2 meter Vhf repeater site I brought home. They are about 30 years old but in decent shape (or so they appear). We've had issues with frying noises and de-sense lately. A friend says he has encountered this with the same type of filters due to corrosion or lightning arc damage between the fingers and the Invar tuning rods. He suggested swapping rods between the TX and RX sides so the finger stock holders would align in different spots on the rods. I'm trying to do that now but I don't know how to actually open the cans.

    This is a picture of one

    20130423_201415_zpsb818bed2.jpg


  2. #2
    Join Date
    Jan 19, 2013
    Location
    In Your Network
    Posts
    2,730
    Thanks
    3,063
    Thanked 1,965 Times in 992 Posts
    Country: Holy See

    Default

    You may have great difficulty as most of these are welded or soldered together depending on what material they are using. I would say swap the entire cavity, but keep the loops matched to the TX side or RX side.

    This should change **where** the rod aligns inside the can, thus hopefully finding an undamaged sector to tune against. If the can is that damaged, you may still not have good results. Cavities should ideally NEVER be tuned under power, and most especially under high power, but it happens all the time....

    Don't forget to keep the stubs associated with the TX side associated with the same side of the duplexer as well. Same with RX. I am assuming you have access to AT LEAST a good tracking generator and return loss bridge or a network analyzer?

  3. #3
    PRC148 No Longer Registered

    Default

    Unreal

    I can't believe I didn't think of that.

    Thanks. That's what I will do.

  4. #4
    Join Date
    Jan 19, 2013
    Location
    In Your Network
    Posts
    2,730
    Thanks
    3,063
    Thanked 1,965 Times in 992 Posts
    Country: Holy See

    Default

    Good luck. Chances are the cavities are NOT damaged, just mistuned. You may have a site problem also. Tune carefully, go slowly, and *think* through each step to make sure you get pass and reject done CORRECTLY. Then double check it.

    Then do it AGAIN once you move the cans (like eggs) to the site.

  5. #5
    PRC148 No Longer Registered

    Default

    I have an P8924 with a minicircuits return loss bridge I used last time but I don't know for sure if I'm seeing the level of detail I should. It seemed to work pretty well. But you are definitely right about movement. I noticed it doesn't take much to knock them out of alignment. The problem has been with power above about 50 watts. It seems to desense above that. We get a frying noise in the system and the sensitivity really drops off. I wonder about the old ass Ts and the cabling. It has the original RG-214/U interconnect cables on it. They look good but who knows. I'm sure this things has probably been nicked by lightning a few times too in its 30+ years.

  6. #6
    Join Date
    Jan 19, 2013
    Location
    In Your Network
    Posts
    2,730
    Thanks
    3,063
    Thanked 1,965 Times in 992 Posts
    Country: Holy See

    Default

    At this point it wouldn't hurt to find an Elmer near you with access to modern gear to do a once or twice over on the cans and cables as a system, and do a thorough look at the cables. If this is a ham box I am sure someone in your area is ham friendly enough to do this on the weekend or on there lunch hour. This is what I do. Of course, I own my own HP8920b and R8000.... (I spend money on radios instead of a wife - you can turn the radios off!) The 8924 is good for that if its calibrated, but lugging it to the site is going to be a chore. Make sure you get a good -90 or better in the notches. If you have 6 cans, I'd want to see closer to -100 in the notches when its all together.

    Move those cables around. Are they noisy? The connectors or shielding might be compromised. Look at wideband noise on your transmitter too. The cans might not be the problem.

  7. #7
    Join Date
    Mar 14, 2012
    Posts
    17
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Country: United States

    Default

    I've had trouble with the lexan tuning rods causing issues as they age. I know they aren't supposed to but I have a hyrbrid ring duplexer that is soon to go away that I finally traced everything back to the lexan tuning stubs going crazy and doing random drifts off frequency. This thing was made in the late 60's and has remained in service till now which still amazes me. I'm not real sure where to source new ones and may not be your problem but given the age of the duplexer it's a posibility as well.
    Keith Dobbins N8KLD
    Mainframe Network Engineer/
    Managed File Transfer Engineer
    W8TAP Repeater Engineer
    Parkersburg, WV 26104

  8. #8
    PRC148 No Longer Registered

    Default

    That's good to know. I hadn't thought about that being an issue. I just figured there really wasn't much there to go wrong with a plastic rod but I will definitely look at those as well. Thanks.

  9. #9
    PRC148 No Longer Registered

    Default

    I was looking at this thing on the HP8924C again. One side appears to be stable. I was looking at the other side and the notch freq above the pass band appears to randomly drop at times. It will dip down about 20 db then pop back up. I can't figure out what the cause it. Moving it or messing with it any any way physically doesn't seem to change the behavior. I'm wondering about the notch rods and loops now. My problem is I can't get the loops out to check them. Everything seems sealed up inside. I don't see any bolts to take anything apart. There are a few on the top that hold the RF connector but that's it. It doesn't allow the loop to come out since it's connected to the side RF connector where the notch rods go. This thing is really frustrating me. The dips I see on the analyzer do seem to correlate to what I've heard on the air from what I can remember. I wonder if there's a bad solder joint at a loop somewhere.
    Last edited by PRC148; Jul 20, 2013 at 12:46 PM. Reason: coherent thoughs

  10. #10
    Join Date
    May 09, 2012
    Posts
    162
    Thanks
    14
    Thanked 28 Times in 19 Posts

    Default

    Hope you figure it out. I have one doing the same thing. We finally replaced it to fix the problem. It would be nice to get the old one working.

    Joe

  11. #11
    PRC148 No Longer Registered

    Default

    I was able to remove the loops. They all appear fine. The solder is strong and clean. I noticed the tuning slugs don't appear to be centered in two of the tubes as if they were dropped at some point. I don't know if that's relevant or not. The SO239 type connections are old but not corroded. If you pick up the small loop assembly the 239 center pins will jiggle a bit in the center of the connector assembly. Again I don't know if that's relevant since the parts are mechanically stabilized when bolted in place.
    There is a very slight amount of corrosion in the bottom of the cans. It appears it is from years of humidity doing its thing. If I could get them open I would try cleaning them. I wonder if it's worth replacing all the RF connector hardware?
    Last edited by PRC148; Jul 21, 2013 at 03:15 AM.

  12. #12
    PRC148 No Longer Registered

    Default

    I took the loops out and cleaned, resoldered and dremel brushed them plus their mounting locations. This is what I'm seeing on the HP8924C. Does this look correct? I don't see the sharp lines like they reference in the tuning documents. Is that a matter of equipment and settings? The generator is set at 10 db and the receiver is set at -50. I'm trying to analyze the notch at 147.390 (tx freq) here. The pass-band is 147.990 at the top there.

    This is looking through the three cans on the TX side with the RX side only connected to the antenna T.

    10 MHz span

    20130723_223606_zps6a76600b.jpg

    .5 MHz span

    20130723_223756_zpsfa2c18ea.jpg
    Attached Images Attached Images
    Last edited by PRC148; Jul 24, 2013 at 04:59 AM.

  13. #13
    Alpha's Avatar
    Alpha is offline T S - Moderator
    CS Forums $upporter
    Join Date
    Feb 12, 2012
    Location
    Directly above the center of the earth.
    Posts
    2,780
    Thanks
    1,235
    Thanked 1,531 Times in 738 Posts
    Country: Christmas Island

    Default

    Make sure you don't forget to terminate the unused end port with a 50-ohm dummy load. This is VERY IMPORTANT to prevent reflections and incorrect readings. Put the gen to the other end port and the analyzer plugs into the "T" in the middle.

    Also it may take more "smoke" (level) to get the signal as far above the noise background as possible. I use a small portable radio as a "signal generator" with higher output capability than the service monitor to do the final steps where you need more than the 60 dBm dynamic range you have with the HP gen.

  14. #14
    PRC148 No Longer Registered

    Default

    That makes sense. I will try to inject some more signal with a 50 ohm load on the other port. The tuning instructions show it left unterminated but I will try that.

  15. #15
    PRC148 No Longer Registered

    Default

    I am now seeing -1.32 loss through both pass bands and rejection levels that bounce between -103 to -108. I think that's about as good as it will get. Thoughts?

  16. #16
    Alpha's Avatar
    Alpha is offline T S - Moderator
    CS Forums $upporter
    Join Date
    Feb 12, 2012
    Location
    Directly above the center of the earth.
    Posts
    2,780
    Thanks
    1,235
    Thanked 1,531 Times in 738 Posts
    Country: Christmas Island

    Default

    That sounds about right. Typical insertion loss is 1.2-1.5 dB and typical reject anything over 100 is decent anything over 110 is great. The terminator trick is a good thing to know, I was taught that by a retired Motorola RF engineer.

  17. #17
    Join Date
    May 23, 2012
    Posts
    785
    Thanks
    442
    Thanked 878 Times in 363 Posts
    Country: United States

    Default

    Agreed on terminating the unconnected port when tuning duplexers. That was one of the first things I was taught when I began doing this. If you don't, you'll see exactly what Alpha mentioned - nonsense readings, reflections, and things that will just aggravate you.

    And I also agree - give it more level. You're nearly in the noise floor with your monitor. I like to begin retuning things with 0dBm into the cavities, then add attenuation as needed and back the levels off from there. Obviously you need to run wide-open without attenuation to see how deep your notches are, less so with added attenuation to see the passband.

    I just did a DB Products 4-cabity BpBr duplexer today. Every time I touched the notch tuning screws with the screwdriver, the readings went way-off. Lots of "insert, adjust, remove" type action with that thing. Wasn't too big a deal once the first cavity was done. Didn't have any ceramic or plastic tuning tools that would fit in the deep well the tuning screws were in.

    A service monitor with the "Tracking Generator" option is a beautiful thing. It makes life considerably easier when tuning filters. If your monitor has that capability, use it. If you don't know how, learn.

  18. #18
    Alpha's Avatar
    Alpha is offline T S - Moderator
    CS Forums $upporter
    Join Date
    Feb 12, 2012
    Location
    Directly above the center of the earth.
    Posts
    2,780
    Thanks
    1,235
    Thanked 1,531 Times in 738 Posts
    Country: Christmas Island

    Default

    Another VERY valuable trick he also taught me was to MEASURE THE DAMN TERMINATOR! EVERY TIME. Make sure it's 50-52 ohms, and that someone hasn't transmitted into it with high power and cooked it up to a higher resistance!! I have seen blown ones that read 70-80 ohms or higher, and needless to say they won't work right. An off-value terminator is nearly as bad as a missing one. Air is about 377 ohms, so an unterminated connection is that or higher impedance.

  19. #19
    PRC148 No Longer Registered

    Default

    It has the tracking generator up to 1 Ghz. I upped the input and used a 50 ohm dummy on the open port. It did settle things down. It appeared to provide more normalized results. I found the notches on the transmit side were way off which was likely leaking RF causing the desense. I moved all the cables around and tried putting pressure on the T connections to see if there was anything intermittent and it all seems rock solid. I'm hoping I can get it as close as possible then do the fine tuning at the site.

    What do you guys think about possibly using the actual site antenna as the dummy load on the open ports when doing the final tuning? Would that help by adjusting to the actual antenna system that it will be placed on or would it be of any use to try that?

  20. #20
    Alpha's Avatar
    Alpha is offline T S - Moderator
    CS Forums $upporter
    Join Date
    Feb 12, 2012
    Location
    Directly above the center of the earth.
    Posts
    2,780
    Thanks
    1,235
    Thanked 1,531 Times in 738 Posts
    Country: Christmas Island

    Default

    First, generally speaking you should not ever do adjustments to your system connected to an antenna.

    Second, you should not try tuning the duplexers under high power. It can cause internal arcing and other issues, and could endanger the TX or RX if they get too far out of whack.

    Third, the general philosophy is to try to tune to a proper 50 ohm resistive load, and if your antenna is NOT exhibiting those characteristics, fix it. Or consider using a circulator to help ensure a better match and to dump any incoming signal or power away from the TX output stages.

  21. #21
    PRC148 No Longer Registered

    Default

    OK, that makes sense. Thanks