Hello Peter,

I have changed the rods in a Q2330E. I used Invar rods. Invar is used for the temperature stability. Not sure about stainless. I'm going to say probably not. I was lucky enough after a long time of searching to find a machinist who had some Invar rod. He put each rod in his lathe and cut the threads. He also smoothed the top end, including a very slight taper. The plungers are thread locked so it takes some work. If I remember correctly, I used acetone to ease the thread lock. I believe that I made the rods 24 inches. You could make them 26 inches to be on the safe side.

Here are some pictures that may help give you an idea of what you are getting into. You will notice the threaded end of the rod is at the bottom of the plunger.



Mick - W7CAT

开云体育

On 2/11/2025 8:45 AM, Peter Campbell via groups.io wrote:

I have 2 duplexers that the rods have been cut and I'm unable to tune them down to the ham bands and I'd like to replace the rods.? I've contacted the manufacturer and the cost of the original rods is very high.? Has anyone been able to do the rods themselves with standard stainless or invar rods?
Thanks in advance!
Peter
VE9PTR

Consider having a machinist extend the rods you have now.? Even if the metal is somewhat dissimilar - as long as it's mechanically sound and compatible for welding / brazing, the difference over length will be negligible.

Kevin W3KKC

"Try to get fine thread rods, if possible."

I don't understand that. The rods should be smooth. And fine thread won't work in the plungers. As I remember they are course thread.

Mick - W7CAT

Kevin n Co

can I assume you will be actually inserting this SS long nut down near the plunger rather than at the top of the rod.
You will require top grade stainless steel, with no ferrous what-so-ever.

Undebar !!!

BUT the coefficient of thermal expansion of Invar is out of this world Mike

I would put your extension rod near the bottom and into the plunger. A couple of things with doing that, your added threaded rod will work with the long nut, and on the bottom of your original rod, it is already threaded so it will screw into the long nut. Also, I would want the smooth rod going through the rod locking part of the duplexer at the top. Don't forget to use a thread lock (Lock-tite) kind of stuff on the rods going into the long nut and the plunger.

Mick - W7CAT

Right on Mike !
you have it first put,
and it's too wet down here,
my Light Sabre would rust

LOL

开云体育

Here you go!

Jack K6YC

I know they have it. The problem is the lengths available. 1 foot is too short and 3 feet would be a lot of waste. Also for a four cavity duplexer it would cost $200 for the rods. The other option which would be a lot less expensive would be to get the 1 foot length and use the long nut on each one.

Mick - W7CAT

开云体育

开云体育

Thanks for all the advice.? I was able to extend the rods using stainless with mixed results:

?

-on "duplexer A" I can only get about 30db of isolation per cavity, which from what I'm reading is not as good as it could be.? Any suggestions?

-on "duplexer B" I can only get 20db on the high side, and on the low side I run out of tuning on notch before it reaches the low point. It bottoms out.? Again, any suggestions? I read somewhere about rotating the capacitor but this is all new to me so I haven't tried that yet. Are they frequency specific (ie a range)?

?

Both duplexers are the same model, but there were differences in the bottom ends of the rods so they're not identical internally.?

?

Thanks again for all the info!

?

Peter

VE9PTR

<-- Hi Peter; I'm? no expert - but, out of respect to my "tutors" Kevin (W3KKC) and Andrew (VE3GTU), I'll have a go and sharing some recent learning / experience.

On each cavity, the tuning loops are secured with 3 Robertson screws.? While using a T connector on each cavity, rotate the tuning loop until you see about 1dB of loss in the pass band - example:

On the TX side, I had to experiment with the IL, I started with 0.7dB, but with help from Kevin, I had to increase this to ~1dB as well in order to improve my reject notch, since we were troubleshooting a desense issue.

Once you have ~ 1dB of insertion loss, tighten the 3 screws; next adjust the capacitor to get your reject notch.? As you can see from the above example, I was able to get ~35dB of rejection on each cavity.

Do that on all 4 cavities.

Once you reconnect the harness, your final plots should look something like this:

Hope these plots give you some idea of what to aim for.

Cheers,

Ramesh, VA3UV, WA3UV

开云体育

I use a similar method that I find a little more accurate for setting the loops. It's very hard to set the pass for 1db IL accurately. I use this method to build VHF duplexer a from anywhere from 4 to 8 cavities.

?

Using the capacitor, center the two notches around the pass. You will have to increase the span to see it. I start at 20 Mhz. Adjust the rod so the pass frequency is at the lowest insertion loss between the notches.

?

Now loosen the three screws holding the loop in place. Rotate the loop until the two notches are the following frequency span across:

?

Notch span.? Ins Loss.? Notch @600kHz

9 MHz.? ? ? ? ? ? ? 0.3 dB.? ? ? ? ? ? 25 dB

8 MHz.? ? ? ? ? ? ? 0.4 dB.? ? ? ? ? ? 27.5 dB

7 MHz.? ? ? ? ? ? ? 0.5 dB.? ? ? ? ? ? 30 dB

6 MHz.? ? ? ? ? ? ? 0.7 dB.? ? ? ? ? ? 32 dB

5 MHz.? ? ? ? ? ? ? 0.85 dB.? ? ? ? ? 35 dB

4 MHz.? ? ? ? ? ? ? 1.2 dB.? ? ? ? ? ? 37 dB

3 MHz.? ? ? ? ? ? ? 1.8 dB.? ? ? ? ? ? 39 dB

?

The insertion loss and notch depth are additive, so you can design your duplexer in this fashion.

?

One thing I also do is rebuild the harness. I believe Sinclair had two harness lengths for these at VHF, one from 137-150 and another from 148-174. If you measure these harnesses, they are actually centered around 138 and 160 MHz respectively. I tear apart the harnesses and rebuild them with approx 14 inch lengths of RG400. I'll actually build the harness with lengths very specific to the frequency used. The intercavity lengths need to be an electrical 1/4 wavelength. This needs to take into account the velocity factor of the coax in use.

?

<-- Hey Dave... long time!

Interesting info.. I agree getting the IL at exactly 1dB is a bit of a game, trial and error; you think you've got it, but by the time you tighten the screws up (and the loops move), the IL changes.... here's a more recent example:

An IL of -1.03dB and a reject notch of -34.91dB.? Seems to be close to your table above.

and the TX side, an IL of -0.89dB and a reject notch of -34.5dB (again pretty close to your table).

So a question for you... with an IL of 1.8dB and reject notches of -39dB per cavity, are you seeing an overall (combined) reject notch of -88dB?

My combined / final plots:

(before anyone mentions the reject notch looking slightly off, the capacitor on one of the cavities seems to be a little touchy... probably could do with being replaced).

Cheers,

Ramesh, VA3UV, WA3UV