Harold,

Thank you for replying to the thread.? For those following along - Harold and I are discussing the Sinclair Q202G and TX-RX 28-37-02A VHF duplexers, and that's different from the subject line because I hijacked the thread.

Comprod was formed by several employees that left TX-RX.? Earlier Comprod cavities were square and more so resembled EMR products.? Later ones look nearly identical to TX-RX.? Earlier TX-RX VHF duplexers used finger stock.? Later TX-RX VHF cavities use a different method where the larger outer conductor is split and reformed, resulting in hardened copper fingers that make the larger electro-mechanical contact area.? Over time, a process causes a "pressure weld" to occur - where the two conductors become one.? As witnessed by myself and others in this group, the latter design is actually superior.? It is, because there is more contact area as compared to the tiny surface mated by the finger stock.? Traditional finger stock provides much more wear on the plated surface of the movable center conductor.? Attached are the highlights of the later TX-RX center conductor manufacturing process and complete explanation of the pressure weld process.? I can't tell you how many cavities I've had to refurbish because the traditional finger stock ate the plating off with movement.? As such, I don't share the same opinion that a duplexer that uses traditional finger stock is superior to the TX-RX large finger design.

Fine tuning the resonator with the rod loads the top (cold) side of the cavity.? Lowering the resonance always lowers the Q, no matter if it's done on the top or bottom of the cavity.? While I agree that the rod lowers the Q, it doesn't do it to a degree that makes a big difference - if done correctly.? The correct way is to minimize the length of the rod, and only use enough rod to get (up) to the resonant frequency doing most of the work with the coarse adjustment.? Significantly reduced Q would show up as the inability to create an anti-resonant notch at a given insertion loss and cavity size.? I will show with VNA produced graphs (attached) that all things being identical (same cavity size, same number of cavities, and same amount of insertion loss dialed in, and same transmit to receive frequency spacing) the TX-RX VHF duplexer exhibits greater notch depth, more symmetrical notch depth, and greater return loss as compared to a Sinclair Q202G.? Some of that is due to the design of the TX-RX BpBr loops, but obviously the Q has to be there in the first place to achieve similar results.

I did a good bit of research on the more complex - double connector - VARI-NOTCH - BpBr coupling loops that TX-RX uses in their VHF cavities, and I created an detailing the results with help from several friends.? There is no need to duplicate the results here, but I'll summarize that greater return loss and greater notch depth is afforded by the more complex (in your words - flaky) coupling loop.? Jeff used 0.6dB of insertion loss for the Agilent E5070B VNA measurements of the TX-RX factory loop shown in the RB article.? Use the article's factory high-pass graph? (the third one down) in comparison to the attached Sinclair Q202G graph tuned to pass 145.250 MHz and reject 144.650 MHz with the same 0.6dB insertion loss.? It was also measured on Jeff's Agilent E5070B VNA.? After thoroughly comparing the two graphs, there is no evidence that Q has been significantly reduced by the fine tuning adjustment, or anything else mechanically for that matter.

You seem to claim that the simple series BpBr loop / capacitor arrangement used by Sinclair is symmetrical.? The truth is, when tuned to opposite sides, they are very asymmetrical.? The loop's design is the same - no matter if the loop is used for high-pass or low pass.? The capacitor is of sufficient value (30pF) to create an anti-resonant notch on either side of the pass frequency.? At 600 kHz, you always see less notch depth on the low-pass side.? That's why TX-RX added inductance to the low-pass Vari-Notch VHF loop.? That inductance is oriented to minimize the coupling to the magnetic field.? Attached is a document for a VHF duplexer tuned to 147.270 MHz where the insertion loss between "sides" is identical, but different notch depths are achieved because of the asymmetry.? This asymmetry is typical of this simple loop design.

So - thank you for sharing your opinion.? Obviously, I don't agree with you, and that's okay...

Kevin W3KKC

On 3/1/2025 10:15 AM, Harold Farrenkopf via groups.io wrote: