I'll be sure to do that on future videos.

The Radio Shop Check prices first :)

Yes, very common in VNAs. The reverse is also available on high-end TDR equipment - a transform to go from TDR/TDT results to S-parameters.

The spectrum analyzer was an Advantest U3641, and the VNA was the Tektronix TTR-506A.

You may want to check out my latest video regarding the NanoVNA and duplexer adjustment - may or may not have enough dynamic range... ukposts.info/have/v-deo/f5mhd4x8qp-iuZc.html

Thanks for the video ideas - I'll add them to my list. In the meantime, you can check out some of the introductory information and videos on the ARRL.org website. It is possible to build a repeater, but most folks/clubs will buy them to ensure they behave properly.

Not this particular one, but I have opened other adjustable cavities - they resemble a piston in a cylinder.

I haven't used a lot of different VNAs, so I really can't form an unbiased opinion. The Tek unit I am using is very nice - performance similar to benchtop units costing north of $20k.

It likely can be done with an LC filter, but it will have to be composed of multiple stages. The higher the Q needed, the higher the "order" of the filter - which means more stages. You can use a program like Elsie to greatly simplify the effort: www.tonnesoftware.com/elsie.html

It depends on the spectrum analyzer. Some tracking generators will cover the full range of the analyzer, others will only cover a portion of it (which is usually the case when the analyzer covers in excess of 10-20GHz). For bandstop or notch duplexers like this, there is a high side and low side. One notches the tx frequency out of the receive path and the other notches the rx frequency out of the tx path. With this type of duplexer, there is no passband adjustment, just the notch adjustments. Using good double shielded coax cables and properly mounted and grounded (and enclosed) housings for the radios, crosstalk from tx to rx is generally not a problem.

Duplexers need to be tuned to the specific frequencies they'll be used for. Once properly tuned, you shouldn't have to touch them again unless the operating frequencies change.

Thanks Alan

Yes - also be sure to terminate the opposite port, and make sure the antenna has a very good match (low SWR, high return loss).

@@w2aew yes, will do. I have tuned duplexers for the 220MHz band when I worked and had access to the equipment. Once you know the procedure it's not that hard.

My limited testing showed me that operation on the 3rd harmonic was possible with reduced performance (about 15-20dB shallower notch, a touchier adjustment, and not as selective (reject BW was wider))

@@w2aewit weas suggested to me that they would be overcoupled there. I'm looking for some 23cm repeater duplexer option that is better than starting from new copper.

@@SteveWrightNZ I'd have to take a look to see, given the broader notch, whether there will be sufficient isolation with the typical split at 23cm.

That's a very good point, thank you!

The VNA is a Tektronix TTR506A.

50 ohm load'

@@w2aew Thank you

They were supplied by Tektronix with the TTR506A VNA.

Thanks! Sadly, I’m guessing that’ll be $800 per cable..? 🤣 They do make fantastic quality stuff,,, just a bit pricey. (then again... you DO get what you pay for) 👍🏻

If you make the cables yourself, use double screened cable. Keep your workbench clutter free and separate the Rx/Tx cables. The interaction can be amazing !

@@s.pedersen1757 I thought I’d come back and let the world know of the BEST jumpers I’ve ever used. RF Industries makes the “UniDapt” multi connector adapter sets. It includes 2x male and 2x female of every type of the following connectors - “N” / SMA / BNC / Mini-UHF / PL-259 / MCX / TNC / “F” / SMA-RP / and a few others I’m forgetting. The kit can be used with the included “couplers” and regular cables OR -> The BEST (IMHO) CABLES. The RF Industries RG-400 ultra flexible cables (3’ or 5’ or 6’) with a female “coupler” type connector on each end. Color BLUE with excellent quality silicone type strain relief boots. They are great AND extremely convenient. I even bought the “UniDapt” flange mount connectors for my Telewave Watt Meter and Bird 43 meter. I also bought their SILVER PLATED connector set to use with the cables. I am not a big fan of the chrome plated type connectors. Just a warning.. Amazon sells the knockoff sets and a friend bought one after seeing mine. He was very disappointed when some of the connectors came with visible corrosion. I bought mine from a legit distributor. I just can’t recall the name. Thank you again for your fantastic videos !

Probably ok to help set the notch, but you'd really want to look at return loss to set the passband.

For some duplexers, it doesn't matter. However, I have seen the more complex multi-cavity bandpass/reject duplexers be affected by the termination on the other port. Just good practice to always test with the unused ports properly terminated.

I am planning to do videos on things like S-parameters, Smith Charts, and more RF topics.

w2aew It would be great to see video's about those subjects Alan! Thanks for this and all your video's!

Usually it lies in a fully characterized directional coupler (at each end), along with a complete vector calibration to remove systematic, tracking and directivity errors and tracking.

You could check with the manufacturer to see if the duplexer you have will tune that far. I would think that most would be adjustable to that range. Seems like a common spec for the 4-cavity band pass/reject duplexers is that they cover 400-470MHz.

@@w2aew Thanks for the quick reply on an older post..

Generally yes, a VNA will usually have more dynamic range for such measurements than a spectrum analyzer + VNA. The other big advantage is the vector calibration of the VNA.

No. In my day job I am a Field Applications Engineer for Tektronix. The videos I produce for my channel are just part of my hobby (and certainly enhanced by the equipment available to me!).

Ideally should be high-quality double-shielded (>95% shield).

Relays won't permit simultaneous transmit and receive, which is required for a repeater application.

With only 60-70dB of dynamic range, you can get close, but you may not be able to get optimal results for duplexers with very good rejection capability.

The tinysa has a built in signal generator would that work with my spectrum analyzer???

The stickers are likely the frequencies that the unit was tuned to when purchased. You would have to re-tune it for your actual operating frequencies.

There are several techniques like this that can be used to drop the noise floor in a spectrum analyzer.

The VNA is a Tektronix TTR506A.

I'm not sure what the phone on-board duplexers look like.

The Tracking Generator and the VNA generate the test signal.

It is a Tektronix TTR-506A.

It is a Tektronix TTR-506A.

Hi Bob. Thanks for the link to Brian's excellent video. I completely agree - using return loss is a better way to tune *Bandpass* cavities. In fact, I did a video on this too, except that I used a directional coupler instead of a return loss bridge. The result is the same, a much more precise measurement of the *bandpass* frequency. Here's a link to my video: ukposts.info/have/v-deo/roamgY96m6Kku6s.html This doesn't really apply in the case of this video (#270), because here I am tuning *band reject* or "notch" cavities. This this case, the return loss is fairly broad, but the S21 (through loss) is quite sharp.

The same measurement can be made with a scalar network analyzer - results would be very similar to the spectrum analyzer with tracking generator (but with better dynamic range in most cases). I would imagine that an 8920 could be used, but I have never used one (nor do I have access to one), so I can't comment on it.

Yes.

One more question .the spectrum analyzer has to be center frequency with 10mhz span what other settings should I need to change before activating tracking generator