Vložit
- čas přidán 2. 11. 2017
- This video shows how to tune a notch-type duplexer using a spectrum analyzer with a tracking generator, as well as with a vector network analyzer (VNA). It also discusses the basics of what a duplexer is, and a quick review of the most popular types. Notes from the video can be found here:
www.qsl.net/w2aew/youtube/Dupl... - Věda a technologie
Wow, I learned quite a bit from watching this video. The follow-along visual representation was super helpful in understanding what's going on inside a duplexer. I'm grateful to knowledgeable amateurs like yourself for posting videos on complex topics for other HAMs to learn from.
Great video! I'm sure lots of folks will benefit with you showing us how to tune that filter!
Substantial difference between the two methods. Have never used or seen it done with a VNA. Thanks for the video.
Thanks for the video. Just got a VNA last week. This helps a lot.
EXCELLENT explanation sir! Thank you. I am assembling a GMRS repeater ,and you answered some questions I had.
Rick
Always enjoy your videos! Years ago a Spectrum Analyzer was unobtanium for me at least because of the cost. I guess my bank account just doesn't keep pace with the state of the art. I'd love a VNA, but they are too new to be affordable. I guess I'll live with -90 db. It's still nice to see why the new gear costs big bucks!
Thanks for this video, I had hoped to see you prepare both your spectrum and vn anaylyzers to make these measurement (i.e. normalize and set a reference) and also view the traces with different spans. I know a portion of your audience may find this tedious but I'm sure I'm not alone here in my desire to see you explain these processes and narrate the changes to the measurements on the screen. I'm looking forward to your next video, keep up the great work!
I'll be sure to do that on future videos.
greetings from Colombia!, thanks for the video, well explained!
I wish I had somebody like you who would make me their apprentice. Nice video. 73
Big thumbs up Alan, Nice explanation. Looks like a VNA is my next bit of gear I need to get. As always thanks for sharing your knowledge.
The Radio Shop Check prices first :)
Nice video. Thanks Alan.
Just rolled it from 999 to 1K likes. Congrats! Thank you for all these awesome vids!
Very good video. You won another entry!
Wow, excellent. My cheaper VNAs probably do not have that noise floor but I will check. Only have one cavity now, and old, tight, GE blue can.
Good video, and informative. Keep up the good work.
I just discovered that they have a 6GHz VNA at my new job, which I now pet and talk sweet words to every morning.
It has one feature that I thought was really cool: just like modern scopes have a fft function that allows you to see a frequency spectrum, that VNA has a inverse fft function that converts your measurement to the time domain. This allows you to see where reflections happen, eg. To me that was mind blowing but it makes perfect sense when you think about it. It has phase and frequency information, all it needs for a transform!
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.
I was looking for information because fiplex does not present tuning instructions. It was helpful. Thanks.
Nice Video Alan very well explain and having the visuals makes it even better . Thats why I enjoying watching the videos and learning . wish Santa drop off one of those Spectrum Anaylers thats what need to complete the Test bench. 73's Alan
Wonderful video, thank you sir.
Thanks a lot , very nice
hola , muy buen video sobre caqlibracion de duplexer, con distintos instrumentos, yo los calibraba en su epoca con un generador de barrido y osciloscopio, saludos
Great video!
Hi Alan,
Great video. Good job with the visuals. 73 WB3BJU
Fascinating and informative. Thanks very much. Time to blow the dust off my nanovna!
You may want to check out my latest video regarding the NanoVNA and duplexer adjustment - may or may not have enough dynamic range...
czcams.com/video/GipCVEsiqXc/video.html
terima kasih ada kesempatan akan kami coba
Thank you so much.
Hi Alan, very nice and didactic video, many thanks!. I have an old Spectrum Analyzer (without Tracking Generator) and a Vector-Network Analyzer (the VNWA from DG8SAQ), and it would be great to see an example or learn from you how to combine them both so that the VNA acts as TG for the Spectrum Analyzer... I have found this topic also in several forums, but never found a proper tutorial or explanation on how-to. Again, many thanks for your videos! I am learning a lot.
Thank you that helped
I would love to see you do a brbp. No one has ever taught me how to do it. They've always done it for me. I've asked but it was too inconvenient to them and so they just did it.
Very good thanks
Nice video. Thumbs up...PEACE
Good video
w2aew I'm glad Alan that you kept referring to it correctly as "duplexer" rather than diplexer which is common.
Thanks for a succinct and informative video. I would like a list of equipment for reference. I know these videos take time to make as does annotation. Very nice work!
The spectrum analyzer was an Advantest U3641, and the VNA was the Tektronix TTR-506A.
i hope you are seeing an S21 plot on the network analyzer and these kind of dips do occur in the S11 input match plots
Hey, Alan!
I'm a recent subscriber and I've watched probably more than a half of your videos already-I really enjoy them! You've inspired me to look into HAM radio as a hobby, so another thanks is in order!
I know that Google is my (dubious)friend, however, I thought that these would make for very interesting video topics and I'd bet that I'm not the only one who'd like to get your personal perspective on them(did I mention that we love your teaching style?):
1). HAM for beginners: what basic equipment and fundamental knowledge is required?
2). Antennas: what types are there and is there any benefit of whacking one in on my roof? Say, for using with a cheap, Chinese, beginner radio?
3). Repeaters: principles of operation. Is it possible to build a homebrew one, at least from the standpoint of parts and components? I know it's most probably too technically complicated for a beginner, but would, say, a HAM adept be able to build one or are we stuck with off the shelf solutions?
Just wanted to throw these out there and please keep making videos!
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.
Hi Alan! Wonderful video ! Should a pad be used on the outside of the tracking Gen??
I would like to see a video on tuning Br/Bp with both a signal generator and a VNA.
Thumbs up and as always, thank you.
fabulous
Hopefully will be able to do this with my cheaper network analyzer. N6GRG
Thanks for sharing! 73! KD2FPM
You can always inject a high level signal from a sig gen at the notch freq and tune for max rejection. Most sig gens can do about +10 -13 dBm.
The tinysa has a built in signal generator would that work with my spectrum analyzer???
I connect HIGH and LOW with VNA and tune the HIGH part for max attenuation. I have to make sure that transmitter signal(from LOW) has max attenuation, coming to the receiver(connector HIGH). When I tune transmitter I connect ANT and LOW and tune it for min attenuation. Because power of tranmitting signal is more inportant than isolation between ANT and LOW on receiving frequency. Doesn't matter if receiving signal -30 and lower dBm comes to the transmitter. So when I did my last duplexer I tuned attenuation between LOW and ANT just 10 dB, but I got the max output power. What do you think?
Thanks for the tuning video. Are you going to show how it functions in a real-world application ?
Hello Alan, Thanks for the video. It's great. Have you ever opened the box of this duplexer's resonance cavity or designed by yourself ? I wonder what it is inside those resonance cavity or how they are lined up in ther, is there any video ?
Not this particular one, but I have opened other adjustable cavities - they resemble a piston in a cylinder.
Good job Alan and thanks for the demo using both pieces of equipment. Question: Do you have opinions on which modern VNA module you prefer? I like the module/laptop VNA set-up like you were using. I think I saw that yours in the video was from Tek. Thanks and 73, Joe W4JF
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.
Hi, Alan does the tracking generator track the same frequency range as the spectrum analyser covers band wise example 9Khz to 3.2Ghz for example. The duplexer tuning was interesting as the depth of the notch. The notch in the duplexer when you tuned it did it also tune the low side as to produce the required split between the transmit and receive side of the filter. I know the reason for duplexers reason to stop RF feedback and receiver protection from RF from the transmitter getting back to the receiver. How do they stop the RF getting into a receiver when mounted on racks because normally both are mounted in close proximity.?
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.
I've tuned mine at the most i can get is -65 db on the notches for high and low. I have the rigol dsa815.
now another question? there is 6 screws and I am thinking that 3 screws for high and 3 screws for low.This is my first time doing this and I want to learn
Did you use any pads on the tracking Gen out put? Did you use an RF bridge or straight in ? Tnx 73 de K8KEM
Got a question - I live near several AM broadcast station. I have a brute force filter that notches out the entire broadcast band. I’ve been wanting to build a tunable notch. Looking for some tips to narrow down the notch. Most simple LC networks cover 50-75% of that band at once. Looking forward to your videos! WA2SQQ ELMWOOD PARK NJ
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
The real impedance of the transmitter, receiver and antenna will be different enough to merrite a final tweak at the repeater site.
I might try this. I have access to a spectrum analyzer and a Sig Gen. Basically, the idea is to feed the Repeater output TX frequency into the port the receiver connects to and notch out the TX freq so
the receiver doesn't see it. Then on the port the Transmitter connects to feed the Repeater Receiver frequency and notch out the Repeater TX frequency. Do I have that right? A friend just picked up a GMRS repeater and needs the duplexer tuned to his Repeater Frequency. I should be able to do this with a spectrum analyzer and a sig generator.
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.
Having a very deep notch is not always best as the Duplexer tuning point changes with temperature. Having a slight flat bottom is better to allow for temperature drift. But nice video - info on your VNA would be welcome. 73' G4SDR.
That's a very good point, thank you!
The VNA is a Tektronix TTR506A.
What would happen if you dont terminate TX connector in case you measure Rx line and Ant connector? Isnt the isolation between Rx and Tx connectors good enough?
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.
Could you recommend a duplexer for a GMRS repeater in the 462-467 range? I have a uhf repeater running on 444.400 with the 4 cans, I don't know how to tell if this duplexer will adjust out to the 462-467 range??
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..
Thanks for the Info. one question: When you say terminate, what do you mean? Cap, short or 50ohm load? Thanks
50 ohm load'
@@w2aew Thank you
Great video! Do you have a shop?
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!).
I have the tinysa which has built in signal generator now can I use that with my spectrum analyzer to tune the duplexer???
Probably ok to help set the notch, but you'd really want to look at return loss to set the passband.
Hi..
Quick Question... Where did you obtain the test cables you used in this video?
(Coax with “strain relief” / “N” male connectors)
Thanks😖!
And..thanks for your videos!
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 !
When you would need to tune a duplexer? Or what my question is what would cause you to tune it in the first place? Would it be because the loss of signal integrity or cross talk between the TX RX signal to the antenna. Great video Cheers
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
Any ideas on converting one of these mobile duplexers into 23cm bandpass notch/reject service? It seems they should work on their third harmonic with reduced coupling. Full-on notch-reject would be required for any serious repeater use, but maybe just not the Q required for low loss use. Keen on your thoughts, and thanks for teaching! :)
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.
What type of cable should I use when connecting the duplexer to the analyzer?
Ideally should be high-quality double-shielded (>95% shield).
Is there an inherent property of a VNA that gives you better dynamic range than the spectrum analyzer?
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.
Does a VNA inherently have more dynamic range than a spectrum analyzer or is it just a better VNA?
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.
Would the NanoVNA be sufficient to do this kind of tuning?
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.
I used a Rohde & Schwarz FSEA30 spectrum analyzer with tracking generator option to tune MANY duplexers. With a zero span noise floor of -160 dBM, I had plenty of dynamic range to tune the duplexers all the way to the bottom of the notch. In spans and bandwidths that were suitable for duplexer tuning, the bottom of the notch was still well above the noise floor.
Before I got the R&S analyzer, I used a Tektronix 494AP spectrum analyzer with a Tektronix TR503 tracking generator and it too was quite adequate for duplexer tuning. Not as good as the R&S, but certainly good enough.
When Chinese duplexers started to appear, the difference in quality was inmmediately apparent. You DO get what you pay for. I would not recommend a Chinese duplexer to ANYBODY for ANY reason. What absolute GARBAGE.
So can a relay or multiple relays substitute this complex duplexer ?
Relays won't permit simultaneous transmit and receive, which is required for a repeater application.
Couldn't you use an amplifier before the D.U.T. to effectively move the noise floor down? Coupled with a limiter to protect the signal analyzer while the sweep is in the passband, of course.
There are several techniques like this that can be used to drop the noise floor in a spectrum analyzer.
next question from a noob. the sticker has high of 467.000 and a low of 462.000 ..now my repeater using 462.675 and 467.675..i have the nano vna that i will be using. will i be able to notch for my repeater or do i go by the sticker???i have not tried to notch yet
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.
Have you done a video that explains S parameters?
This one was great but your comment at the end revealed a gap in my understanding :-(
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!
Hi Alan About 5 years Brian Swann N1BS did a video show how the use of a Return Loss bridge improves the resolution of a Spectrum Analyzer when tuning a duplexer. Your Spectrum Analyzer has very good Dynamic Range but you had to resort reduce RBW in order to get increased Dynamic Range. For those with a Spectrum Analyzer with modest specs a Return Loss Bridge can be acquired for considerably less cost than a new VNA. Here is the link to Brian's video : czcams.com/video/ghJHPX9f5aE/video.html
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:
czcams.com/video/vVuMYCdlsZw/video.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.
What model of VNA did you use?
The VNA is a Tektronix TTR506A.
What vna are you using.
It is a Tektronix TTR-506A.
what VNA do you use?
It is a Tektronix TTR-506A.
How do the phone ones work?
I'm not sure what the phone on-board duplexers look like.
Are you injecting a frequency type Generator. To tune it ?
The Tracking Generator and the VNA generate the test signal.
Was hoping for something a bit more simplified using an instrument more commonly available than a vector analyzer. A scalar would have been something more likely and certainly less expensive. Even these are way out of most folks price reach let alone any clue how to use it and of course the directional bridge that must accompany either. A communication analyzer like an 8920/21 would be a more realistic for the Amateur Radio class or possibly the advance technician type. I believe the filter tuning can be accomplished using either instrument above if the proper settings are established with an equipment price an order of magnitude cheaper. Nice video however but if it’s necessary to explain what a duplexes is and how it works, you are so far over the heads of any audience that could begin to comprehend this discussion let alone how to perform the tuning.
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.
Somehow I missed passband attenuation ...
copy paste da li ima pomoci.
"Imam jedan prilicno glup problem, koji ne znam kako da izracunam. :ohno:
Napravio sam manje vise prirucni VNA. Ukratko sta je sklepano, signal generator(u ovom slucaju modul za koji radim antenu) na 1-3000MHz RF most. Njegov Ref port na kanal A od AD3802, OUT(ovde je reflektovan signal) na kanal B, DUT na antenu.
Sa AD3802 imam signal za fazu i odnos amplituda dva signala. Odradio sam bazdarenje sa otvorenim portom, kratkim spojem i sa 47 i 50R. I to je sve jasno, da moram na fazu dodati tu fazu sa bazdarenja, u sustini kompezacija do ravni koja se meri. Isto tako moram dodati na gubitke mosta, jer za kratki spoj i otvoren dobijam 10dB. Dalje mali problem je faza, jer na izlazu dobija +/- ugao. Ali moze da se odradi dodavanjem kalema ili kondezatora paralelno anteni, pa da se vidi kuda ugao mrda.
I to je sve ok, ali dalje nastaju problemi.
Kako pretvoriti ta dva podatka u nesto korisno, u prevodu kako ih pretvoriti u kompleksnu impedansu, ili kako ih upisati u smith chart.
Ako treba vise informacija i linkovi ka ebay, rado postavim.
Hvala na svakom odgovoru
"
www.elektronika.ba/forum/viewtopic.php?f=11&t=17512
Terminate is that with a 50 ohm dummy load
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
LOL I thought that was a retrofit of a FX control foot panel :D