Excellent Ralph ! This is the best demonstration of duplexer tuning I have ever seen. Your knowledge of both the theoretical and practical points of electronics are why your videos are so valuable.
This is a great refresher! Thank you. I haven’t tuned a duplexer in a decade and had forgotten about the reject! I guess that’s what I get for just tuning up pass cavity only combiners. Oof.
He needs either more power out of the signal source or a more sensitive receiver to tune the notch. That is why there is not a smooth curve at the notch frequency and it is showing what I may call 'grass' here. Similar to the noise of a FM receiver when a signal fades below the sensitivity of it.
@@ralphmowery2898 I was using a professional, high end VNA to do this tuning as you can see. Yes, it was an older Hewlett Packard VNA which was recently calibrated and validated at the time of this shooting. For the sake of the requirements (looking at the specifications of the duplexer), this instrument has more than enough dynamic range to tune a duplexer to meet and exceed their specifications. 🙂
@@eie_for_you Then you either did not have the generator set for enough power out or the receiver part was not set for enough sensitivity. I have often seen this when I am using HP 8924c and do not set the sa/tg to the correct range for what I am doing. If the unit is set up correctly and has enough dynamic range you will get a sharp curve at the notch instead of that broad one.
Well, mobile duplexers are a bit of a different story. They only have a single adjustment: the only thing that is tunable here is the operation frequency as revealed by the reject frequency. The offset between the pass and reject frequencies is pretty much set in stone without a LOT of unpleasant disassembly (in my experience). So, the setup is pretty much the same. VNA Port 1 to the receive or transmit port and port 2 to the common/antenna port. Adjust the cavities for coincidence of reject frequency (maximum reject) at the desired reject frequency. Check the return loss looking into the rx and tx ports as well as the insertion loss at the pass frequency for each side. This will give you an idea of the health of the duplexer. Hope this helps. Oh ... and by the way, I'd have to borrow one such duplexer to produce the video.
You may find that the nanVNA doesn't have the dynamic range to adjust the reject frequency real well. They often only have a dynamic range of around 70 dB and duplexers want to run -80+ dB (sometimes as much as -100 dB). I'd be interesting in hearing the news when you are done.
I am making progress Ralph. I am still working with the Nano VNA. I thought I tuned it almost to perfection but in operation, sensitivity is very low on RX side. Its a TX RX 4 cavity duplexer model number 47-37-78005. Its quite old and had been kept unused for a long time. I am trying to tune it up for 145.6 negative shift 600 Khz. I have asked spectrum analyzer with TG from a friend of mine. Tonight, I will be making another attempt with the Nano VNA. The notch capacitors are a bit "jumpy" - probably they need to be replaced... I will update further in this forum! Again, thanks for super video!
@@rajeshkannan634Boy! I cannot find a datasheet for that duplexer anywhere! First question is, "What is it 'Minimum Separation' specification?" Hopefully you will be able to get it with the spectrum analyzer. The problem with the nanoVNA is its dynamic range just isn't quite high enough to do a truly good job on a duplexer. Then there is the frequency calibration of the nanoVNA. You might have the duplexer tuned right on as far as the nanoVNA is concerned, but its frequency calibration might have you tuning the duplexer to the wrong frequency. It doesn't take much of a variation to be a problem. Looking forward to your results with the SA. 🙂
Yesterday I typed a long reply to yours and posted it but somehow it doesnt show! OK,... I couldnt find spec sheet for this repeater either. I will write to TX RX systems and ask. I did a google search and found that they have many similar (looking) models and they are 0.5Mhz separation. This ones probably has similar. I am trying to get 600 separation - Repeater RX-145.000 Mhz and TX-145.600. I am still at it because the legend is that this duplexer was shipped to our city some 35 years ago with the same frequency and it was in operation. I got this far with the tuning - www.qsl.net/vu2ow/765f55fb-80eb-481a-a863-62a3040b15d7.jpg. With this tuning, sensitivity on RX side was low with 8-9W output from the repeater. At present, a notch capacitor is broken. I am waiting for that to be replaced.
Well, a quadplexer is the same concept as a duplexer only with one common port and four (vs 2 of the duplexer) "output" (really bidirectional) ports. I've never had the pleasure of tuning a quadplexer. Nonetheless, I am of the opinion that, if it is adjustable/tunable, the process would be virtually the same just with four ports to tune instead of two.
Forgot... a great way to “rehab” an old duplexer with crumbling harness cables & corroded (NOT oxidized silver... ) crappy nickel plated (nickel shouldn’t be used on duplexers amyway... intermod etc..).. is FleaBay! Honest. 😳 I have found a few US vendors selling good brand name “jumpers” at various lengths. Remember that length IS critical. Example. RG400 with Silver plated “N” male connectors on each end. 12” jumpers were only $60 for 12 pieces. Just a thought.
After some of the awful stuff I have received supposedly made by professionals (usually crimp-type), I prefer to build my own. Then I know it is right and can fix it if it isn't.
The cables that connect a duplexer to your radio can be of any convenient length. The key here is to use good quality, double shielded cables for this. Normal, everyday coax has enough signal leakage to degrade the overall performance of the system. The cables that interconnect the cavities and that join the two halves of the duplexer together, now *THAT* is quite a different story. These also need to be good quality, double-shielded cables. They are all an odd multiple of 1/4 wavelength long (electrically speaking) **including** the innards of the duplexer cavity; so, they are actually a bit shorter. This is NOT a fun project to recreate a missing set of duplexer cables. 🙂
I have a doubt: some videos from others show the S11 on the duplexer's antenna port and the S21 on either the TX or RX ports. On this configuration, you can see one of the two notch filters and the attenuation is around -90 dB on my nano VNA for a 6-cavity mobile duplexer. However, having S11 on TX and S21 on RX (or the reverse) like you show here, shows you both notch filters, the one from the TX side notching out the RX frequency and the one from the RX side notching out the TX frequency. But the reading on the VNA is -78 dB. Why is there such difference? Thank you.
With port 1 on the tx or rx port with the opposing port and antenna ports properly terminated in 50 ohms and measuring S11 (return loss) measures the match of the duplexer to your radio. With VNA port 1 connected to the tx or rx port, the opposing port properly terminated with a quality 50 ohm load and VNA port 2 connected to the antenna port and making an S21 measurement (thru or response) you are seeing only the rejection as seen by your radio. I learned a lot about this process when I was first starting out through a video by Sinclair, a duplexer manufacturer, on how to tune a duplexer. If you watch this video carefully, you can see that he has his Port 1 (source) connected to the rx/tx port and Port 2 (detector/rcvr) connected to the antenna/common port of the duplexer. Here is a Google link to that video: www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwjB37HUxbz_AhXUE1kFHW_DANsQwqsBegQIDxAF&url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DU7QZzzB2LZU&usg=AOvVaw1NjmvyYASdczPOUpmNLyxt Why the difference? I’m not sure of that answer. I hope this helps.
The Rigol series of test instruments sure gives you a good “bang for the buck”. The DSA815TG is a real steal at approx $1200. That’s what some of the “top tier” brands charge for cables and probes and accessories. Not that it’s gonna hold up to tower site work or heavy field use... but most of the time this stuff is tuned on the bench anyway. Fantastic videos ! I’m guessing your background was in the teaching field.?
Yes, I like my spectrum analyzer! Bang for the buck, a great piece of equipment. Actually ... I was an Electronics Engineer, product development for 20+ years and a minister for 13+ years. I still teach Adult Sunday School.
Excellent Ralph ! This is the best demonstration of duplexer tuning I have ever seen. Your knowledge of both the theoretical and practical points of electronics are why your videos are so valuable.
This is a great refresher! Thank you. I haven’t tuned a duplexer in a decade and had forgotten about the reject! I guess that’s what I get for just tuning up pass cavity only combiners. Oof.
I'm so glad that you found this video helpful! It's so easy to forget things when we are away from them for a while. 🙂
He needs either more power out of the signal source or a more sensitive receiver to tune the notch. That is why there is not a smooth curve at the notch frequency and it is showing what I may call 'grass' here. Similar to the noise of a FM receiver when a signal fades below the sensitivity of it.
@@ralphmowery2898 I was using a professional, high end VNA to do this tuning as you can see. Yes, it was an older Hewlett Packard VNA which was recently calibrated and validated at the time of this shooting. For the sake of the requirements (looking at the specifications of the duplexer), this instrument has more than enough dynamic range to tune a duplexer to meet and exceed their specifications. 🙂
@@eie_for_you Then you either did not have the generator set for enough power out or the receiver part was not set for enough sensitivity. I have often seen this when I am using HP 8924c and do not set the sa/tg to the correct range for what I am doing. If the unit is set up correctly and has enough dynamic range you will get a sharp curve at the notch instead of that broad one.
@@ralphmowery2898 Never thought of setting the stimulus level at a different level than the default, 0 dB. It is worth giving it a try. 🙂
Thanks Ralph, as always a great video
Great video thank you so much.
Thank you and you are very welcome! 🙂
I'd love to see the or a full video of tuning the mobile duplexer
Well, mobile duplexers are a bit of a different story. They only have a single adjustment: the only thing that is tunable here is the operation frequency as revealed by the reject frequency. The offset between the pass and reject frequencies is pretty much set in stone without a LOT of unpleasant disassembly (in my experience).
So, the setup is pretty much the same. VNA Port 1 to the receive or transmit port and port 2 to the common/antenna port. Adjust the cavities for coincidence of reject frequency (maximum reject) at the desired reject frequency.
Check the return loss looking into the rx and tx ports as well as the insertion loss at the pass frequency for each side. This will give you an idea of the health of the duplexer.
Hope this helps.
Oh ... and by the way, I'd have to borrow one such duplexer to produce the video.
Thank you so much Ralph! I am attempting to tune a VHF duplexer with NANO VNA. I will add more comments once I finish that! Rajesh VU2OW
You may find that the nanVNA doesn't have the dynamic range to adjust the reject frequency real well. They often only have a dynamic range of around 70 dB and duplexers want to run -80+ dB (sometimes as much as -100 dB). I'd be interesting in hearing the news when you are done.
I am making progress Ralph. I am still working with the Nano VNA. I thought I tuned it almost to perfection but in operation, sensitivity is very low on RX side. Its a TX RX 4 cavity duplexer model number 47-37-78005. Its quite old and had been kept unused for a long time. I am trying to tune it up for 145.6 negative shift 600 Khz. I have asked spectrum analyzer with TG from a friend of mine. Tonight, I will be making another attempt with the Nano VNA. The notch capacitors are a bit "jumpy" - probably they need to be replaced... I will update further in this forum! Again, thanks for super video!
@@rajeshkannan634Boy! I cannot find a datasheet for that duplexer anywhere! First question is, "What is it 'Minimum Separation' specification?"
Hopefully you will be able to get it with the spectrum analyzer. The problem with the nanoVNA is its dynamic range just isn't quite high enough to do a truly good job on a duplexer. Then there is the frequency calibration of the nanoVNA. You might have the duplexer tuned right on as far as the nanoVNA is concerned, but its frequency calibration might have you tuning the duplexer to the wrong frequency. It doesn't take much of a variation to be a problem.
Looking forward to your results with the SA. 🙂
The label on the repeater says its for 144-174Mhz. @@eie_for_you
Yesterday I typed a long reply to yours and posted it but somehow it doesnt show! OK,... I couldnt find spec sheet for this repeater either. I will write to TX RX systems and ask. I did a google search and found that they have many similar (looking) models and they are 0.5Mhz separation. This ones probably has similar. I am trying to get 600 separation - Repeater RX-145.000 Mhz and TX-145.600. I am still at it because the legend is that this duplexer was shipped to our city some 35 years ago with the same frequency and it was in operation. I got this far with the tuning - www.qsl.net/vu2ow/765f55fb-80eb-481a-a863-62a3040b15d7.jpg. With this tuning, sensitivity on RX side was low with 8-9W output from the repeater. At present, a notch capacitor is broken. I am waiting for that to be replaced.
Excellent!
Thank you! 😃
Thanks. Could you talk more about Quadplexer?
Well, a quadplexer is the same concept as a duplexer only with one common port and four (vs 2 of the duplexer) "output" (really bidirectional) ports. I've never had the pleasure of tuning a quadplexer. Nonetheless, I am of the opinion that, if it is adjustable/tunable, the process would be virtually the same just with four ports to tune instead of two.
Great video. Thanks sir.
You, my friend, are very welcome! 🙂
Forgot... a great way to “rehab” an old duplexer with crumbling harness cables & corroded (NOT oxidized silver... ) crappy nickel plated (nickel shouldn’t be used on duplexers amyway... intermod etc..)..
is FleaBay! Honest. 😳
I have found a few US vendors selling good brand name “jumpers” at various lengths.
Remember that length IS critical.
Example. RG400 with Silver plated “N” male connectors on each end. 12” jumpers were only $60 for 12 pieces.
Just a thought.
After some of the awful stuff I have received supposedly made by professionals (usually crimp-type), I prefer to build my own. Then I know it is right and can fix it if it isn't.
How do you determine the length of the patch cable going to the radio? Is it dependent to the frequency band you'll be using?
The cables that connect a duplexer to your radio can be of any convenient length. The key here is to use good quality, double shielded cables for this. Normal, everyday coax has enough signal leakage to degrade the overall performance of the system.
The cables that interconnect the cavities and that join the two halves of the duplexer together, now *THAT* is quite a different story. These also need to be good quality, double-shielded cables. They are all an odd multiple of 1/4 wavelength long (electrically speaking) **including** the innards of the duplexer cavity; so, they are actually a bit shorter. This is NOT a fun project to recreate a missing set of duplexer cables. 🙂
I have a doubt: some videos from others show the S11 on the duplexer's antenna port and the S21 on either the TX or RX ports. On this configuration, you can see one of the two notch filters and the attenuation is around -90 dB on my nano VNA for a 6-cavity mobile duplexer. However, having S11 on TX and S21 on RX (or the reverse) like you show here, shows you both notch filters, the one from the TX side notching out the RX frequency and the one from the RX side notching out the TX frequency. But the reading on the VNA is -78 dB. Why is there such difference? Thank you.
With port 1 on the tx or rx port with the opposing port and antenna ports properly terminated in 50 ohms and measuring S11 (return loss) measures the match of the duplexer to your radio.
With VNA port 1 connected to the tx or rx port, the opposing port properly terminated with a quality 50 ohm load and VNA port 2 connected to the antenna port and making an S21 measurement (thru or response) you are seeing only the rejection as seen by your radio.
I learned a lot about this process when I was first starting out through a video by Sinclair, a duplexer manufacturer, on how to tune a duplexer. If you watch this video carefully, you can see that he has his Port 1 (source) connected to the rx/tx port and Port 2 (detector/rcvr) connected to the antenna/common port of the duplexer.
Here is a Google link to that video:
www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwjB37HUxbz_AhXUE1kFHW_DANsQwqsBegQIDxAF&url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DU7QZzzB2LZU&usg=AOvVaw1NjmvyYASdczPOUpmNLyxt
Why the difference? I’m not sure of that answer.
I hope this helps.
The Rigol series of test instruments sure gives you a good “bang for the buck”. The DSA815TG is a real steal at approx $1200.
That’s what some of the “top tier” brands charge for cables and probes and accessories.
Not that it’s gonna hold up to tower site work or heavy field use... but most of the time this stuff is tuned on the bench anyway.
Fantastic videos !
I’m guessing your background was in the teaching field.?
Yes, I like my spectrum analyzer! Bang for the buck, a great piece of equipment.
Actually ... I was an Electronics Engineer, product development for 20+ years and a minister for 13+ years. I still teach Adult Sunday School.
:( we did not get to see the actual tune or results on the screen.
I thought I did tell everyone what the final results were. I guess I should have also shown it. Sorry about that.
A picture (video) is worth 10,000 words. Thanks for the best video with walkthrough and explanations.