Phelps Dodge duplexer.

[Colin927]

Hello all,

 

Hoping I can get a few questions answered here. I recently built my first repeater and just started off with a cheap Chinease notch style duplexer. When I got the repeater online and did a few range tests I noticed the transmit range was about 12 miles but the receive range was less than a mile. After talking to a few people in my local repeater group I was told to ditch the notch style duplexer for a more favorable cavity style duplexer. I got on Ebay and found this Phelps Dodge 4 cavity duplexer model 696 and had it tuned for my repeater frequency. When I received it I hooked it up and got some strange results. When an SWR meter is hooked up to it between the transmit side of the repeater and the duplexer I was now getting 40 watts out of the repeater at the back of the repeater instead of the 48 watts i was getting with the notch duplexer 1.13 swr at the back of the repeater box. When I moved the meter to the end of the jumper cable and screwed the meter into the duplexer side I got 36 watts still connected between the repeater and duplexer 1.28 swr at this point. I then moved the meter to the antenna output at the tee connector on the duplexer and got the same wattage as I was getting on the input side at the duplexer 36 watts except now I was getting 1.01 swr. I then thought it was tuned poorly so I connected my VNA to it and found it was slightly out of tune and did a slight adjustment following the original tuning instructions I found on the repeater builder website and the numbers didn't really change. I then started looking at the cables between the cavities and found that a couple of the connectors were loose and one was failing so i replaced the 1 end paying close attention to not change the length of the cable as apparently it changes the performance of the duplexer adversely, the cables from the end cavities to the tee were electrical taped at the ends for some reason so I removed the electrical tape and they seemed firm. Other things to mention that might be helpful are the repeater is a Kenwood TKR 850-1 and the Antenna is a Comet 712-EFC. The jumper cable used between the repeater and duplexer's transmit side is a 36" piece of lmr400 with N connectors on both ends, the receive cable is a piece of RG-142 with an N connector on 1 end and an N on the other. No adapters are used on the jumpers. On the antenna side I am using LMR400 with an adapter from so239 to N on the duplexer side, N connecter on the antenna side. " I know I know"  I need Heliax but I need to save up for the feedline, I should have it in place next month. The cables used between the cavities are pieces of RG-8U, on the Celwave version of this duplexer the cables are supposed to be RG-214 but I have seen some pictures of this duplexer on Ebay and they all seem to use RG-8U on the cavity cables. and according to the Celwave spec sheet for this duplexer the measured lengths of the cables are all correct. So my questions are: 1. are those reading normal from the SWR meter? I know there is usually loss through a duplexer but this seems weird. 2. Should I look into changing these cavity cables out due to poor fittings and electrical tape? 3. Are the jumper cables I have chosen ok for their purpose or should I change those out to something different. 4. Is the LMR400 to the antenna affection the duplexer performance? Oh and my range did change slightly, not nearly as much as I had hoped.

 

Thanks in advance

Colin.

[WROZ250]

I would first test the duplexer with the output terminated in a known good 50 ohm load.  I would also check to ensure each and every connection is tight.  One thing about duplexers is that even a slightly tight/loose connector can give you fits.  That particular duplexer, as I recall, has a piston (plunger) type notch capacitor.  So like the main tuning rod, you need to ensure that it too is securely tightened (once tuned). 

If it''s an older duplexer with some 'mileage' on it, the cables and/or their connectors could also be suspect.  I seem to recall the 696 used RG-214 or similar jumpers with soldered on N-type connectors.  Don't be afraid to wiggle these to test them, like the connectors being tight enough, a bad cable or one of it's connectors can be a headache.

I don't know what VNA you are using, but I will say that if it's one of the NanoVNA types, as good as they are, they simply do not have enough dynamic range to accurately tune a duplexer.  You can get really close to be sure, but it gets really difficult to accurately tune the notch side and that, can make a difference of several dB of loss elimination.  IMHO, the notch is almost more important than the pass side.

While certainly not the only way, the best method to check/tune a duplexer is with a professional VNA (Tektronics, etc...) or, with a quality/decent spectrum analyzer with at least 120 dB of dynamic range (preferably more).   I use a directional couple with the Spectrum analyzer and tracking generator to tune and to check the return loss.  The latter is to a large extent, is like vswr, but is far more obvious about where you pass or notch is and how good you have it.

If you get good results testing into a known load, then and only then check the system with the antenna.  Never check a duplexer with the antenna system connected as it can skew the results.

Sorry for the long message, but I hope it helps you out!

BTW, those CCDs (Cheap Chinese Duplexers)... it's not (necessarily) because they're Chinese... Those things are crap no matter who makes them!

 

[Colin927]

@WROZ250 The VNA I am using is a Nano type but it is supposed to do 90db no problem, it's not a cheap $65 unit, it's an actual branded vna. https://www.amazon.com/gp/product/B099Z9J7ZC/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1   Also per the tuning instructions I was to remove and set aside the cables " and make sure they go back in the same order and orientation" and tune each can individually, pass first then notch, when doing that the notch for each individual can only drops down into the 40-45 db range so the VNA was well within its specs from what I can tell. I did buy replacement N type connectors and was going to pick up some new rg-8U cable if someone recommended to do so. Do these SWR readings seem odd? or is this the norm for this type of duplexer, It may be normal I just dont know enough.. testing into a 50ohm dummy load doesn't change the results, same numbers. also I did check all connections and shook everything while on the analyzer and no weird shifts in the waves.

 

Thanks

Colin

[WROZ250]

2 hours ago, Colin927 said:

@WROZ250 The VNA I am using is a Nano type but it is supposed to do 90db no problem, it's not a cheap $65 unit, it's an actual branded vna. https://www.amazon.com/gp/product/B099Z9J7ZC/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1   Also per the tuning instructions I was to remove and set aside the cables " and make sure they go back in the same order and orientation" and tune each can individually, pass first then notch, when doing that the notch for each individual can only drops down into the 40-45 db range so the VNA was well within its specs from what I can tell. I did buy replacement N type connectors and was going to pick up some new rg-8U cable if someone recommended to do so. Do these SWR readings seem odd? or is this the norm for this type of duplexer, It may be normal I just dont know enough..

 

Thanks

Colin

The thing about vswr is that 50 ohms isn't quite/exactly 50 ohms between devices.  There is always minor differences (Very Minor) between say the analyzer/VNA and the repeater itself. So not having a perfect vswr is normal.  Having a vswr that jumps around, not so normal.  Most duplexer specifications will include a nominal vswr to expect/consider normal.

I agree with you about the using VNA with the individual cavities. 

However, when assembled as a whole, there is typically some interaction between the cavities and the cables and so tuning a duplexer is more than simply tuning the individual cavities by themselves.  Indeed, bad inter-cavity cables can really cause some issues.  Bottom line, no matter how well the individual cavities are tuned (independent of the duplexer as a whole), most of the time when connected them together to for the duplexer, some additional tuning is required.

Additionally, if you repair or replace the inter-cavity cables, know that to some extent, their length is critical and so will also impact the overall tuning.  If you use a different type of cable (RG-8 verses RG-214) the length may be different depending on the velocity factor.

There are many excellent videos on Youtube for explaining and performing duplexer tuning.

If you want the best possible performance, a duplexer has to be tuned properly. 

The mere fact that it is a commercial grade base/repeater duplexer is irrelevant.  I had people send me these units to tune and some were so poorly tuned and/or repaired, the CCD types were actually better by comparison.  The goal is to squeak out every possible dB of rejection and pass to get optimum performance.  That again, can only be done through the proper procedure with the correct equipment.

All that said, I have seen people tune a duplexer using a signal generator, a step attenuator, and a good receiver with either an S meter or SSI output.  It's a 'down and dirty' way to do  it, but done properly and carefully, one 'can' get some pretty good results.  Not nearly as good as the proper procedure, but not horrible either.

 

[Colin927]

@WROZ250 Ok, that is some great info! I have watched the youtube videos, probably too much lol. The swr is not really jumping around unless I'm not understanding, its just reading less going into the duplexer that the notch style one did. if I go directly into the swr meter and into a dummy load I get 48 watts out of the repeater just as I did on the notch duplexer, the new duplexer seems to be drawing about 8 watts from the repeater at the back of the repeater and more like 12 watts at the end of the jumper cable going into the duplexer.

 

Colin

 

[WROZ250]

11 minutes ago, Colin927 said:

@WROZ250 Ok, that is some great info! I have watched the youtube videos, probably too much lol. The swr is not really jumping around unless I'm not understanding, its just reading less going into the duplexer that the notch style one did. if I go directly into the swr meter and into a dummy load I get 48 watts out of the repeater just as I did on the notch duplexer, the new duplexer seems to be drawing about 8 watts from the repeater at the back of the repeater and more like 12 watts at the end of the jumper cable going into the duplexer.

 

Colin

 

Pretty sure the loss you described is not normal. 

I've never encounter that to my recollection, at least not with a properly tuned and working duplexer.  It would cause me to investigate further.  It could be a sign that the duplexer (as a complete unit) is not truly tuned correctly. 

It is possible that it could also be the difference between your RF load and the duplexer (impedance-wise), but even in that scenario, 8 watts across a jumper is excessive IMHO.  Sorry, it's really difficult to diagnose a problem through messaging!

The reason I suspect the duplexer is that without being fine tuned as a unit, as I stated previously, simply tuning the individual cavities and connecting them together rarely (if ever) works and, any mismatches within the chain of cavities will be seen at the ports.  This again, might explain what you are experiencing.


 

[Colin927]

@WROZ250 Ok, I am going to try to find somebody locally here that can retune it, I'm sure it wasnt tuned correctly when I received it even though the guy that sold it to me said he would "super tune" it to my frequency, whatever that means... I calibrated my meter and checked the duplexer when I got it and it was only set at around 70db on the reject and 2.5 db or so on the pass on the receive side. I am in the Sacramento area here in California if you have any suggestions on where to go...

 

Thanks again for the info.

Colin

 

[WROZ250]

42 minutes ago, Colin927 said:

@WROZ250 Ok, I am going to try to find somebody locally here that can retune it, I'm sure it wasnt tuned correctly when I received it even though the guy that sold it to me said he would "super tune" it to my frequency, whatever that means... I calibrated my meter and checked the duplexer when I got it and it was only set at around 70db on the reject and 2.5 db or so on the pass on the receive side. I am in the Sacramento area here in California if you have any suggestions on where to go...

 

Thanks again for the info.

Colin

 

Unfortunately, short of a 2-way radio service center, no.  I have no idea what that would cost these days.  Alternatively, if you know any hams or perhaps even a local club, there is almost always one or two people who know how to do it and, have the proper equipment.

LOL! Super Tune to me says more bad than good about the person saying that.  There is tuned and there is untuned.

That said, general issues with the NanoVNA aside, if this duplexer is not giving you the expected performance, and you've already played with it, you really have nothing to lose and everything to learn by trying again.

There are ways to configure the nano settings to allow you to better see the combined tuning (as apposed to just one cavity).  The view takes more time in these modes, so patience is required, but it should help the resolution a lot.  For example, the bandwidth, sweep speed, and #sweep points can all be reduced or increased as you get closer to 'fine tuning', to allow you to see a more defined reject notch.

I believe W2AEW has at least a couple of Youtube videos on duplexer tuning with the Nano, and this is a guy with a lot of experience.  I'd suggest going to his list and search for Nano and duplexer to find them.

Nothing is a better learning tool than doing and trying.  Worse case scenario, you pay a service shop to tune it.

Sorry I can't be of much more assistance.
 

[n4gix]

13 minutes ago, WROZ250 said:

LOL! Super Tune to me says more bad than good about the person saying that.  There is tuned and there is untuned.

Well, I suppose the guy may have used an IFR/Aeroflex 1200 Super S to tune the duplexer.. That could be called a "Super Tune!" 

[WROZ250]

33 minutes ago, n4gix said:

Well, I suppose the guy may have used an IFR/Aeroflex 1200 Super S to tune the duplexer.. That could be called a "Super Tune!" 

Right up there with the guys bragging about how their CB was 'tweaked/tuned' and now has 120% modulation. 

I made a small fortune in my early 20s fixing these 'tweaked/tuned' CBs. 
"This guy told me if I tighten down the screws in those little metal cans I'll get more power". 
My all time favorite: "if you stick pins in your coax, they'll act like little antennas and improve your signal". 
Yeah, right, sure it does. (yes, people actually did stuff like that)
LMAO ROTF!!!