Keyboard Shortcuts
Likes
Search
PD526 Duplexer Tuning
|
I have an issue with tuning a duplexer. As you can see from the pictures the high pass looks good but the low pass has a double peak. I am having an issue trying to merge that double peak into one peak. Any suggestions? Do I have an issue with a high side can or do I just need to keep at it? Thanks Robert |
开云体育Are you tuning each cavity separately for both pass and notch, then connecting first two together and then the third for each side. The connecting cables should be 11.5 inch, face to face, then look at each side with a 50 ohm termination at
the opposite end. If you the capability to look at RL, you could touch that up too. I have tuned many of this model with no problem. Good luck.
Charlie Adams K3HKZ
Charles Adams
Industrial Communications Co.
Mobile 610-360-0050?
Office 610-253-1214?
cadams@...
From: [email protected] <[email protected]> on behalf of Part 15 Engineer via groups.io <kc8gpd@...>
Sent: Tuesday, April 22, 2025 6:11:50 AM To: [email protected] <[email protected]> Subject: [repeater-builder] PD526 Duplexer Tuning ?
I have an issue with tuning a duplexer. As you can see from the pictures the high pass looks good but the low pass has a double peak. I am having an issue trying to merge that double peak into one peak. Any suggestions? Do I have an issue with
a high side can or do I just need to keep at it?
Thanks
Robert
|
|
One version of the factory tuning instructions says to remove all cables, tune each pass/notch separately, attach all cables and put into service. I was skeptical but tried this and it resulted in a really good working duplexer. This is how I usually tune a 526 these days while also taking a look at the overall pass/notch after cables are attached and a repeater desense test. So far I've tuned three 536's this way and have not had to do any touch up tuning using this method. If I can find these instructions I can post them.? |
|
Well, per the suggestions here I did get it tuned up and it looked textbook on my r2600 but when I put it on the quantar the quantar went into pa fold back high vswr. I've never had that happen before when a duplexer looked good on a service monitor. Back to the drawing board. Try different known good cabes and retune it and see if maybe I had a bad cable throwing off readings. On Tue, Apr 22, 2025, 9:49?AM Mike via <prcradio=[email protected]> wrote:
|
开云体育Remove all cables and tune each cavity individually - tune the pass for best match (return loss), then set the notch.? As long as the notch wasn’t far off from where it needed to be, adjusting the notch will not “pull” the pass.? Then harness everything back together and check the results – it should meet or exceed spec. ? The factory tuning instructions assume you only have a signal generator and some means of detection.? It is decidedly a dumbed-down approach that is going to lead to mediocre results at best. ? ??????????????????????????????????????????????????????????????????????????????? --- Jeff WN3A ? From: [email protected] <[email protected]> On Behalf Of Part 15 Engineer via groups.io
Sent: Wednesday, April 23, 2025 9:02 AM To: [email protected] Subject: Re: [repeater-builder] PD526 Duplexer Tuning ? Well, per the suggestions here I did get it tuned up and it looked textbook on my r2600 but when I put it on the quantar the quantar went into pa fold back high vswr. I've never had that happen before when a duplexer looked good on a service monitor. Back to the drawing board. Try different known good cabes and retune it and see if maybe I had a bad cable throwing off readings. ? On Tue, Apr 22, 2025, 9:49?AM Mike via <prcradio=[email protected]> wrote:
|
|
That was actually the procedure I used and got it looking good on the 2600s tracking gen. I will try it with band new swept known good cables and do the procedure again and see if I get a different result. I will also throw it on the nanovna and see what it looks like there. I only have 4 or 5 duplexers under my belt. Certainly not the experience many of you all have. I'm still in the learning stage. On Wed, Apr 23, 2025, 7:11?AM Jeff DePolo WN3A via <jd0=[email protected]> wrote:
|
|
No I did not. I think I have a 2, 3, 10, 15, 20 db pads but no 6. I'll ask my friend if he has a 6db 50 ohm pad. On Wed, Apr 23, 2025, 9:31?AM Mike via <prcradio=[email protected]> wrote:
|
开云体育The value isn’t critical, just that the pad exists. Compensate your measurements by the pad value you have. The pad provides a known reference impedance, isolating the unknown impedance of the test equipment. ? Matt AL0R ? From: [email protected] <[email protected]> On Behalf Of Part 15 Engineer via groups.io ? No I did not. I think I have a 2, 3, 10, 15, 20 db pads but no 6. I'll ask my friend if he has a 6db 50 ohm pad. ? On Wed, Apr 23, 2025, 9:31?AM Mike via <prcradio=[email protected]> wrote:
|
|
Oh ok, Roger Roger. I'll try that when. I get home later today. On Wed, Apr 23, 2025, 9:44?AM Matt via <al0r=[email protected]> wrote:
|
|
The pass response curve from antenna port to TX port (and RX port) is much broader for the pass frequency than for the best return loss curve, and the 2 curves probably do not line up in your case. It may be that your nano VNA will show you this. You just don't have the dynamic range on most nanoVNAs to measure the notch depth, but it seems you can interpolate between the skirts it's center frequency,. If the duplexer cables from antenna port to TX side and RX side are the optimum lengths, then it is possible for both best pass and best return loss to match up. That is why Jeff said it is usually a compromise? not to use, for BP/BR duplexers, a return loss bridge or VNA.
?
It seems to me that if you find, with your NanoVNA, that with the TX frequency notch on the RX side of the duplexer, the skirts of which are centered on the TX frequency (with the entire duplexer assembled) and at the same time the best return loss is say on the high side of the best pass frequency, the coax length between the antenna T and the RX side cans is too short, etc.
?
John
|
|
The exact value is not critical, just has to be 50 ohms.? The 2 and 3 in series or the 10,15, and 20 would be ok but not many service monitors would have enough power to over come the loss of much over 6 db. Ralph ku4pt
On Wednesday, April 23, 2025 at 11:40:14 AM EDT, Part 15 Engineer via groups.io <kc8gpd@...> wrote:
No I did not. I think I have a 2, 3, 10, 15, 20 db pads but no 6. I'll ask my friend if he has a 6db 50 ohm pad. On Wed, Apr 23, 2025, 9:31?AM Mike via <prcradio=[email protected]> wrote:
_._,_._,_
,_._,_
|
|
I have 2.0 meters of 1/4 inch andrew hardline I'm going to use for measurements and then for when the duplexer is permanently installed. On Wed, Apr 23, 2025, 9:48?AM John via <jhaserick84=[email protected]> wrote:
|
开云体育You don’t want to use pads when you’re tuning for return loss…which is how you should be tuning the pass. ? From: [email protected] <[email protected]> On Behalf Of Part 15 Engineer via groups.io
Sent: Wednesday, April 23, 2025 11:40 AM To: [email protected] Subject: Re: [repeater-builder] PD526 Duplexer Tuning ? No I did not. I think I have a 2, 3, 10, 15, 20 db pads but no 6. I'll ask my friend if he has a 6db 50 ohm pad. ? On Wed, Apr 23, 2025, 9:31?AM Mike via <prcradio=[email protected]> wrote:
|
|
I put the pad between duplexer and Input on service monitor? Or do I put one of both into duplexer and out to service monitor? On Wed, Apr 23, 2025, 10:52?AM George Csahanin via <george=[email protected]> wrote:
|
开云体育Pads are not needed with a VNA.? They might be needed if you’re using a sig gen or tracking generator as a source that doesn’t hold its output level constant when presented with a mismatch, but you wouldn’t be turning for return loss then, would you? ? Pass gets tuned for return loss. ? Notch gets tuned for maximum insertion loss. ? From: [email protected] <[email protected]> On Behalf Of George Csahanin
Sent: Wednesday, April 23, 2025 12:53 PM To: [email protected] Subject: Re: [repeater-builder] PD526 Duplexer Tuning ? I've used 3db for this and 10db pads. Successful either? way. But the pads are needed.? ? I'm working on a 900 Quantar right now folding back power but it is an internal issue with the PA, not duplexer not antenna system ? GeorgeC W2DB |
|
Check that the center pins of all your connectors on the cables are at the correct height. Also give a wiggle to the connectors at the cavity to see if your readings change significantly.? I've had a set of Wacom cavities where the connectors had cold solder joints on the center pins and With the PD-526 I've had a set where someone had pushed the center pin of the connector down to where it was barely making contact, and I've had another where the notch capacitor and coupling loop was damaged with the solder ripped off the ground end. On Wed, Apr 23, 2025, 11:36 Jeff DePolo WN3A via <jd0=[email protected]> wrote:
|
