GAIN with a right angle connector ???

Hardwire001 · Post by **Hardwire001** » Wed Jun 28, 2006 1:49 pm

Ok, I have not been drinking yet...

While trying to clean up the look of a GR300 repeater,( and keep the coax out of the fan) I stuck a 90 degree "N" on the back of the duplexer to the TX radio... and I got more power out of the unit ????

It went like this I plugged the radio into the service monitor and measured tx power out ( 21.4 W) before the duplexer...

I then plugged the duplexer in and measured tx power out ( 17.4 W) out of the duplexer, not great but not bad...

I then added the afore mentioned 90degree into the mix and measured tx power out agian to see how much I would loose by adding the 90 but I got 18.2 W... now maybe some else can tell me WTF ???

Know I have been doing COM's for a long time, and I remember the rule of thumb, every connector in the line expect to loose 3dB's....

I have tried several test cables ( thinking one maybe faulty )... but the number were almost the same ( 0.5W to 0.9W ) in GAIN

Thanks in advance

HW

NOW I am going to start drinking....

Will · Post by **Will** » Wed Jun 28, 2006 2:06 pm

Your transmitter can be changing the TX power level due to matches or mismatches on the output. Even after the duplexer changes in the accuall load would be sensed by the radio.

I would leave the N rt angle adaptor in there!!!!

A good quality N right angle connecctor should not mess up the match, EXCEPT that the lenth of the transmission line has changed by a mire 1/2 inch. The input impeadance of the duplexer TX side may not be right at 50 ohms, depends on the duplexer type and alignment.

Loss would be way less than 1/2db with the N right angle adaptor, everything else matching 50 ohms.

There are some cheeeep N connectors and adaptors out there, some are even 75 ohm.

Now before we both have to need a drink, while working on the JPL antenna range, there was this cable about 20" long that everytime we put it into the test setup, we got a gain! The cable with it's N connectors is still in the Arroyo where I slung it! Another cable worked fine...

Post by **wavetar** » Wed Jun 28, 2006 2:38 pm

Yep, Will's correct...the duplexer isn't presenting a perfect 50-ohm load to the radio, and the length offered by the rt-angle connector is helping it in that regard. Tuning the duplexer should provide a proper match & maximize your output power...without the rt-angle connector.

Todd

N9LLO · Post by **N9LLO** » Wed Jun 28, 2006 3:39 pm

Not sure what you are using for a wattmeter but the Bird Wattmeter manual has information on the specific cable length to use with the Bird Thruline measurement device to equal 1/4 wavelength. Using this length with the Thruline wattmeter will allow you to read the true SWR and power without introducing errors due to test cable length.

Chris
N9LLO

AEC · Post by **AEC** » Wed Jun 28, 2006 6:03 pm

As a side note, I always connect the watt meter directly to the duplexer output without a cable.

Male on the watt meter's input, female connector on the output directly attached to the meter, this way, there are no 'extra' cables in the system aside from the line in the meter itself.

You also get no reradiation with this setup and the readings are far more accurate as well.

You do need to hold the meter while doing this, the only drawback to the setup...

If the system was designed without 'spare' cables, then I do not add more.

BpBr: the only way to go.

WPA6MDuplexer · Post by **WPA6MDuplexer** » Wed Jun 28, 2006 7:04 pm

N9LLO wrote: Using this length with the Thruline wattmeter will allow you to read the true SWR and power without introducing errors due to test cable length.

I would really like to see this laid out on a Smith chart. With the output of the wattmeter going to a good 50 Ohm dummy load that is. I need to see how line length affects an impedance transformation from the center of the 'chart' I guess is what I am saying.

As to an extra 1/4 wavelength line being placed in a line to anything - this is going to 'transform' that impedance where low becomes high and high becomes low or inductive becomes capacitive and vice versa (as the quater wave line 'rotates' the impedance value half way around the Smith chart at a constant radial distance from the middle from whereever one starts).

At this point I will defer to a source more eloquent than myself for further explnation:

How do you turn a short circuit into an open circuit, or a capacitor into an inductor? Here we will describe some of the magic that happens when you use the distributed properties of transmission lines, in particular when you use lines of length one-quarter wavelength, or multiples of a quarter-wavelength.

http://www.microwaves101.com/encycloped ... erwave.cfm

Mike B · Post by **Mike B** » Wed Jun 28, 2006 8:52 pm

At least you know you have a good right angle connector. We have occasionally run into cheap UHF right angles and T connectors that were put together using a coiled spring as the internal center conductor. If the obviously "bad" connectors had not been taken apart we would never have realized the design/manufacture was defective.

N9LLO · Post by **N9LLO** » Thu Jun 29, 2006 5:10 am

rwave.cfm[/quote]

I would really like to see this laid out on a Smith chart. With the output of the wattmeter going to a good 50 Ohm dummy load that is. I need to see how line length affects an impedance transformation from the center of the 'chart' I guess is what I am saying.

Sorry I meant to say 1/2 wavelength, as I'm sure you know. You dont ADD 1/2 wl to the meter you add a length of cable to the meter to total 1/2 wl as outlined in the Bird manual, this prevents the Thruline section from transforming the impedence and causing measurement errors
.
See

http://www.bird-electronic.com/products ... aspx?id=81

Every radio tech here has several of these cables on the truck, one for every land mobile band with pl259 on one end and a N on the other, the meter is equipped with an SO239 and N female. This gets you hooked up to 99% of the base station equipment out there.

Chris
N9LLO

WPA6MDuplexer · Post by **WPA6MDuplexer** » Thu Jun 29, 2006 8:44 pm

N9LLO wrote: Sorry I meant to say 1/2 wavelength, as I'm sure you know.

Ahh yes ... I see this when tuning duplexer stub pass bands on the stubs either side of the one in the middle (three stubs each leg); the two on the outside will effect each other (these are notch duplexers) and the one in the middle is separated from, and isolated from, each of the outside two stubs/notches via the 1/4 wavelength lengths of cable. (The two on the outside are connected via 1/2 wavelength with a 'stub' in the center; and, of course, the 1/2 WL doesn't impedance transform or 'isolate' as the 1/4 WL does.)

I've never owned a Bird straight from the factory, so I have to confess I have never read the manual! I always sweep antennas/check antennas with other gear and only use my Bird to measure power into a dummy load usually ...

440roadrunner · Post by **440roadrunner** » Fri Jun 30, 2006 7:11 am

Years ago Radio shack got a REAL bad rep for selling poor quality adaptors. Some of them actually had SPRINGS inside. Can you spell inductive?

One thing you might try with that right angle, is to rig the meter as radio-coax-meter-ANGLE-dummy load. Measure into the dummy load with the angle in, and out. A good angle should have no measurable change in reflected power