Friday, June 16, 2017

The PL259 myth

You often read or hear that every PL259 connector on your coax adds 0,5dB loss. This might be the truth on UHF and higher. But on HF/VHF that's not the case.

Recently I was thinking about the 50m Aircell7 coax I run to my antenna now through a homemade disconnect/patch panel. So, this coaxcable to my HF antenna contains 4 PL259 connectors and a female/female barrel to connect the 2 pieces. 50m Aircell7 has a loss of about 1,8dB on 28MHz and 2,4dB on 50MHz. I've no interest in VHF/UHF so leave that out.

At the new antennamast I want to have another patchpanel at the bottom of the mast like I had before in my previous QTH. I want to add some lightning protectors and a galvanic isolation transformer for the highest (HF) antenna in my mast. Now I was afraid all the PL259 connectors would add a significant loss to my transmission line. But after all I think it is not that much.

Steve Katz, WB2WIK/6 demonstrated at the Dayton Hamvention in 1985 that the loss in an 83-1SP PL-259 connector averages .0435 dB per connector at 28 MHz.
The primary difference between a "zero loss" PL-259 installation and a lossy one is how the connector is installed

PE4BAS experimental test site
However, I'm not a believer. I like to test this out. Although I haven't any high sensitive equipment, only my trusty MFJ-259B. Is the PL259 loss a myth or can I bust this? I did some test this evening and made some interesting discoveries.

My not-so-scientific test setup: MFJ-259B, 2 pieces of 10m long Aircell7 coax laid down in the too long grass with PL259 connectors both sides. Only soldered at the center. Braid is screwed into the connector when I assembled it in 2005.

10m Aircell7 has a calculated loss of 0,36dB. 2 PL connectors have a calculated loss of 0,087dB = 0,3687dB (0,37dB including the female/female barrel connector). Total loss of the 2 pieces calculated 0,74dB. Lets see:

Left: 1 Piece of coax - Right 2 Pieces connected
Of course the MFJ-259 is not very accurate and has losses itself. However, this is not a bad result.
Ready to do a practical real life connector test...

13 connectors in total
I connected the 2 pieces of coax with as much connectors and connecting barrels as I had. Besides 2 times PL259 I added 11 connectors. This should add a calculated loss of about 0,4785 dB (0,48dB). But in real life it only add 0,1dB.

Another test....add 4 times a 90 degree connector. Now this is going to be serious!!! There could be a faulty connector. So test just one of them:

Still higher loss. 1 90 degree connector 0,7dB loss,  The 90 degree corner does probabely affect the impedance of 50 Ohm?

PE1BVQ Hans is always searching for silverplated old PL (and other) connectors at radio rallies. Preferable the original amphenol ones. I asked him why, he says they are the best, others can add a lot of loss in your system. Could that be true? Or is this another myth that's going to be busted?

Let's see:

I changed the 90 degree connector for a silverplated original Amphenol one. Gone is the loss! How strange is that?

Well, of course I don't use any superduper sensitive low loss equipment but it shows that there can be a difference in connectors. As long as the connectors are mounted in a straight line it hardly gives any loss. Very important of course is a good soldering connection between the coax and the PL connector.

A few tips:

* Solder both core and braid. Best way in my opinion is how it is done by K3LR.
* Keep connectors clean so they can make a good lossless connection
* Tighten them well. I checked that as well this evening, a loose connector gives about 1dB or more loss and propabely a lot of unwanted RF around the loose connection!
* Prefer silverplated (Amphenol) connectors, they probabely are the best you can get.
* Amphenol too expensive? Look for them on radio rallies. I payed about 50 cents for the 90 degree connector.


VE9KK said...

Good afternoon Bas, great post on the PL 259 connector and funny thing I just ordered some new RG8X and silver plated PL 259 connectors and in 2 weeks we have our Canada Day long weekend (celebrating 150 years) and I have taken a few extra days off. I am going to use this time to replace the very old patch cables and feed cable to the antenna.

PE4BAS, Bas said...

Hello Mike, you're right. I think RG8X is the same as Aircell7. Just another name. Good luck! 73, Bas

Hans said...

Hallo Bas, leuk en interessant artikel. Dit zal een hoop mensen aan het denken zetten.
73 Hans, PE1BVQ

Photon said...

Very useful post! I had forgotten to use my own analyser (SARK-110) to test for losses!

The SARK reports that six PL259 connectors plus a lighning arrestor produces 0.02dB loss. Together with the WSPR 100mW results, I don't think I need to worry!

PE4BAS, Bas said...

You don't have to worry about normal PL259 connectors as long as they are mounted correctly. It's the PL-PL elbows that add loss to your system. If you really need to use one (for example behind your tuner or radio) you really need good quality connectors. 73, Bas

PE4BAS, Bas said...

Hallo Hans, gewone PL connectors geven eigenlijk weinig tot geen verlies. Het probleem zit hem vooral in die haakse PL stekkers. Sommige amateurs gebruiken die vaak achter de set of tuner (ruimte gebrek?) het kan je veel verlies geven + warmte die ontstaat in de connector denk ik. Lijkt me niet goed voor het hele systeem. 73, Bas

Unknown said...

I have encountered both right angle and tee adapters that exhibited high InsertionVSWR, worse with increasing frequency. Those that I cut apart used a small spring to make a contact on the centre conductor, they essentially inserted a small inductor in series with the centre conductor. This may be what you have... Owen.

PE4BAS, Bas said...

Hello Owen, those adaptors I tested went right into the old iron box. You'll probabely right about the construction, I've seen pictures on the internet lately. 73, Bas