Comparing connection losses

[SCPD]

Apologies if this has been answered elsewhere but I've not been able to find any good references to my specific need (700/800 MHz) when it comes to PL/259.

As it stands today I've got a 770 MHz Yagi and a 2M/440 J-Pole set up in my attic. They feed a Comet Duplexer. I'm using the Yagi because the system I primarily monitor is running simulcast and my house is in a bad spot to monitor the system with a normal antenna. At the time of the installation I cobbled together the connectors I had on hand. LMR 400 runs through the attic to the office wall where it terminates in a PL259 bulkhead. On the other side I've got a PL259 right angle adapter feeding the LMR 200 scanner coax which has 259 on the one end and connects to the scanner with BNC.

Honestly things are working well enough right now but I would like to try to pick up an 800 MHz system that is pretty much in line with my Yagi's current orientation. I realize that the bandwidth isn't ideal but the real trouble is the distance. At times I can copy it and at other times the signal is just barely adaquate so there is a lot of distortion.

With all of that said, here's what I'm wondering, how much signal loss could I eliminate at the 800 MHz band if I reduced several connectors from:

PL/259 -> Bulkhead -> PL/259 right angle adapter -> PL/259 Plug -> BNC -> scanner

to

N -> Bulkhead* -> BNC -> scanner

* Pasternack makes an N to BNC bulkhead that would probably suit my needs.

For what it's worth the antenna side of the attic line is already N so I don't plan to change that.

Thanks for any insight you can offer!

 

[prcguy]

It depends on the type of PL-259 and who installed it. Its been awhile since I've measured PL-259 loss but two good quality PL-259s and the bulkhead as an assembly probably have about .5dB loss at 800MHz. The big unknown is the PL-259 right angle adapter and some cheap Chinese versions have a spring inside as the conductor that makes the bend, which is really bad at high frequencies.
prcguy

 

[SCPD]

Thanks. The right angle is Amphenol. I've pulled it out of the circuit and it has not made any appreciable difference, granted I don't have any test equipment to measure that by. So it sounds like at best I'll eliminate less than 1dB of loss. Doesn't seem worth it.

I did one of the 259s myself. I did tons of those years ago so I think I've got those down pat. On the LMR 400 I had the commerical two way shop that sold me the cable do it since they were attaching the N connector at the other end anyway.

 

[jim202]

Not trying to throw sand on any of the conversation here, but a PL-259 is not a very good connector to use on signals much over 100 MHz. or so.

You would be much better off using a type N connector for the higher frequencies. That is what they were designed for. They work well up into the GHz. range.

You will also find that the center conductor material has a good bit to do with how high in frequency a coax connector will work at. This is why you will find that you can get the PL-259 connectors with Teflon insulators. But still the PL-259 connector was designed for the HF and lower VHF frequencies.

 

[Voyager]

I've never found a good UHF connector at 800 MHz no matter who installed it.

UHF connectors are like a brick wall at those frequencies.

 

[SCPD]

Yep I'm well aware of the limitations of UHF connectors. What it came down to was what I had on hand when I put everything together. Since it worked acceptably well for what I was trying to monitor I didn't see a point in spending an extra $75-100 in connectors that I didn't have on hand.

Now I'm just trying to get a perspetive on how much of a performance hit I'm actually taking to decide whether it's worth it to go back and use Ns just to try to pick up a system that I'm moderately interested in monitoring. Since the system is marginal to gone at times I'm questioning whether even a 3dB improvement would be enough to make it an enjoyable listening experience.

Kind of the old "if it ain't broke..." philosophy since my primary system of interest is rock solid.

 

[Voyager]

Really, if it's rock solid, why bother making any change even if you get 100 dB of improvement of the signal? Any improvement may reveal unintended consequences - such as overload issues. Or are you just wondering how much loss there is in theory?

 

[w8prr]

You also have some loss in your duplexer. If possible, try running straight cable with N connectors from your antenna direct to your radio. If you can then receive the desired signal, you know its possible at your location and you can work to get the signal to your radio with whatever connectors, adaptors and duplexers suit your needs.

 

[SCPD]

Really, if it's rock solid, why bother making any change even if you get 100 dB of improvement of the signal? Any improvement may reveal unintended consequences - such as overload issues. Or are you just wondering how much loss there is in theory?

Because there is another system that I'm somewhat interested in listening to if it can be reasonably accomplished. I guess it's really a matter of prioritization which is why I'm leaning towards just living with it for now...

 

[prcguy]

Many years ago I measured the insertion loss of two PL-259s and a barrel adapter with a scaler network analyzer at around .5dB loss at 1GHz and if my memory is still intact I used Amphenol silver plated connectors with Teflon insulation.

Motorola used PL-259 connectors on their UHF duplexers up to 500MHz and with jumpers to the transmitter and receiver I believe you would have six connectors in the receive path and six in the transmit path plus an SO-239 "T" adapter and there was not enough loss there to worry about at 500MHz.

I would not use them in an 800Mhz system but the actual loss is probably not that bad.
prcguy

Not trying to throw sand on any of the conversation here, but a PL-259 is not a very good connector to use on signals much over 100 MHz. or so.

You would be much better off using a type N connector for the higher frequencies. That is what they were designed for. They work well up into the GHz. range.

You will also find that the center conductor material has a good bit to do with how high in frequency a coax connector will work at. This is why you will find that you can get the PL-259 connectors with Teflon insulators. But still the PL-259 connector was designed for the HF and lower VHF frequencies.

 

[n5ims]

While it's true that an N connector will provide less loss than a typical UHF connector, the additional loss will not really matter. If it does, then you need to upgrade your components where the gain (well, lower loss in this case) will actually be worth it. If the UHF connector is so bad, why do the major land mobile radio makers use it (or the smaller mini-UHF connector) on their radios?

Most likely changing to a better coax (with lower loss connectors while you're at it) will give you a change that you will actually be able to see. Assuming a 100' run (at 860 MHz, a typical 800 MHz system scanner frequency) you'll have 2.8 dB of loss using your LMR-400, but moving to a slightly larger (7/8" vs. 1/2") size of heliax (a better grade of coax), LDF5-50A will reduce the loss to only 1.15 dB (you could still do better using an even larger size of heliax like LDF6-50A where it would be only 0.823 dB of loss.

 

[SCPD]

The LMR 400 run (antenna to wall) is about 15 feet. The LMR 200 (wall to scanner) is about 6. Probably not much to be gained from a cable perspective. Even hooking the scanner directly to the antenna only yielded a very modest improvement. I think I'm going to keep things as they are for now. Doesn't seem that I have abundant opportunities for significant gain, no pun intended.

Thanks all for the responses!