Polyphaser Directionality

[synegy3488]

So I understand fully the installation literature and what the device does....however, being a curious fellow, and knowing people on here are likely smarter then me I want to pose this question to satisfy my curiosity:

• What exactly, other then perhaps distance inside the device, makes this a directional device?

It would appear that the only reason for "directionality" is to put the antenna connection closer to both the DC block and the Gas Tube. Otherwise, this device should work essentially the same in both directions (albeit perhaps with slightly less protection backwards). I can't imagine that just reversing the direction would cause either the DC block or the Gas Tube to not do their designed jobs (yet this is what the installation literature from Polyphaser states).

I partially disassembled one half of one of the ones I have:

https://imgur.com/a/4LsQemh

 

[synegy3488]

Nevermind. I'm a moron and a fool today -- and my brain totally failed on both Science and Electronics 101...

The DC blocking cap DIRECTLY impacts directionality; by preventing DC transients from reaching the downstream device, and when a transient surge current is high enough it will suppressing that currents thru the gas discharge tube.

 

[RFI-EMI-GUY]

If I recall correctly the amount of Joules "let through" was dependent upon the frequency range. So if you had one for HF range it would have a larger cap and therefore more Joules would escape to the radio. On the other hand one for 800 MHz would have a smaller value cap and thus far less Joules presented to the radio. And yes, the cap should be on the radio side.

 

[CARC383]

If I recall correctly the amount of Joules "let through" was dependent upon the frequency range. So if you had one for HF range it would have a larger cap and therefore more Joules would escape to the radio. On the other hand one for 800 MHz would have a smaller value cap and thus far less Joules presented to the radio. And yes, the cap should be on the radio side.

With a direct strike the coax will arc over at some point and probably take the brunt of the energy heading towards the radio on the coax center conductor. That will lessen the stress on a Polyphaser or other coax arrester downstream. What's just as important is the overall grounding of the site and making sure all grounds are at the same potential during the strike. That may be 5kV or more above normal for a few hundred milliseconds but if all grounds do it as one and the site AC entry point has adequate protection and the overall site ground is large enough and at a low enough impedance to dissipate the strike then you have a chance of surviving.

 

[Astro Spectra]

Best avoid GDT arrestors, don't take my word for it see Motorola R56.

GDT types let thru hundreds of volts during a nearby strike and are a maintenance nightmare, do you test your GDT arrestor strike voltage regularly as part of your PM? Of course not, no one does... so how do you know they're not blown?

Use only quarter wave shorted stub types, you be glad you did (except on HF where everything changes). You can even make your own if on a budget using RG214 and type-N T pieces.

 

[CARC383]

Best avoid GDT arrestors, don't take my word for it see Motorola R56.

GDT types let thru hundreds of volts during a nearby strike and are a maintenance nightmare, do you test your GDT arrestor strike voltage regularly as part of your PM? Of course not, no one does... so how do you know they're not blown?

Use only quarter wave shorted stub types, you be glad you did (except on HF where everything changes). You can even make your own if on a budget using RG214 and type-N T pieces.

At risk of being spanked I'll add that it depends on the station type. For a single antenna directly feeding a radio with no duplexer, filters, etc, a system with shorted stub should have a long trouble free life, at least for the antenna input side.

For a repeater station that has a duplexer, cavity filters, etc, there is a lot of additional stuff to protect the radio. Assuming everything else at the side is done right to mitigate a direct hit, the duplexer and other filtering will reduce or eliminate energy except within the frequency range of the duplexer and filters. The frequency spectrum for lightning can extend to 300MHz but the peak energy is down in the audio range, so not that much energy within the VHF band and less at UHF. If a Polyphaser fails and opens on a UHF repeater system, it may not matter that much since its only protecting the coax center conductor and whatever is connected to it. I would be more concerned with the rest of the system and if the AC powered components will survive.

 

[PSEhub]

In a lot of installs, you can't go making your own protection devices, even if they may be superior.

Polyphaser gas based aresstors are a COTS option with paperwork and a recognizable name. So for liability, and competitive bidding situations, sometimes you have to stick with industry standards and follow the herd.

 

[Astro Spectra]

There are quarter wave types COTS from a range of manufacturers including PolyPhaser.

Just because everyone else does it doesn't make it right. Have a read of R56 or talk to the Midland Oil and Gas folks.

 

[RFI-EMI-GUY]

Best avoid GDT arrestors, don't take my word for it see Motorola R56.

GDT types let thru hundreds of volts during a nearby strike and are a maintenance nightmare, do you test your GDT arrestor strike voltage regularly as part of your PM? Of course not, no one does... so how do you know they're not blown?

Use only quarter wave shorted stub types, you be glad you did (except on HF where everything changes). You can even make your own if on a budget using RG214 and type-N T pieces.

Not sure I have ever seen any 1/4 wave stub VHF or UHF protectors?

 

[synegy3488]

In a lot of installs, you can't go making your own protection devices, even if they may be superior.

Polyphaser gas based aresstors are a COTS option with paperwork and a recognizable name. So for liability, and competitive bidding situations, sometimes you have to stick with industry standards and follow the herd.

I had no intention of building my own arrestor (that's asking for trouble) and I hope no one ever builds a homebrew arrestor -- it was more of a curiosity question, some what due to ignorance and some what due to my neurons failing me.

However, and it wouldn't/shouldn't be a exercise for the home user (i.e. DON'T DO THIS EVER), it would seem to swap the protected side, simply moving the GDT from one connector to the other would "flip" the sides, since the DC block is always in between the center conductor on these.

 

[Astro Spectra]

Enough of the building your own arrestors already, did I say you had to make you own? I simply pointed out that you could make your own if on a budget.


Starting with PolyPhaser quarter wave types, the go to device I specify and use (like hundreds) for 700 and 800 MHz sites is the TSK series available in both type-N and DIN versions. Here's the N-N TSX-NFF:



For longer feeder runs on 100 and 300' towers we generally use DIN connectors on larger sizes of Heliax so the DIN version of TSK PolyPhaser (mostly TSK-DFF but -DFM shown) is most useful. These are mounted thru a bulkhead panel with smaller DIN to type-N coax jumpers inside the shelter:


The TSK is a great arrestor and can handle 500W so can be used for large multi channel combiner setups where powers can reach those levels.

Bulkhead types are great when you have to handle large diameter cables. Here's a nice product from Times Microwave that is a ready made panel for their own range of protection devices that gives the idea:



The angled entries helps shed water.



For smaller feeders (less than 30mm or so) you can bring them in but pay attention to grounding). Here's a site I worked on as a good example of an indoor arrestor grounding plate (the arrestor grounds are the black cables).




This is not necessarily the best arrangement as I prefer to bulkhead mount arrestors thru the shelter wall and then ground the wall plate outside the shelter to reduce radiated EMI from strike currents travelling thru the ground wire. Of course you'll have installed a dedicated site lightning ring ground [reference 1].



@RFO-EMI-GUY Not sure I have ever seen any 1/4 wave stub VHF or UHF protectors?

I well understand that not everyone has migrated to 700/800 MHz (the PolyPhaser TSK low end limit is 698 MHz). So for legacy VHF/UHF (I say legacy with my tongue firmly cheek) here's the well priced LGP range from Amphenol Procom that cover 140-200 and 350-520 MHz. Again available with NFF or DFF connector options:



Obviously the quarter wave sections are longer in these lower frequency devices. In the case of the LGP series the overall length ranges from 196mm (UHF) to 355mm (a bit over a foot in stone age measurements) for the VHF type. These guys also handle 500W. There are other vendors out there.

Now to revisit the downsides of "industry standard" GDT types:

GDT types require regular testing, they are not install and forget devices [reference 2]​

GDT tubes have to be sized for the RF power so that they don't strike with the applied RF voltage​

Gas tubes do not completely clamp surges, typical ‘let through’ is 1,500 V peak at strike​

On clamp voltage, here's what Motorola says in R56 (see section 7.6.1):

Do not assume that a 90V gas tube will provide a 90V protection level for the equipment, because the voltage rating is based on a static DC voltage measurement. The Gas Tube is essentially a voltage dependent switch that reacts to the dv/dt of a lightning impulse. The typical voltage breakdown level for a 5kV/µs impulse is approximately 700 Vpk.

Voltages will be higher with GDT tubes designed for higher powers. A 90 volt GDT will strike with RF power a bit over 70W. For 100W transmitters you need to be using a higher strike voltage so the let through will be >700 Vpk.

As Motorola points out and as @CARC383 noted in his post, a decent band pass duplexer greatly limits the amount of broadband power from an event being coupled into the radio. However, protection on every feeder is still good practice.


References
[1] Telcordia Technologies Generic Requirements, Generic Requirements for Network Elements Used in Wireless Networks Physical Layer Criteria, Document Number GR-3171 CORE, Issue Number 02, December 2014
[2] Motorola R56 Committee STANDARDS AND GUIDELINES FOR COMMUNICATION SITES, Motorola publication 68P81089E50
[3] ITU Recommendations SERIES K: PROTECTION AGAINST INTERFERENCE, Protection of radio base stations against lightning discharges, ITU-T K.56, January 2010
[4] Military Handbook 419 GROUNDING, BONDING, AND SHIELDING FOR ELECTRONIC EQUIPMENTS AND FACILITIES, MIL-HDBK-419A, December 1987

 

[RFI-EMI-GUY]

Clever.

LGP

amphenolprocom.com

 

[PSEhub]

In my experience of several direct (and indirect hits, the feedline and duplexers get damaged/disintegrate well before anything happens to the repeater. At most, a new receiver FRU.

Usually commonly ordered items are stocked by Talley and Tessco and available on their websites. The fact they don't seem to have UHF quarter wave stub protection devices is an indicator of their popularity.

The Polyphaser gas based cover a broad frequency range so you can use multiband antennas, are inexpensive, and compact for VHF and UHF. UHF T band is very popular with some of the largest metro area public safety agencies in the USA, and it isn't going anywhere. There are also hill top APX8500 remote bases and consolletes using multi-band antennas.

Many of the largest airtime services are also UHF. I would disagree that UHF or VHF is older than 7/800. That's like saying UV is older than Infrared.

There isn't room for everyone on 800 in many metro areas

The quarter wave stubs have advantages, but not without drawbacks