3

I'm sure this seems like a Google question, but I've BTDT and come up empty.

The problem I'm trying to solve is to obtain cables that are as "leak proof" as possible. Tried Pasternak "100% shield" and they leak like a sieve. Tried Andrew FSJ1-50 (F1B-PNMBM-1M) which is pretty good, but leaks at both ends, apparently due to the joining to the connectors.

Given that a commercially made cable isn't good enough here, I'm tempted to make my own, and if I'm going to that much trouble I want the best. I keep seeing references to "Superscreened" or "double Superscreened" coax, which has braid and mu-metal, but I can't find anywhere that sells such cable.

I'm hoping someone here knows where I can find it, and the appropriate BNC ends so I can fab some quality cables. I'm not too concerned about cost.

Premade 1m cables, BNC male to BNC male would be best, but only if the connectors are put on so that they don't leak.

I am working inside a faraday cage sending test signals to multiple antennas, and I don't want the coax radiating anything significant. The ferrite bar antennas have a gain on the order of -80db so leakage from the cables can dominate the signal at the receiver, and I can't use any other type of antenna here.

If there's another way to solve this, that would be great.

Frequency of interest is <1MHz.

mike65535
  • 1,035
  • 1
  • 11
  • 25
user103218
  • 319
  • 1
  • 6
  • 2
    Could you add some more information about what your application is that is observing such leakage? We'd like to avoid purely "shopping" questions, and so it's better to state the problem and leave the question open to alternative solutions. (I'd particularly be concerned that your _equipment enclosures_ are possibly “leakier” than your coax.) – Kevin Reid AG6YO Dec 10 '19 at 18:06
  • The leakage from common coax, even Pasternak "100% shield" coax, is pretty severe. It's easy to see if you have a small CW reciever handy. Terminate the far end of the coax (50 or 75 ohm as appropriate) and connect the close end to a signal generator set to your frequency of interest. Bring the reciever close to the coax, and you'll hear the carrier. In my application, I am working inside a faraday cage sending test signals to multiple antennas, and I don't want the coax radiating anything significant. If there's another way to solve this, that would be great. – user103218 Dec 10 '19 at 20:22
  • 1
    "Does it exist, what is its most common name ..." is a good frame to use, in my opinion. The info about the faraday cage should also go into the question (comments are meant to be disposable). Thanks! – Kevin Reid AG6YO Dec 11 '19 at 14:58
  • Trying to understand what you are using... is there a page that describes Andrew FSJ1-50? Your application sounds like you might need semi-rigid or rigid coax. Or perhaps you are already doing that? – Chris K8NVH Dec 11 '19 at 17:38
  • 2
    Perhaps your problem is common-mode currents on the outside of the shield. Are your shields bonded to the Faraday cage where it exits? – Mike Waters Dec 11 '19 at 17:46
  • Yes they are, but the problem also shows up simply attaching the cable to a signal generator. – user103218 Dec 11 '19 at 17:49
  • @Chris: FSJ1-50 is a "superflexible" hardline. https://www.commscope.com/catalog/cables/product_details.aspx?id=1342 It's shield effectiveness falls off some below 1MHz due to the transfer impedance. http://www.repeater-builder.com/antenna/andrew/andrew-shielding-effectiveness.pdf – user103218 Dec 11 '19 at 17:50
  • That solid shield looks like it as good as can be reasonably expected. Have you tried taking the pre-constructed coax and wrapping copper tape where the BNC meets the shield? (Possibly need to remove some insulator) If it is just leaking through there, this would seem to be easier than starting over. – Chris K8NVH Dec 12 '19 at 22:07
  • Thought about it.. I haven't taken the andrew jumpers apart yet, but that's really a separate problem. I want to find the supershield coax so as to minimize the uncontrolled variables in my setup. The transfer impedance is supposed to be something like four orders of magnitude lower. – user103218 Dec 13 '19 at 13:27
  • I wonder if you could get by with a decent quality coax cable inside a flexible metal conduit that's grounded at both ends. Has anyone tried something like this? – mrog Jul 15 '20 at 16:31
  • LMR-400 is also double shielded. and heavy! – user10489 Jul 15 '20 at 16:49

2 Answers2

5

It turns out that these cables were made by TE Connectivity. Some time back, they sold that business to Commscope. After talking to Commscope, it looks like those cables no longer are made. :(

Their recommendation is to use FSJ1-50, even though RG-214 has a lower transfer impedance at this frequency, because the shield coverage is better. So I will have to fight the leakage at the joints of the connectors and cables.

Anyway, if someone else can benefit from this info, here it is.

user103218
  • 319
  • 1
  • 6
0

What you are looking for is Andrew Heliax®. This page from the full PDF indicates that it has in excess of 120 dB of isolation from 10 MHz to 10 GHz. It has a solid yet somewhat flexible copper shield.

Andrew HELIAX leakage vs. frequency

Have you looked at double- or quad-shielded coax with the inner foil bonded to the dielectric? It is more flexible than Heliax®.

Unless you are willing to give up on BNC connectors, you will likely need solid adapters at each end of the cable. However, perhaps you can find BNC connectors for this cable.

Mike Waters
  • 7,843
  • 4
  • 16
  • 50
  • 1
    Confirmed with Commscope what I am looking for is no longer made. Heliax (which I already have) is the best currently available, so that's what I'll be using. I'll have to work out the issue with the connector leaks. – user103218 Dec 20 '19 at 00:39