Coax for Q100 - lots of questions....

[radiogareth]

Having finally got my Q100 station up and running I was quick to find shortcomings and so attempted upgrades.
Working back from the POTY on my dish, I have 1m of RG402 (90 degree SMA to N plug) that connects to a piece of 4-50 heliax. That makes it to the shack where I have an N-type joiner, N-to-BNC adaptor and then a 8" BNC-BNC coax patch lead salvaged from a UHF diplexer (silver plated core, PFTE dielectric, silvered braid OD 3.8mm. Then a BNC to SMA on the Stealth amp. The 8" pigtail gets hot in a few minutes. The amp is running about 50 watts DVBS2.
I've another piece of salvaged coax, from a military 4CX250 amp, conveniently ended in a N-male which will remove one N joiner and a N to BNC adapter - worth hile gain I'd expect. I can fit a BNC on the other end. It appears to be silver plated too, brown PTFE-like - slippy outer covering, 4.5mm diameter. Very stiff, so maybe silver plated steel core?
I'd assume this will be better that the braided jumper, but if not, what are people using for short patches? I have some rigid PTFE but it's also 3.8 diameter and NOT silver plated so I can't see that helping.
Finally, I have some nice 300mm Suhner RG223U SMA to SMA, I could cut one end off and fit an N male.
So some choices.....what are people using? Heat = loss which I'd clearly like to remove if possible. Its not the connectors getting warm.
Thanks
Gareth G4XAT

**** Update***
Found a SMA to N male adaptor so put the whole Suhner patch in line. Still gets warm, but not as warm. Also 3 less joints in this short part now, maybe 0.1 gain. I've nothing accurate enough to check exactly.....

[g0mjw]

Too many transitions. BNCs don't work well at 2.4GHz. TNCs are better.

I have a 6" piece of 0.141" semi-flexible from the Stealth PA. Right angle SMA on one end, N female on the other. Then a 4m length of FSJ4-50 with N males on each end, then another identical semi flexible from the Heliax to the feed. The semi-flex gets warm at high power but the FSJ4-50 doesn't. I think the loss is around a dB.

Try measuring the power at the feed vs the power at the amp. It can be done at low power or with a VNA. See what you are losing. Miss-match at transitions is an issue but very frequency / length dependent.

Mike

[g8gtz]

Hi Gareth

First of all congratulations and look forward to seeing you on Thursday night!

Without accurate power measurement you can't really tell but you would be amazed how much power you lose in connectors, adapters, right angle adapters which should be avoided and what looks like good co-ax.

Not sure how long your piece of 4-50 but there is a reason why most people have their power amplifiers outside under the dish and just a very short lead to the feed - even so I have burnt out what looked like a good SMS to SMA fly lead with just 50 watts of power. To get the amp outside should be your next aim!

Silver braided co-ax is normally good, BNCs are not good at 2.3 GHz and semi rigid is normally good as it has a PTFE inner.

So keep it short and avoid adapters at all costs are the rule.

Not sure if this helps much.

Noel

[g0mjw]

LDF4-50 has a very low loss at 2.4 GHz, its about 0.13dB/m so even quite long lengths are viable. Where the power gets lost is in the connections and when the cable is not good, for example it's got water in it or has been coiled up or bent one or more times to many.

I find this table of cable losses useful from time to time.

Clipboard01.jpg (123.11 KiB) Viewed 3097 times

[g8gtz]

Useful chart Mike but folks should be aware that it represents the cable as manufacturered - even second hand cable that looks fine may have got damp or damaged (as I found out a few months ago).

Plus it's unlikley you will acheive those numbers in practice after you've put connectors on the end, particularly 2nd hand ones from a rally or nice shiny Chinese ones off ebay!

Noel

[radiogareth]

Thanks for the pointers
Its not a very long run of 4-50. I measured it as 10m and for the moment the system 'works'. A bigger amp is in the 'works' but if 333ks is making it that's fine for the time being.
Just sorting out the shack - re-arrangement to accommodate TV and clear away 3+ years of bits/scraps/etc from building stuff!!

2 days so far.......

Gareth

[g7ocd]

I assume 4-50 means LDF, not FSJ? The latter has a fair bit more loss.

The most important thing with used cable is to measure it. Then you will know what your system limits are.
If the loss is a little more than 'book' loss then you may choose to live with it. If larger, you may want to
investigate to stop PA damage or even a fire! Of course 1dB loss in 8" is going to be far more of a concern
than 3dB over 30'.

With connectors, it's fairly obvious that more connectors = more loss but different connector types can
make a bigger different. For example, a cheap N with poor dielectric and 2 or 3 leaves on the
centre pin recepticle may have a lot more loss than a good BNC with decent dielectric and 6 leaves
on the centre in recepticle. For used connectors, the number of times it have been mated and the
cleanliness is also a big factor.

In my case, I currently have FSJ4-50 with surplus right-angle N types on both ends. I used an N connector
on the amp (~50W) and also on the back of the POTY so there are no transitions.
I had to print a replacement feed clamp with a slight cut-out to allow the right-angle connector to fit.

I'm still unhappy with the loss in this arrangement so am looking to move the amp into a small
weatherproof box under the dish to reduce feeder length and possibly replace feeder.

poty-feed.jpg (130.71 KiB) Viewed 2760 times

[g7ocd]

LDF4-50 has a very low loss at 2.4 GHz, its about 0.13dB/m so even quite long lengths are viable.

The published spec is even less than that, about 0.118dB/m and LDF5-50 is about 0.68dB/m at
2400. The problem with both of these is bending radius which often forces additional connectors
and high loss tails.

[g0mjw]

The problem with connecting the coax directly to the patch is the rigidity of the stuff. Dealing with differential expansion with temperature, wind flexing etc. Possible I am sure with care. I like that connector arrangement, you might even be able to use the coax as the feed support. I hope you are going to weatherproof it properly. Water has the ability to seep into everywhere and spoil everything. I have a 20m length of LDF4-50 with 3 dB loss on 70cm to prove it.

Mike

[g7ocd]

Yes, there is an overall radome which was removed to take the photo