Jump to content
Sign in to follow this  
snooks

Patch Antenna

Recommended Posts

Hi There

I've built a couple of the patch antennas to run with my gigaair 2727 tx/rx settup to get video transmitted in swimming pool see earlier topic. For a number of reasons I have had to use bnc connectors and bnc/bnc couplers to fix them direct to tx and receiver. Is their any way (other than simply using them) to test whats happening beam width and performance wise. I'm not looking for power or range simply directionality to prevent multipathing break up with this ground based application.

For practical purposes I've had to to fit a chassis style bnc connector to the reflector plate and therefore the nut lies on the surface of the reflector i.e. between reflector and collector, does this matter as it is not a totally flush surface therefore on the reflector? (dont forget I'm not after power simply a 60/80 degree directionality to eliminate multipathing in a partly steel building)

Finally (at last) even with the aerials removed tx and rx still 'talk' to each other when seperated by 20/30 feet with no problem so there is obviously signal leakage within the chassis/plastic carcass they come in? I would like to know how to reduce the power output of the transmitter to just the bare minimum for a signal over say a 30/40 metre range clear line of site?

Many thanks any info appreciated on any of the above points. PS where can u buy a cheap diversity receiver in th UK? in case the above doen't work

Share this post


Link to post
Share on other sites
Is their any way (other than simply using them) to test whats happening beam width and performance wise.

The typical beamwidth for the GPP is about 80 degrees. But keep in mind that antennas do not have brick wall beam patterns. In other words, there is plenty of RF signal further beyond their -3dB beam spread cutoff specs.

... does this matter as it is not a totally flush surface therefore on the reflector? (don't forget I'm not after power simply a 60/80 degree directionality to eliminate multipathing in a partly steel building)

This may affect the dielectric and cause reduced efficiency. I doubt it would be much of a threat to directionality, at least not that would be any concern in your app.

... even with the aerials removed tx and rx still 'talk' to each other when separated by 20/30 feet with no problem so there is obviously signal leakage within the chassis/plastic carcass they come in?

Such a test is only valid if proper 50 ohm RF terminators were installed on the RF ports.

I would like to know how to reduce the power output of the transmitter to just the bare minimum for a signal over say a 30/40 metre range clear line of site?

You would need an attenuator in series with the Tx antenna feedline. These are difficult to build for microwave RF. However, you can emulate one using RG-178 or RG-174 coax. Assume about -2dB attenuation for every meter of coax.

I don't think you will solve the problem merely by reducing RF power. In an interior space the reflective surfaces cause strong microwave multipathing. Even with reduced power, in these sort of environments the unwanted reflections are often strong as the original source.

Besides a diversity Rx, you could also look into circularly polarized antennas. When used correctly, these can offer reasonable immunity to multipathing. However, you have to be careful since some vendors claim circular polarization but really do not achieve it. A good pair (left hand and right hand set) can be expensive.

Share this post


Link to post
Share on other sites

I have had problems with BNC connectors at 2.4Ghz and now only use N type or SMA. I had a very hard to find intermittant loss of signal with BNC.

I think there is a lot of things you can try to cure this problem and you have a good chance of success but it may take a while. Keep us informed its an interesting problem.

Terry

Share this post


Link to post
Share on other sites

Brief update gentlemen on progress (or rather lack of it).

Built the two gp patch antennae raised them to above head height and aligned them at swimming pool but problem (as RC-CAM suspected) still exactly the same. Attenuated signal so when the two antennae were at max distance of 30m picture was just about to dissapear but still no joy. The roof of this building (coupled to waters surface?) really seem just too reflective and condusive to multipathing. The roof is profile steel sheeting in an arch format with steel ties across at about 5m centres (sounds almost like an antenna in its own right!!)

I really dont have enough knowledge I feel to progress much more short of buying a diversity receiver and thats a big gamble a) as I would be risking club's money and I dont have enough confidence or knowledge to know if this has a high success rate.

The alternative circular polarised aerials sound a possibility and I have read up to try and understand them but am unsure how they would be used. Is it one on transmitter one on receiver, or a pair on receiver? Do you have a clockwise on one and anticlockwise on another etc etc. Any further comments would be greatly appreciated as indeed has been the discussion too date.

Kind regards

Share this post


Link to post
Share on other sites

Thats a shame I was banking on more success than that, sorry I was wrong.

It may be one particular wall, roof or whatever causing most of the problem, have you tried moving the patches so they are not pointing directly at each over ?

The patches I guess have too wide a beam for your needs, I have had some luck fitting horn feeds to patches which may help you. I made mine from cardboard covered in tin foil until I was happy with it then made it in thin aluminium sheet. Nothing to hard to do just a square 60deg horn with the small end 11cm where the patch goes.

The helicals are not too hard to make and plans are on the net still Im sure. You need one each as with the patches and both must be wound in the same sense.

If I was closer to Norfolk I would give you a hand but Im in sunny Somerset. Good Luck.

Terry

Share this post


Link to post
Share on other sites

When using circular polarized patches {for multipathing reduction}, both ends need to have one. One side must be RH circular, and the other must be LH circular. It doesn't matter which is which.

Circular polarization does not totally cure multipathing, but will reduce it a lot. And as mentioned, some vendors claim circular polarization, but do not achieve it, so be careful.

Share this post


Link to post
Share on other sites
When using circular polarized patches {for multipathing reduction}, both ends need to have one. One side must be RH circular, and the other must be LH circular. It doesn't matter which is which.

Can you explain why this is different than for helicals ?

Thanks, Terry

Share this post


Link to post
Share on other sites
Can you explain why this is different than for helicals ?

It is the same if the both helicals are fully circular polarized. If the ends are not matched correctly then > -20dB loss will be encountered.

Share this post


Link to post
Share on other sites

I dont agree, maybe Im going mad but Im sure when I tested a home made data link using helicals a few years back we used a RH helix on both ends and it worked very well.

Terry

Share this post


Link to post
Share on other sites

That web page does indeed say that both sides need to be the same. I'm not sure why I was confused; I appreciate the correction!

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×