Jump to content
ThomasScherrer

The RC-CAM PAtch antenna

Recommended Posts

I have constructed the Patch antenna discribed here:

http://www.rc-cam.com/gp_patch.htm

at the bottom of this page you can see closeups of my Patch

http://www.webx.dk/rc/video-wireless/video.htm

Here it is RTN loss sweeped:

Antenna-RTN-Sweep-Patch-original-design.png

Here is original black whip antenna.

Antenna-RTN-Sweep-Whip.png

The Whip has got impressive good RTN loss curve,

The Patch is close to ok, but perform much better than the whip in range test,

I think the Patch can be even better if the poor RTN loss can be fixed,

it is however not so easy to adjust, now it's glued and soldered.

Anyother have got RTN curves on their Patch ?

Share this post


Link to post
Share on other sites

ARGH ! I just realized two of the corners of my patch,

the distance is 0.2mm to small, in the documantation it say 5.0mm

so I will fix this, and make a new curve, I hope this helps.

Share this post


Link to post
Share on other sites

Another measurement cable was found less ripple, here the original unmodified:

path56x56x5org.png

the feed patch is 56x56 and 5mm height from bottom plate.

Then I started to experiment with the distance to ground:

pat56x56x4.png

the feed patch is 56x56 and 4mm height from bottom plate,

ok so less distance improve the impedance, but did not move the resonance frequence enough.

So I changed the size of the patch to see if I could get resonance closer to our video frequency band:

pat55x55x4.png

the feed patch is 55x55 and 4mm height from bottom plate.

The Last try was my lucky one:

pat54x54x4.png

the feed patch is 54x54 and 4mm height from bottom plate.

so the original patch is 2mm to big on both sides, and the height i 1mm to much.

Share this post


Link to post
Share on other sites

Nice data. Thanks.

Can you perform a field test to see how much more practical range the idealized patch antenna gives over the original?

Share this post


Link to post
Share on other sites

I can not, now I dont have the original anymore :-)

but I dont think it will give any huge difference,

kind of hard to prove, from day to day to set up your transmitter one speciel place, see the noise on the screen several km away, then change the antenna, then see if less noise is there ? hum kind of complicated,

that is why I like lab measurements more, if they prove better in lab, they will most likely prove better in real life, but it is easy to change your own and see if you can prove any difference, I promise it will not be worse :-)

You see a 10dB better signal will give you 3 times the range,

this antenna return loss fix will give you 1.6 db better signal,

the same as if you changed your transmitter from 500 to 700mW

Nice extra power for free, just by changing the antenna a few mm :-)

But why not change the drawing so that future antennas made will be in perfect match, at the right frequency band.

Return loss of 5dB equals 1.65dB reflection power loss, equals 68% of the power is used.

Return loss of 12dB equals 0.28dB reflection power loss, equals 94% of the power is used.

Edited by ThomasScherrer

Share this post


Link to post
Share on other sites

I have done a lot of testing of my own design patch so I will share my findings as they will be simular to the GPP.

The GPP as with my patch have both been designed for receiving and I have found that in field tests you have to have a very poor match before you can measure the difference. When connected to the TX its a different story but thats not what we are intrested in.

The GPP has been made by lots of people using lots of different materials and different connectors. I have found that changing the connector type or using say brass instead of aluminium can also drastically change the performance when tested on the bench but still give great performance in the field.

The point I am trying to make is that the Goof Proof Patch is a simple cheap upgrade used for receiving only that often wont be constructed to precision standards so the improvement will not be seen.

Before we worry about the exact size of the element a lot more testing would need to be done using different connectors, materials and in the field to determine what the exact parts should be. Different parts may require a different size resonator. Of coarse it would have to chage its name to the PMP (Precision Made Patch). :rolleyes:

Having said that, it's always good to read the findings of others and maybe a link to your results would be interesting to anyone thinking of building the GPP, maybe somebody will take it further ?

Thanks for posting your info Thomas :)

Terry

Share this post


Link to post
Share on other sites

Elossam:

I used a HP network analyzer at work :-)

Terry:

Yes, as I said 1.6dB is most likely not possible to prove much better in field test,

same as if changing TX power from 500 to 700mW also 1.6dB better, and also most likely not possible to prove much better in field test.

I dont agree on your TX vs RX values,

return loss miss match is the same on a tx as on a rx signal,

bad swr will damage a tx, ok so it may be more important due to that.

but bad swr on a rx antenna will loose signal, and my calculations show the same values.

to mr cam I suggest updating the drawing, and most important is to add measurements on all parts, so they can be verifed much more accurate,

you see testing them on a paper is not the most perfect way,

it you add 54 x 54 mm and also the inch value, more people will get a perfect result, the backplate is 17,4 x 17,4 mm but not at all as critical,

and the distance is uha uha critical :-)

If I had more time, I would be happy to find out the best feed point also,

it is only a few mm off, but I dont know in what dirrection :-)

so there is 10dB more to get on the RTN loss factor, but who cares now, when we transfer 94% of the signal, and not 68% as the previous.

Edited by ThomasScherrer

Share this post


Link to post
Share on other sites

I just had a nice conversation with a Patch antenna designer,

after looking at my page and mr cam page about all this he said:

1) nice result, 10-12dB rtn loss is fine, we normally get 12-18dB

but we also spend much more time.

2) the feeder pin on your (thomas) patch, is a prof SMA make shunner,

the pin is thick and in one piece, this is what I would recommend !

this is important ! if you change this thickness the rtn loss and resonance will change. use 1mm pin

Share this post


Link to post
Share on other sites
I dont agree on your TX vs RX values,

return loss miss match is the same on a tx as on a rx signal,

bad swr will damage a tx, ok so it may be more important due to that.

Yes I agree in theory it should not matter if the patch is on the TX or the RX but when out in the field testing it is much more noticable on the TX. It maybe the way I did the test ? I used a 10mW TX with a 3db att set on a tripod then walked away and marked the point on the ground were the picture started to break up. The difference between my near perfect match patch and the one that was reflecting 1/6 of its power on the bench was hard to see, maybe 5m at a range of 300ish metres. When I swapped round and put the patch on the TX there was an easy 40m difference. I'm not saying my test was perfect but with the equipment I have it was as close to real use as I could get. I have not repeated the test since (this was a few years ago) so maybe there was another factor I missed. If anyone else has done a simular test I would be interested in there results ?

2) the feeder pin on your (thomas) patch, is a prof SMA make shunner,

the pin is thick and in one piece, this is what I would recommend !

this is important ! if you change this thickness the rtn loss and resonance will change. use 1mm pin

That is interesting to know, it may explain why my patches with 'N' type connectors are better than the SMA ones. I first did all my testing to find the sizes using an 'N' type connector and was wondering why the SMA ones were not as good, I had not thought the thickness of the pin would make much difference.

Terry

Share this post


Link to post
Share on other sites
... to mr cam I suggest updating the drawing,

I'll consider that. But, I would want to redo all my testing too. I agree that perfection is the best goal, but from what I can see in the data, the revisions would not be noticed by the builder.

... and most important is to add measurements on all parts, so they can be verified much more accurate

The entire theme of the project was to eliminate numbers and measurements. In effect, the antenna construction has been reduced to a "paper doll" cutout project. Although this makes engineers uncomfortable, since they like to see hard numbers, it has been well received by all other mortals. Frankly, I thought this was a clever way to present the antenna design. :)

... and the distance is uha uha critical :-)

The air dielectric gap sets the bandwidth, which I wanted to be sufficiently wide. The wide gap does reduce gain a little, but makes the antenna system more immune to the effects of Rx input impedance (which will widely vary). Or so the theory goes. Long story short, to change the feedpoint impedance, the location of the SMA should be moved instead. Just my personal opinion.

BTW, could you please post the dimensions to the SMA feedpoint on your modified Patch? This location will be needed if others want to build with your revised dimensions. They will also need the part number to the SMA you used.

Share this post


Link to post
Share on other sites

I did not change the feed point, the patch is made smaller on all 4 sides equaly,

and the corner edges grinded also so they looked exactly the same,

well I dont see why adding measurenents numbers on the drawing could confuse anybody, if the values are there, I am sure people will check they got it right,and the change for a perfect result will be much better.

I have an idea: if you edit your drawing, and mail a checkup version to me, I will be happy to unsolder my patch and verify it fits perfectly on top of the paper,

and to check the solder hole is exactly right.

The SMA connector I use is shunner male large mount with 4 holes,

and long about 17mm the pin is 1.3mm diameter,

the teflon isolating thing went along way up the pin,

I have removed that teflon.

Edited by ThomasScherrer

Share this post


Link to post
Share on other sites

Thomas, I'm considering building a GP patch to complement my current selection of antennaes. I have access to hydraulic shears to cut the metal accurately to the specs you posted on your site, so I'm considering using the connector you used.

You never provided the part number you used, however, I found this one that seems to closely match the information you posted. Is this the same one or suitable?

http://www.amphenolconnex.com/Assets/Produ..._pdf/132144.pdf

Share this post


Link to post
Share on other sites

Hi rob, EXACTLY ! this is the connector you need.

remove the isolation from the center pin stick, all the way to the connector.

there exist an updated drawing:

www.webx.dk/rc/video-wireless/PATCH4.pdf

it is not in final release state yet, but will be in a few days,

so stay tuned :-)

we will not change any measurements, but only how the page is alligned

and how the PDF was created and converted.

Share this post


Link to post
Share on other sites

there exist an updated drawing:

www.webx.dk/rc/video-wireless/PATCH5.pdf

now have a look.

I suggest Mr Cam could take a copy and place online in the RC-CAM projects dir, for all your users.

specially good idea, if my site goes offline or what ever.

Share this post


Link to post
Share on other sites

I have a question about this GP setup... In the old days when we built television antennas, the "teachers" warned us about connecting the copper wire directly to the aluminum dipole, since the copper-aluminum connection would create some kind of static electricity - just because of the metals are different. Now I am not sure how true that was, but we always covered with solder all the exposed copper wire which came in contact with the aluminum dipole.

In the case of the GP modified by Thomas, we have 3 different types of metals - Aluminum for the reflector which has the SMA plug - I think with a gold surface (at least the ones I've got from mouser are gold) - also the copper dipole connected to the golden middle pin.

I understand that the connector cannot be any better than if it is with gold surface, but wouldn't the above problem apply here as well?

Thanks, Ox.

Edited by oxxyfx

Share this post


Link to post
Share on other sites

I've never heard of that personally oxxyfx. Nor can i really see, chemically, how that is possible. I'm sure someone else here has far more experience than me though...

Thomas, your avatar is really trippy after practicing 3d heli moves on a sim. dunno why, making my head spin hahaha.

Share this post


Link to post
Share on other sites

different materials used in outdoor antenna installations is a know problem.

Take two different metals and stick them in a cup of water,

and measure the voltage, ok you see !

now if the two different metarials touch, and get wet everytime it rains,

you will have a tiny current eating up the metarial, years later, problems.

Either avoid different metals or avoid water.

if water consist of a salty mixture it will be much worse,

like if you live near the sea.

I just stay home when it rains so I am happy.

Share this post


Link to post
Share on other sites

Thomas,

How about 70-95% humidity - which is not uncommon in the east cost of the US and Canada? Does that count? Should one stay home those days?

Thanks, Ox.

Share this post


Link to post
Share on other sites
I'll consider that. But, I would want to redo all my testing too. I agree that perfection is the best goal, but from what I can see in the data, the revisions would not be noticed by the builder.

The entire theme of the project was to eliminate numbers and measurements. In effect, the antenna construction has been reduced to a "paper doll" cutout project. Although this makes engineers uncomfortable, since they like to see hard numbers, it has been well received by all other mortals. Frankly, I thought this was a clever way to present the antenna design. :)

I just stumbled over this post as I was looking to make a patch antenna. Thomas has done all of this research and published it to boot. I can't really see why there is so much resistance to incorporate his findings and improve things. Isn't that what open-source and community projects are all about?

As for the suggestion of displaying measurements on a plan; I hardly think a few XXmm here and there are going to confuse anyone. If someone can be baffled by that then they really shouldn't be flying RC aircraft at all!

Just my 2 cents.

Share this post


Link to post
Share on other sites
Thomas has done all of this research and published it to boot. I can't really see why there is so much resistance to incorporate his findings and improve things. Isn't that what open-source and community projects are all about?

I guess our antenna designs will remain different. I have enough confidence in the rc-cam GPP project to know it will not offer disappointing results. The beauty of it all is now builders have two choices. :)

As for the suggestion of displaying measurements on a plan; I hardly think a few XXmm here and there are going to confuse anyone. If someone can be baffled by that then they really shouldn't be flying RC aircraft at all!

The basic premise behind the GPP is that it is built without having to measure anything. It's perhaps the first antenna project to do such a thing. Given the endless number of success stories, it seems to have worked out well.

Frankly, the Internet is full of patch designs that have dimensions. It's nice to be different. :)

Share this post


Link to post
Share on other sites

OK guys, I know this is slightly off topic, but I figured it fits in with some of the discussion here. I was looking at this page:

http://www.cst.com/Content/Applications/Ar...VE+STUDIO%C2%AE

considering something similar to a GP patch design for my 900MHz system (Actual 908MHz for Ch1). Note: I've already built a scaled up version of the GP Patch antenna and it's working pretty good....but it's HUGE!.

What I like about the antenna on the page I listed is that it claims to be circularly polarized.

They show two versions, one that uses "expanded polystyrene" (styrofoam?) between the antenna patch and the ground. The other uses a type of Nylon. I have no idea what the dielectric constant is of the styrofoam, and I can't seem to find the exact same nylon with the same dielectric constant.

Anyone with better tools care to take a look at this design and modify it for easy to obtain (read off-the-shelf) materials? Really it's just the dielectric material that needs to be chosen. I like the idea of the smaller patch, but based on what they show it has a narrower frequency band. That's fine if it matches the frequency I'm using!

Share this post


Link to post
Share on other sites
What I like about the antenna on the page I listed is that it claims to be circularly polarized.

I could be wrong, but I doubt the cst design is circularly polarized (at least not enough to help out). I'm a bit tainted since I tried the "corner truncations" trick and it did not appear to be effective on the GPP. I never say never, but I would be surprised if the same technique worked on cst's design.

I like the idea of the smaller patch, but based on what they show it has a narrower frequency band.

The dielectric distance affects the bandwidth. The chosen dielectric material does indeed allow the patch to be reduced in size. But at a cost -- the smaller antenna will have proportionally less gain. The reason the GPP uses an air dielectric is so that the antenna elements can be as large as possible for higher gain.

Lastly, the cst antenna is optimized for 50 ohms. From what I have seen, the typical wireless A/V receiver designs are mostly 75 ohms (not all, but most). If you do build the cst patch, also try one optimized for this impedance to see if you like it better. Impedance is controlled by moving the patch's feedpoint (don't change the dielectric distance to do that).

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

×