Discuss EVDO PC Cards, ExpressCards, EVDO USB, PDAs, Phones, Coverage and Mobile Broadband Cards. Discuss Verizon and Sprint Coverage. CradlePoint CTR350 / PHS300/ MBR1000 & Kyocera KR1 / KR2 & LinkSys WRT54G3G-ST Forums!
EVDOforums.com
Discussion forum for EVDO users
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Our sites Include: EVDO Info :: EVDO Forums :: EVDO Maps :: EVDO Blog :: 3Gstore.com

To purchase your EVDO Card / Antenna / Amplifier / Router from the EVDO Experts, just contact us!

Making a corner reflector for Wilson Trucker omni?

 
Post new topic   Reply to topic    EVDOforums.com Forum Index -> EVDO Signal Enhancers with Antennas & Amplifiers
View previous topic :: View next topic  
Author Message
thebordella
EVDO User


Joined: 12 Mar 2007
Posts: 26

PostPosted: Thu Oct 04, 2007 6:30 pm    Post subject: Making a corner reflector for Wilson Trucker omni? Reply with quote

I am experimenting with (cheap) ideas to squeeze out every dB I can get in a fringe coverage area.

Currently using a Wilson Trucker omni on the roof. Soon going to raise it for more elevation. I'm intrigued by the DIY corner reflector described in another post:

http://www.evdoforums.com/thread5179.html

Suppose I want to place the Wilson omni in the PCS focal point of the corner reflector. Which part of the omni is the "hot" part? This antenna has a 'whip' at the top, with a bulbous thing partway down (load coil?), and then more 'whip' antenna, below which are several metal spikes which supposedly act as a ground plain.

Where would you orient the Wilson omni inside the corner reflector construct? Does this plan even make any sense?

I can't/won't place the USB720 itself in the corner reflector, as proposed in the original post, because this needs to be mounted outdoors.

thanks!
Back to top
View user's profile Send private message
n6gn
EVDO Junkie


Joined: 22 Aug 2006
Posts: 347
Location: Northern California

PostPosted: Thu Oct 04, 2007 10:58 pm    Post subject: Reply with quote

Try placing the length of the omni, both above and below the "thing", along the line where you would have placed the dipole or U720, had you used that instead. This means that you want the the whole thing spaced out from the corner about 2" at PCS and about 4.5" for 850 MHz. The presence of the corner (correctly aimed) should increase the overall gain of the omni by 6-8 dB. You'll want to be sure that the overall height of the corner is at least 2-4" greater than the length of the omni to get this improvement.

n6gn
Back to top
View user's profile Send private message
118morpheus
EVDO Newbie


Joined: 05 Jan 2008
Posts: 4

PostPosted: Sat Jan 05, 2008 10:29 pm    Post subject: Corner Reflector for wilson trucker - need details please! Reply with quote

I wondering how the reflector worked for you if you tried it out. I have a few q's for an expert. I noticed a distance of 2" from the corner was mentioned. Where is that derived from (why 2").

I've done some research on corner reflectors and for a 1920Mhz corner reflector. The research i've come up with says that the best position for the element is .6 wavelength from each panel. If a 1920Mhz wave is 156.25mm then a .6 wave is 93.75mm. The distance from the corner is about .85 Wave or 132.56mm...much farther than 2". These numbers were calculated based on specs for a "3D corner reflector" found here:

yu1aw.ba-karlsruhe.de/shorten3dcra.pdf

I don't know if the difference is that there is a bottom reflector???

I have a wilson trucker as well and want to make it a little more directional with some signal gain. I was going to put the element for the trucker at the distances I calculated from the pdf file above, but am wondering if the 2" would work just as well.

Any guidance would be helpful. Or even refer me to where I can find some measurements for a corner reflector in the PC band 1850-1990Mhz

thanks
Back to top
View user's profile Send private message
rlw
EVDO User


Joined: 29 May 2007
Posts: 37
Location: Near Lancaster, OH

PostPosted: Sat Jan 05, 2008 11:40 pm    Post subject: Reply with quote

Take a look at http://www.evdoforums.com/viewtopic.php?p=39708#39708 -- I made my 12 db parabolic reflector based on the designs at http://www.freeantennas.com. I'm also in a fringe area -- I was getting 400-500kbps downloads before I used the reflector, now I get 800-1200kbps and 300-400kbps up.

Be sure to use some kind of ground plane (I use an old aluminum modem case).

Hope this helps,

RLW
Back to top
View user's profile Send private message
n6gn
EVDO Junkie


Joined: 22 Aug 2006
Posts: 347
Location: Northern California

PostPosted: Sun Jan 06, 2008 12:09 am    Post subject: Re: Corner Reflector for wilson trucker - need details pleas Reply with quote

118morpheus wrote:
I wondering how the reflector worked for you if you tried it out. I have a few q's for an expert. I noticed a distance of 2" from the corner was mentioned. Where is that derived from (why 2").

I've done some research on corner reflectors and for a 1920Mhz corner reflector. The research i've come up with says that the best position for the element is .6 wavelength from each panel.
...
yu1aw.ba-karlsruhe.de/shorten3dcra.pdf

I don't know if the difference is that there is a bottom reflector???
...

Any guidance would be helpful. Or even refer me to where I can find some measurements for a corner reflector in the PC band 1850-1990Mhz



Nice article by yu1aw, I hadn't seen it before. Thanks for pointing to it.

The 3D reflector he describes is considerably different from a conventional (2D) corner reflector. By using a much longer element (3/4 wave) and a passive element (mainly to modify the match) he ends up with both more gain and also a main lobe that is at an angle quite different from the 2D version.

So yes, in this case, the bottom makes a great deal of difference as compared to a quarter wave monopole using the bottom only as a ground plane to reflect it's image, and thereby create a dipole.

For the simpler case, which you can find described many places but Kraus' <Antennas> is a good one, the position of high gain is pretty broad and closer to the reflector than for the 3D corner.

I built a pair of corner reflectors with dipole and carefully measured them with a vector network analyzer. Each worked as Kraus suggests and has about 10 dBi gain. I also replaced the dipole with a U720 modem/s antenna and a Treo 700 PDA in the 2" offset position and measured gain increase over the modem/PDA alone. This also agreed well with Kraus.

My expectation is that if you put an omni antenna (like the Wilson, I think, but I've never examined one) in front of a corner, you will probably find that maximum gain improvement is on the order of 6 dB and will occur when the antenna is about 2" away from the corner, but that it will not be terribly sensitive to this position. See Kraus or any other good antenna reference book to get a plot of gain vs. spacing.

A difference between the 3D corner antenna and the one you would be creating above is in the feed. The 3D antenna uses a 3/4 wavelength element without any phasing. On the other hand, the omni would be expected to be similar to a monopole phased with a dipole above it. The peak lobe for the .75 lambda antenna doesn't occur at right angles to the element while the peak for the omni should. Thus when you put these antennas with different patterns in front of a corner reflector, whether 2D like the simplest Kraus or like the 3D one you referenced, you will get very different patterns and different gains.

I don't doubt that the 3D antenna can have considerably more gain than the 2D corner, which for a 90 degree corner maxes out at a bit over 10 dBi for a dipole feed. However the 3D corner is more complex to construct and will have to be bigger to achieve this increase in gain.

I might add that if you are interested in building antennas that deliver a lot of gain, you may find that a parabolic antenna or a parabolic approximation like a cone are actually easier to construct and likely provide more gain for a given antenna size.

n6gn
Back to top
View user's profile Send private message
118morpheus
EVDO Newbie


Joined: 05 Jan 2008
Posts: 4

PostPosted: Sun Jan 06, 2008 12:37 pm    Post subject: Reply with quote

Excellent response! thanks for the info. I figured the 3D had a difference and noticed that the direction of the signal radiation was about 45 degree from the bottom plane.

Do you suggest any place for DIY cone antenna. I am working on building a bi-quad with 1920 specs and may try to use it as a feed with a dishnetwork antenna or just alone. But It seems that the bi-quad is a little picky with dimesions and building. We'll see how it goes.

Do you know why 2" is the sweet spot on a 2D corner reflector. Is a a percentage of wavelength or some other variable?

Thanks again,

Jim
Back to top
View user's profile Send private message
n6gn
EVDO Junkie


Joined: 22 Aug 2006
Posts: 347
Location: Northern California

PostPosted: Sun Jan 06, 2008 2:56 pm    Post subject: Reply with quote

118morpheus wrote:

Do you suggest any place for DIY cone antenna. I am working on building a bi-quad with 1920 specs and may try to use it as a feed with a dishnetwork antenna or just alone. But It seems that the bi-quad is a little picky with dimesions and building. We'll see how it goes.

Do you know why 2" is the sweet spot on a 2D corner reflector. Is a a percentage of wavelength or some other variable?


The best reference I know for a DIY conical-approximation-of-a-parabola was in a very old amateur radio VHF magazine. It was for about 1.3 GHz and was just a 32" circle of rabbit wire with a pie-section cut out. The cut edges were then pulled together to create a cone. The feed was just a dipole in front of a passive reflector mounted at the focal point. The size of the pie slice that was removed was picked to make an F/D ratio of something on the order of 0.5 which the dipole/ +reflector fed moderately well.

I've thought of building up something similar using a bi-conical/dipole feed so that both 850 MHz and PCS would be covered with a single antenna but haven't done it yet. I think it is possible to get 12 dBi on 850 MHz and 18 dBi on PCS that way. The hardest part for many people is going to be in building and connectoring/cabling the dipole. I haven't yet figured out how to make that easy to do well in the average garage and without soldering equipment.

The 2" (for PCS) does come from the dimensions of lambda ( wavelength = ~ 11.8/Frequency in GHz). As I said, the distance is quite uncritical. Page 331 of John Kraus' <Antennas>, Edition 1 McGraw-Hill, NY 1950, shows that for a 90 degree corner that the max gain is about 10.3 dBi but that the gain is above 10 dBi from .2 lambda to .4 lambda. At PCS 2inches is about .32 labmbda. You can go out to .7 lambda without the gain dropping below 8 dBi. It's really quite uncritical.

n6gn
Back to top
View user's profile Send private message
118morpheus
EVDO Newbie


Joined: 05 Jan 2008
Posts: 4

PostPosted: Sun Jan 06, 2008 4:02 pm    Post subject: Reply with quote

Thank you. I finally found someone with an understandable answer!

Jim
Back to top
View user's profile Send private message
Jayson
EVDO Heavy User


Joined: 17 Feb 2007
Posts: 101
Location: ohio , in the sticks

PostPosted: Sun Jan 06, 2008 6:43 pm    Post subject: Reply with quote

118 I would like to see this project if you get it finished up ....I to live in an area like you speak of.... always looking to better my downloads....normally there around 500 down and only about 100 up.....thought of buying a grid antenna maybee in the near future to see what i can get outta that !!!!! hope to see your results soon !!!! and good luck with it ...... Smile
Back to top
View user's profile Send private message
118morpheus
EVDO Newbie


Joined: 05 Jan 2008
Posts: 4

PostPosted: Sun Jan 06, 2008 8:16 pm    Post subject: Reply with quote

Just a little info about my connection. For the past month since signing up for sprint evdo using a pantech px500 and Kyocera kr1 I had been getting 400-900k down and 150-170k up.

My signal with card in my laptop is -89 to -87 and pings from 85 to 175ms (prett good) with a Wilson trucker on roof, and 30 feet of RG8 ultra low loss (heavy stuff). Signal is going through tree tops to tower about less than 2.5 miles away. My router's signal meter is next to useless. It fluctuates from 3 of 10 bars to 7 of 10 bars even though when the card is in my laptop the signal is steady according to Sprint Connection Manager.

The past three days I have had erratic pings and speeds from 150-700k down and 80k to 170k up with pings 99-350ms. I have also had several unexpected disconnects where I have had to restart the router and reconnect.

I got a CM2000 3/2 watt dual band amp 2 days ago, it doesn't give me any gain increase...still -89 to -87.

Today 1-6-07 I put a parabolic reflector on my wilson trucker. I didn't check the db gain in my laptop using connection manager, but I got a more steady 8 of 10 bars on the router signal meter, with drops to 3 on occasion (don't know why?). Before reflector I got 6 of 10 bars with drops to 3.

Speed tests are still about the same as the erratic speeds and pings I had been having over the past few days (sprint says all is ok on their end).

Another odd thing I have going on is my pantech says I am connected to rev A when looking in the diagnostics window of the connection manager, but I was told by a local Sprint employee that my area has not yet upgraded to Rev A? My uploads lead me to believe I am not rev A.

In my case signal increase does not appear to equate to faster speeds.

Sorry about all the boring info. I just am wondering why my quality of service took a dive and wish could get an upload and download speeds to be like some others 1300 by 400+.

Since this is a thread about reflectors:

The parabolic reflector I made was of 1/4" wire mesh the bottom rests on the "ground plane" (metal rods) at the base of the trucker antenna. The reflector is set for 1.6" which is the focal point for my particular parabolic antenna. It seems to work, but has not changed my speeds any.
Back to top
View user's profile Send private message
xrayman
EVDO Junkie


Joined: 23 Feb 2007
Posts: 259
Location: Kansas City

PostPosted: Sun Jan 06, 2008 11:59 pm    Post subject: Reply with quote

n6gn wrote:
I've thought of building up something similar using a bi-conical/dipole feed so that both 850 MHz and PCS would be covered with a single antenna but haven't done it yet. I think it is possible to get 12 dBi on 850 MHz and 18 dBi on PCS that way. The hardest part for many people is going to be in building and connectoring/cabling the dipole. I haven't yet figured out how to make that easy to do well in the average garage and without soldering equipment. n6gn
Sounds like a great project, I would love to see you build one and post the how to build instructions. I understand what you are saying about DIY soldering. If your project didn't require a solder bath or ovens but only simple Rat Shack equipment, you could post a short 101 soldering tutorial. It would be a great time for people to learn.
Over the years I have installed hundreds of various types of antennas but I have very limited knowledge about antenna engineering. Your postings about antenna design has been very educational.
I have seen in postings on forums that some have used a large metal coffee can to build an USB EVDO device antenna. They place the EVDO device inside the can and say it will greatly increase the signal strength. Could you give an easy to understand explanation of how this device works and what gain could be expected? I understand the concept of the conical, parabola, 3D and 2D reflector but the coffee can design I need help with. If it works like a wave guide I would think the size of the can should be critical. This is a little off the thread title, post to a different thread if needed.
Back to top
View user's profile Send private message
n6gn
EVDO Junkie


Joined: 22 Aug 2006
Posts: 347
Location: Northern California

PostPosted: Mon Jan 07, 2008 1:45 am    Post subject: Reply with quote

xrayman wrote:

...
I have seen in postings on forums that some have used a large metal coffee can to build an USB EVDO device antenna. They place the EVDO device inside the can and say it will greatly increase the signal strength. Could you give an easy to understand explanation of how this device works and what gain could be expected?

I've been thinking about this for about an hour now and at least for the first part of your question I think the answer may be "I can't" ! However, fools rush in so here goes...

The only way I can think to make it easy to understand is to describe something it is *somewhat* like. This means that my explanation is going to be wrong - it will fall apart for many aspects of understanding. However, I like an old saying I heard years ago "All models are wrong, some are useful" so let me try this one ...

Think of an antenna as a "wave converter"; as a device that goes between waves in free space and turns the wave energy into useful voltage and current in some conductors somewhere.

This of course brings up the question of "what is a wave?". For that, consider a wave you can see - waves on the surface of water. When you throwa big rock into a previously still lake it gets "disturbed". That disturbance moves (propagates) from the point where it started outward, as an expanding circle. The falling rock hitting the water contributed energy to the lake which starts a propagating disturbance or "wave". If you have a fishing line with a bobber on it in the lake, when the wave goes by it, the energy from the wave moves the bobber up and down but the bobber and the local water itself doesn't move other than up and down. Once the wave is past the bobber is as it was. When the wave hits the shore, it might be possible to recover some of the initial energy by letting the wave move some driftwood, sand or something. If that happens, energy of the falling rock would end up doing work on some remote object at lake's edge. So the wave might be thought of as a disturbance in a medium which can transfer energy across a distance.

Antennas couple energy of electro/magnetic waves. In fact, they do it in both directions; an incoming wave can be converted to voltage/current in a conductor or voltages and currents within the antenna's conductors can *generate* waves. The antenna is an adapter of sorts between electrical energy in a conductor and wave energy propagating in a medium.

There's a parallel between antennas and loudspeakers. If you hook a 100 watt audio amplifier 'transmitter' to your subwoofer, the cone in the speaker moves, generating a disturbance which can propagate across a medium (air) in the form of (pressure) waves. Similarly, you could use your subwoofer on 'receive' as a microphone to convert pressure waves generated elsewhere into electrical voltage and current in a resistor or amplifier.

The cylindrical (Pringles?) can works as an antenna containing a wave guide. A wave guide is a structure that bounds, directs or contains a moving wave (normally keeping it from expanding as it would if the guide were not present).

For receive, the incoming wave gets "caught" by the end of the circular wave guide, travels along it according to the terms of the propagating mode in that guide and is eventually coupled to a conductor, causing currents and voltages which can be applied to a receiver. It turns out that there is a fundamental rule (Lawrence reciprocity theorem) which says (except for some pretty special cases) an antenna works equally on transmit and receive. Currents/voltages at one end produce waves at the other and if those same waves impinge on the antenna the result will be the original currents/voltages.

The gain you see for an antenna relates to how much of the incoming wave it "catches" and turns into useful current/voltage. Interestingly, it isn't possible to build an antenna (out of perfect conductors, anyway) that doesn't effectively catch an area almost as big as a dipole does. In fact even the theoretical and minimum gain 'isotropic' antenna (the 'i' in dBi) which is an antenna with zero directivity, catches an area about as big as 1/12 of a wavelength squared. At PCS, this amounts to about three square inches.

Thus, if you had a (10*3 sq in =) 30 square inch antenna that perfectly caught an incoming PCS wave and converted all of the energy that it intercepted into usable voltage/current, it could catch 10 times the amount of power of an isotropic antenna and would then be said to have 10 *log(10)=10 dBi gain. If it were 300 square inches it would have 10*log(100)=20 dBi gain.

Of course the reality is that the best of antennas does not perfectly convert the wave to voltage/current. For large antennas/feeds like in big radio telescopes, the really good ones might have 75-80% aperture efficiency - which is a measure of how good a job they do. Typical parabolas may be only 55% efficient and I suspect that the 'Pringles' variety are worse than this, though I haven't measured one.

Putting rough numbers on all of this, if you had 4" diameter cylinder waveguide that could be made to have 55% aperture efficiency you would have a 'catcher'

area pi*r^2= pi*(4/2)^2 = 12.6 square inches
and effectively
(55% of 12.6)/3= about 2.3 times as big as an isotropic antenna. It's gain would then be
10*log(2.3) = 3.6 dBi at PCS. If it was twice that diameter, 8", it would have 4 times the area and 6 dB more gain or about 9.6 dBi.

I'll now repeat my caution that the above model is wrong. Maybe someone else can correct and/or improve on it for us. Still, I hope it is useful.

n6gn
Back to top
View user's profile Send private message
xrayman
EVDO Junkie


Joined: 23 Feb 2007
Posts: 259
Location: Kansas City

PostPosted: Mon Jan 07, 2008 11:09 am    Post subject: Reply with quote

n6gn,
Thanks for the elegant technical reply.
Back to top
View user's profile Send private message
matt10099
EVDO Fledgling


Joined: 09 Dec 2007
Posts: 15

PostPosted: Tue Jan 08, 2008 12:40 pm    Post subject: Reply with quote

I also live in a fringe area with trees.. I have a 7db gain 18" antenna, which I made a corner reflector with dimesions of approx. 20"H x 17"L, with plywood surfaces, then covered in aluminum foil for the reflector

I now get very good to excellent signal or 4-5 bars (-77--81db)

Getting that signal was very hard it took 2 people & about 1-2 hrs of adjustments to get right... there was little room for error & 1/8" meant the difference between a good signal or low signal.

Very cost effective solution if you already have these supplies & 2 people & couple hours.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    EVDOforums.com Forum Index -> EVDO Signal Enhancers with Antennas & Amplifiers All times are GMT - 6 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum








Sprint Rev A USB: Compass 597



Sprint Rev A ExpressCard: Merlin EX720


Purchase an V740 Rev A ExpressCard


Sprint U727


CTR350 Router


CradlePoint PHS300


Purchase a MBR1000


Purchase an EVDO Booster Antenna

Purchase a LinkSys 3G Router



Purchase an EVDO Amplifier


Your Mac EVDO Experts









EVDO Antenna Booster





Digg Us :: del.icio.us :: technorati :: furl

4G :: 4G Forums :: PHS300 :: MBR1000 :: KR2