[EE] Antenna design?

[Neil]

Okay antenna gurus, I can use some help here please...

I'm using a 5.8Ghz video transmitter/receiver system on a robot (2D
driving, indoor). These systems are popular with the drone-racing
crowd, but most of the good antennas are either circular-polarized (so
they are uniformly effective at most angles) or patch antennas which are
apparently very unidirectional. Linear antennas have mostly gone away
for this purpose.

I'm trying to maximize signal strength/reception so the operator can be
in a different room and on a different floor. For my purposes, I'm
thinking a linear antenna on the transmitter side (robot) may be best
though, as I can mount it vertically in the robot body, and perhaps use
a patch antenna on the receiver side as that would have minimal
movement. Or linear as well for the receiver. I understand that whip
antennas are really crappy so would a straight piece of wire work
better? I also discovered "collinear antennas" while searching.

I have about 18-20" of room to put an antenna vertically inside the
robot ... would having an antenna length of a multiple of the wavelength
work better than just a single-wavelength antenna?

So what say ye?

Cheers,
-Neil.

[Jason White]

To clarify you are looking for a 5.8GHz directional "high gain" antenna?

The problem with directional (ie. not whip or patch) antennas on
vehicles/robots is that they are directional. If the vehicle turns 90
degrees to the left then the radio beam will also be pointed 90 degrees to
the left. Unless your system is significantly different than I imagine I
would advise against a highly directional antenna.

If you are tying to transmit video indoors I would recommend boosting the
transmit power - and staying with your omnidirectional whip/patch antenna.
If you have a HAM license then you can probably do that legally (depending
on where you are).

Just my 2¢
-Jason White

Post by Neil
Okay antenna gurus, I can use some help here please...
I'm using a 5.8Ghz video transmitter/receiver system on a robot (2D
driving, indoor). These systems are popular with the drone-racing
crowd, but most of the good antennas are either circular-polarized (so
they are uniformly effective at most angles) or patch antennas which are
apparently very unidirectional. Linear antennas have mostly gone away
for this purpose.
I'm trying to maximize signal strength/reception so the operator can be
in a different room and on a different floor. For my purposes, I'm
thinking a linear antenna on the transmitter side (robot) may be best
though, as I can mount it vertically in the robot body, and perhaps use
a patch antenna on the receiver side as that would have minimal
movement. Or linear as well for the receiver. I understand that whip
antennas are really crappy so would a straight piece of wire work
better? I also discovered "collinear antennas" while searching.
I have about 18-20" of room to put an antenna vertically inside the
robot ... would having an antenna length of a multiple of the wavelength
work better than just a single-wavelength antenna?
So what say ye?
Cheers,
-Neil.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Van Horn, David]

There are many things that work a bit. Working well is more difficult.
Gain comes at a cost, you can trade bandwidth or directionality for it. It never comes for free.
You can trade directionality for gain with a colinear antenna, but the more gain you get, the more sensitive it is to antenna tilt taking the receiver out of the beam.
With a colinear, it's like an omnidirectional search light. No energy straight up or down, very little at 45 degrees, and lots at 0 degrees.
Return loss is a really bad metric of antenna performance. A 50 ohm resistor has great return loss (in a 50 ohm system) but totally sucks as an antenna.

Remcom's XFDTD would be my choice to go after this problem, but it is expensive.

Detuning by nearby structures is a major problem here, XFDTD allows you to model those structures and even use them as part of your antenna.
See their study on holding your car remote at your chin. 😊

An inherently broadband antenna may have low gain, but it won't suffer so much from detuning.

http://www.wa5vjb.com/references/PlanarDiskAntennas.pdf
I did something very similar for 2.4 GHz using XFDTD a couple years ago.



-----Original Message-----
From: piclist-***@mit.edu <piclist-***@mit.edu> On Behalf Of Jason White
Sent: Wednesday, July 11, 2018 12:57 PM
To: Microcontroller discussion list - Public. <***@mit.edu>
Subject: Re: [EE] Antenna design?

To clarify you are looking for a 5.8GHz directional "high gain" antenna?

The problem with directional (ie. not whip or patch) antennas on vehicles/robots is that they are directional. If the vehicle turns 90 degrees to the left then the radio beam will also be pointed 90 degrees to the left. Unless your system is significantly different than I imagine I would advise against a highly directional antenna.

If you are tying to transmit video indoors I would recommend boosting the transmit power - and staying with your omnidirectional whip/patch antenna.
If you have a HAM license then you can probably do that legally (depending on where you are).

Just my 2¢
-Jason White

Post by Neil
Okay antenna gurus, I can use some help here please...
I'm using a 5.8Ghz video transmitter/receiver system on a robot (2D
driving, indoor). These systems are popular with the drone-racing
crowd, but most of the good antennas are either circular-polarized (so
they are uniformly effective at most angles) or patch antennas which
are apparently very unidirectional. Linear antennas have mostly gone
away for this purpose.
I'm trying to maximize signal strength/reception so the operator can
be in a different room and on a different floor. For my purposes, I'm
thinking a linear antenna on the transmitter side (robot) may be best
though, as I can mount it vertically in the robot body, and perhaps
use a patch antenna on the receiver side as that would have minimal
movement. Or linear as well for the receiver. I understand that whip
antennas are really crappy so would a straight piece of wire work
better? I also discovered "collinear antennas" while searching.
I have about 18-20" of room to put an antenna vertically inside the
robot ... would having an antenna length of a multiple of the
wavelength work better than just a single-wavelength antenna?
So what say ye?
Cheers,
-Neil.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

--
Jason White
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist

[Sean Breheny]

I would recommend reading appendix A of my master's thesis :)

https://ecommons.cornell.edu/bitstream/handle/1813/8346/thesis%20formatted.pdf?sequence=1&isAllowed=y

It is an 11-page summary of antenna theory (that I had to include because
my thesis advisor was not an antenna expert)

With an efficient passive antenna (no amplifiers and no major losses that
you can reduce), you can't get more gain without being more directive. In
other words, if you cannot be directional, then making your antenna bigger
than a half wavelength does nothing for you. This is true for BOTH transmit
AND receive because passive antennas are reciprocal devices.

A patch antenna is probably best for you. Any kind of linear antenna is
going to have a null along the long axis of the antenna. This can be a
problem for the case where the robot is on one floor and the operator is on
another floor.

Colinear arrays only make this worse - they direct more radiation in a flat
disk and have an even wider null along the long axis. They are commonly
used for cell towers because all of the cell phones that need to access the
tower are in a narrow range of vertical angle - radiation going up into the
sky (or listening for radiation from the sky) is a waste.

Sean

Post by Neil
Okay antenna gurus, I can use some help here please...
I'm using a 5.8Ghz video transmitter/receiver system on a robot (2D
driving, indoor). These systems are popular with the drone-racing
crowd, but most of the good antennas are either circular-polarized (so
they are uniformly effective at most angles) or patch antennas which are
apparently very unidirectional. Linear antennas have mostly gone away
for this purpose.
I'm trying to maximize signal strength/reception so the operator can be
in a different room and on a different floor. For my purposes, I'm
thinking a linear antenna on the transmitter side (robot) may be best
though, as I can mount it vertically in the robot body, and perhaps use
a patch antenna on the receiver side as that would have minimal
movement. Or linear as well for the receiver. I understand that whip
antennas are really crappy so would a straight piece of wire work
better? I also discovered "collinear antennas" while searching.
I have about 18-20" of room to put an antenna vertically inside the
robot ... would having an antenna length of a multiple of the wavelength
work better than just a single-wavelength antenna?
So what say ye?
Cheers,
-Neil.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Van Horn, David]

Very nice!

-----Original Message-----
From: piclist-***@mit.edu <piclist-***@mit.edu> On Behalf Of Sean Breheny
Sent: Wednesday, July 11, 2018 1:56 PM
To: Microcontroller discussion list - Public. <***@mit.edu>
Subject: Re: [EE] Antenna design?

I would recommend reading appendix A of my master's thesis :)

https://ecommons.cornell.edu/bitstream/handle/1813/8346/thesis%20formatted.pdf?sequence=1&isAllowed=y

It is an 11-page summary of antenna theory (that I had to include because my thesis advisor was not an antenna expert)

With an efficient passive antenna (no amplifiers and no major losses that you can reduce), you can't get more gain without being more directive. In other words, if you cannot be directional, then making your antenna bigger than a half wavelength does nothing for you. This is true for BOTH transmit AND receive because passive antennas are reciprocal devices.

A patch antenna is probably best for you. Any kind of linear antenna is going to have a null along the long axis of the antenna. This can be a problem for the case where the robot is on one floor and the operator is on another floor.

Colinear arrays only make this worse - they direct more radiation in a flat disk and have an even wider null along the long axis. They are commonly used for cell towers because all of the cell phones that need to access the tower are in a narrow range of vertical angle - radiation going up into the sky (or listening for radiation from the sky) is a waste.

Sean

Post by Neil
Okay antenna gurus, I can use some help here please...
I'm using a 5.8Ghz video transmitter/receiver system on a robot (2D
driving, indoor). These systems are popular with the drone-racing
crowd, but most of the good antennas are either circular-polarized (so
they are uniformly effective at most angles) or patch antennas which
are apparently very unidirectional. Linear antennas have mostly gone
away for this purpose.
I'm trying to maximize signal strength/reception so the operator can
be in a different room and on a different floor. For my purposes, I'm
thinking a linear antenna on the transmitter side (robot) may be best
though, as I can mount it vertically in the robot body, and perhaps
use a patch antenna on the receiver side as that would have minimal
movement. Or linear as well for the receiver. I understand that whip
antennas are really crappy so would a straight piece of wire work
better? I also discovered "collinear antennas" while searching.
I have about 18-20" of room to put an antenna vertically inside the
robot ... would having an antenna length of a multiple of the
wavelength work better than just a single-wavelength antenna?
So what say ye?
Cheers,
-Neil.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist

[Sean Breheny]

Thanks!

On Wed, Jul 11, 2018 at 4:05 PM, Van Horn, David <

Post by Van Horn, David
Very nice!
-----Original Message-----
Sent: Wednesday, July 11, 2018 1:56 PM
Subject: Re: [EE] Antenna design?
I would recommend reading appendix A of my master's thesis :)
https://ecommons.cornell.edu/bitstream/handle/1813/8346/
thesis%20formatted.pdf?sequence=1&isAllowed=y
It is an 11-page summary of antenna theory (that I had to include because
my thesis advisor was not an antenna expert)
With an efficient passive antenna (no amplifiers and no major losses that
you can reduce), you can't get more gain without being more directive. In
other words, if you cannot be directional, then making your antenna bigger
than a half wavelength does nothing for you. This is true for BOTH transmit
AND receive because passive antennas are reciprocal devices.
A patch antenna is probably best for you. Any kind of linear antenna is
going to have a null along the long axis of the antenna. This can be a
problem for the case where the robot is on one floor and the operator is on
another floor.
Colinear arrays only make this worse - they direct more radiation in a
flat disk and have an even wider null along the long axis. They are
commonly used for cell towers because all of the cell phones that need to
access the tower are in a narrow range of vertical angle - radiation going
up into the sky (or listening for radiation from the sky) is a waste.
Sean

Post by Neil
Okay antenna gurus, I can use some help here please...
I'm using a 5.8Ghz video transmitter/receiver system on a robot (2D
driving, indoor). These systems are popular with the drone-racing
crowd, but most of the good antennas are either circular-polarized (so
they are uniformly effective at most angles) or patch antennas which
are apparently very unidirectional. Linear antennas have mostly gone
away for this purpose.
I'm trying to maximize signal strength/reception so the operator can
be in a different room and on a different floor. For my purposes, I'm
thinking a linear antenna on the transmitter side (robot) may be best
though, as I can mount it vertically in the robot body, and perhaps
use a patch antenna on the receiver side as that would have minimal
movement. Or linear as well for the receiver. I understand that whip
antennas are really crappy so would a straight piece of wire work
better? I also discovered "collinear antennas" while searching.
I have about 18-20" of room to put an antenna vertically inside the
robot ... would having an antenna length of a multiple of the
wavelength work better than just a single-wavelength antenna?
So what say ye?
Cheers,
-Neil.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at http://mailman.mit.edu/
mailman/listinfo/piclist
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Ryan O'Connor]

What about evolving an antenna using a genetic algorithm
https://en.wikipedia.org/wiki/Evolved_antenna

Ryan

Post by Sean Breheny
Thanks!
On Wed, Jul 11, 2018 at 4:05 PM, Van Horn, David <

Post by Van Horn, David
Very nice!
-----Original Message-----

Sean

Post by Van Horn, David
Breheny
Sent: Wednesday, July 11, 2018 1:56 PM
Subject: Re: [EE] Antenna design?
I would recommend reading appendix A of my master's thesis :)
https://ecommons.cornell.edu/bitstream/handle/1813/8346/
thesis%20formatted.pdf?sequence=1&isAllowed=y
It is an 11-page summary of antenna theory (that I had to include because
my thesis advisor was not an antenna expert)
With an efficient passive antenna (no amplifiers and no major losses that
you can reduce), you can't get more gain without being more directive. In
other words, if you cannot be directional, then making your antenna

bigger

Post by Van Horn, David
than a half wavelength does nothing for you. This is true for BOTH

transmit

Post by Van Horn, David
AND receive because passive antennas are reciprocal devices.
A patch antenna is probably best for you. Any kind of linear antenna is
going to have a null along the long axis of the antenna. This can be a
problem for the case where the robot is on one floor and the operator is

on

Post by Van Horn, David
another floor.
Colinear arrays only make this worse - they direct more radiation in a
flat disk and have an even wider null along the long axis. They are
commonly used for cell towers because all of the cell phones that need to
access the tower are in a narrow range of vertical angle - radiation

going

Post by Van Horn, David
up into the sky (or listening for radiation from the sky) is a waste.
Sean

Post by Neil
Okay antenna gurus, I can use some help here please...
I'm using a 5.8Ghz video transmitter/receiver system on a robot (2D
driving, indoor). These systems are popular with the drone-racing
crowd, but most of the good antennas are either circular-polarized (so
they are uniformly effective at most angles) or patch antennas which
are apparently very unidirectional. Linear antennas have mostly gone
away for this purpose.
I'm trying to maximize signal strength/reception so the operator can
be in a different room and on a different floor. For my purposes, I'm
thinking a linear antenna on the transmitter side (robot) may be best
though, as I can mount it vertically in the robot body, and perhaps
use a patch antenna on the receiver side as that would have minimal
movement. Or linear as well for the receiver. I understand that whip
antennas are really crappy so would a straight piece of wire work
better? I also discovered "collinear antennas" while searching.
I have about 18-20" of room to put an antenna vertically inside the
robot ... would having an antenna length of a multiple of the
wavelength work better than just a single-wavelength antenna?
So what say ye?
Cheers,
-Neil.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at http://mailman.mit.edu/
mailman/listinfo/piclist
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[RussellMc]

Post by Sean Breheny
I would recommend reading appendix A of my master's thesis :)
​Your thesis and mine differ in most respects except for the dedication -

both have the same general intent.
Mine is 5 words.
No prizes for guessing it correctly :-)


Russell.

[Jean-Paul Louis]

Neil,
A lot of pro drone users use cloverleaf antennas to do exactly what you
do. Google "cloverleaf antenna" and you will see a good list of designs.

Just my $0.02,
Jean-Paul
N1JPL

Post by Neil
Okay antenna gurus, I can use some help here please...
I'm using a 5.8Ghz video transmitter/receiver system on a robot (2D
driving, indoor). These systems are popular with the drone-racing
crowd, but most of the good antennas are either circular-polarized (so
they are uniformly effective at most angles) or patch antennas which are
apparently very unidirectional. Linear antennas have mostly gone away
for this purpose.
I'm trying to maximize signal strength/reception so the operator can be
in a different room and on a different floor. For my purposes, I'm
thinking a linear antenna on the transmitter side (robot) may be best
though, as I can mount it vertically in the robot body, and perhaps use
a patch antenna on the receiver side as that would have minimal
movement. Or linear as well for the receiver. I understand that whip
antennas are really crappy so would a straight piece of wire work
better? I also discovered "collinear antennas" while searching.
I have about 18-20" of room to put an antenna vertically inside the
robot ... would having an antenna length of a multiple of the wavelength
work better than just a single-wavelength antenna?
So what say ye?
Cheers,
-Neil.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[David VanHorn]

Aka "turnstile"?

Post by Jean-Paul Louis
Neil,
A lot of pro drone users use cloverleaf antennas to do exactly what you
do. Google "cloverleaf antenna" and you will see a good list of designs.
Just my $0.02,
Jean-Paul
N1JPL

Post by Neil
Okay antenna gurus, I can use some help here please...
I'm using a 5.8Ghz video transmitter/receiver system on a robot (2D
driving, indoor). These systems are popular with the drone-racing
crowd, but most of the good antennas are either circular-polarized (so
they are uniformly effective at most angles) or patch antennas which are
apparently very unidirectional. Linear antennas have mostly gone away
for this purpose.
I'm trying to maximize signal strength/reception so the operator can be
in a different room and on a different floor. For my purposes, I'm
thinking a linear antenna on the transmitter side (robot) may be best
though, as I can mount it vertically in the robot body, and perhaps use
a patch antenna on the receiver side as that would have minimal
movement. Or linear as well for the receiver. I understand that whip
antennas are really crappy so would a straight piece of wire work
better? I also discovered "collinear antennas" while searching.
I have about 18-20" of room to put an antenna vertically inside the
robot ... would having an antenna length of a multiple of the wavelength
work better than just a single-wavelength antenna?
So what say ye?
Cheers,
-Neil.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[RussellMc]

Post by David VanHorn
Aka "turnstile"?

​Related.

A turnstile antenna usually has linear/rod 'arms'.

A cloverleaf antenna has curved close loops - "leaf" shaped, sort of.

Cloverleaf antenna images
<https://www.google.co.nz/search?q=cloverleaf+antenna&num=50&source=lnms&tbm=isch&sa=X&ved=0ahUKEwi2mLbU-JrcAhVWQd4KHcGYARsQ_AUICigB&biw=1680&bih=917>

Turnstile antenna images
<https://www.google.co.nz/search?q=turnstile+antenna&num=50&source=lnms&tbm=isch&sa=X&ved=0ahUKEwiSgPXc-JrcAhVM7mEKHZGgAWYQ_AUICigB&biw=1680&bih=917>



Russell