Searching \ for '[EE] Eagle printed inductor as antenna' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/pcbs.htm?key=eagle
Search entire site for: 'Eagle printed inductor as antenna'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Eagle printed inductor as antenna'
2010\12\21@144124 by David

flavicon
face
Has anybody used print-inductor.ulp that comes with Eagle to design a
PCB antenna?

I would like a 433Mhz antenna and I have plenty of board space free, but
not enough board width to print a dipole.

As an alternative to a 17cm piece of wire hanging out of the (plastic)
case, I thought this might be a neat idea.  I could create a spiral
trace as near to 172mm length as possible, fed from the RFM12 module.
However if the output from that ULP is not at all suitable, I shouldn't
even go down that route.

I don't need huge range (10m to 20m max) and can accept that it wont be
omni-directional.

Davi

2010\12\21@151058 by Matt Pobursky

flavicon
face
How about an antenna like this one?

<http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=ANT-433-HETH-ND>

Small, inexpensive and generally works better than a PCB trace antenna.
Relatively easy to hide in all but the tiniest enclosures. We've had very
good luck with this line of antennas. Also available in an SMD version if
you're so inclined.

Matt Pobursky
Maximum Performance Systems


On Tue, 21 Dec 2010 19:41:19 +0000, David wrote:
{Quote hidden}

> David

2010\12\21@151330 by Olin Lathrop

face picon face
part 1 1425 bytes content-type:text/plain; charset="iso-8859-1" (decoded quoted-printable)

David wrote:
> I would like a 433Mhz antenna and I have plenty of board space free,
> but not enough board width to print a dipole.

There are many antenna topologies beyond a dipole.  Also, you can insert
components into the antenna pattern on the PCB to make it resonate or have
the desired impedence at a much smaller size.

A loop antenna is pretty easy.  You can cut the loop in one place and put a
capacitor in series to get resonance or all-real impedence at a particular
frequency.  With a loop you can change the driving impedence by where (how
far apart) you feed it.  A loop has the same radiation pattern as a dipole,
with the nulls being at right angles to the loop.  A dipole has the nulls in
the direction parallel to the antenna.

Attached is a piece of a board drawing showing a loop antenna.  The grid
lines are 100 mil pitch.  This was for a bench test unit.  I didn't know the
best feed points for the loop at the time, so I left lots of holes.  Someone
else found how many holes apart the feedpoints should be and what resonating
components to insert at the other end to get the best result for 434 MHz.
If I remember right, the feedpoints were a little less than 1/4 circle apart
and the resonating capacitor was a few pF.

This stuff is real hard to guess, so leave room for different parts and
choices, then find the best ones by trial and error.


part 2 23642 bytes content-type:image/gif; name="b.gif" (decode)


part 3 181 bytes content-type:text/plain; name="ATT00001.txt"
(decoded base64)

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

2010\12\21@153302 by RussellMc

face picon face
This looks reasonably apposite and useful.
Unlike most such this is 100% accessible.(creative commons, attribution).

        http://www.hindawi.com/journals/ijmst/2008/287627.html



<http://www.hindawi.com/journals/ijmst/2008/287627.html>     Russell

On 22 December 2010 08:41, David <spam_OUTlistsTakeThisOuTspamedeca.net> wrote:

> Has anybody used print-inductor.ulp that comes with Eagle to design a
> PCB antenna?
>
> I would like a 433Mhz antenna and I have plenty of board space free, but
> not enough board width to print a dipole.
>

2010\12\21@154443 by David

flavicon
face
On 21/12/2010 20:10, Matt Pobursky wrote:
> How about an antenna like this one?
>
> <search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=ANT-433-HETH-ND>
>
> Small, inexpensive and generally works better than a PCB trace antenna.
> Relatively easy to hide in all but the tiniest enclosures. We've had very
> good luck with this line of antennas. Also available in an SMD version if
> you're so inclined.

I had not seen those, only the small SMD chip antenna.  They look neat,
will investigate as an alternative.

Thanks!

David

2010\12\21@164249 by David

flavicon
face
On 21/12/2010 20:10, Matt Pobursky wrote:
> How about an antenna like this one?
>
> <search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=ANT-433-HETH-ND>
>
> Small, inexpensive and generally works better than a PCB trace antenna.
> Relatively easy to hide in all but the tiniest enclosures. We've had very
> good luck with this line of antennas. Also available in an SMD version if
> you're so inclined.

OK, taking a look at these they seem like they might be very useful.

In your experience using them, how directional are they?  The
propagation diagrams in http://tinyurl.com/2a4mwan are interesting.

If I read this right, then parallel to ground it is essentially
radiating completely away from the plane and therefore would be highly
directional, with nothing being sent "backwards" (excuse the
terminology).  This would mean, I assume, that the unit would need
pointing at the others.

What sort of range do you get and with what TX/RX?

And finally, how do you feed them?  The datasheet mixes units here,
which isn't helpful.  It seems the minimum distance to ground plane is
..5", but max feed trace is .25" (I assume inches here), suggesting the
ground trace can't be underneath the radio unit either.  My ground plane
is entirely on the bottom of the board, not the top.

They also have a few other neat antenna including "the splatch"
(http://tinyurl.com/3yqnjgf), but sadly no propagation diagrams for that
one.

A PCB antenna or something like this would be great, but if it gets too
confusing I might resort to a piece of wire!

Davi

2010\12\21@175324 by IVP

face picon face
David,

some information I've gathered for a current project

http://www.rfm.com/

http://www.rfm.com/support/short_range_radios.php

App Notes

http://www.rfm.com/products/apnotes/antenna.pdf

That pdf has many examples of PCB antenna and their effectiveness,
eg shows the gain and radiation pattern

http://www.ingecom.ch

Application Note 01 - V/UHF Antenna Design

I've been comparing PCB traces, small PCB helicals, whips and 'other'

One of the 'others' is the Linx Splatch

http://www.antennafactor.com/

http://search.digikey.com/scripts/DkSearch/dksus.dll?Cat=3540022&k=planar

ANT-433-SP_-ND is very nearly as good as an external 1/4 wave
whip. Range for a 434MHz transmitter said to be 7dBm is 320m
with a whip, a little over 300m with a Splatch (using a ground plane
of 38mm x 90mm)

AFAICT so far, order of increasing efficiency is PCB, short helix, Splatch
and whip

Microchip have design notes for rfPIC PCBs

http://ww1.microchip.com/downloads/en/AppNotes/tb069a.pdf

and in the rfPIC12F675F (434MHz) datasheet DS70091A

Jo

2010\12\21@180305 by IVP

face picon face
> My ground plane is entirely on the bottom of the board, not the top

David, AIUI a ground plane under a PCB trace will short it out at
RF frequencies (what a purposely-used ground plane is meant to do)

> A PCB antenna or something like this would be great, but if it gets
> too confusing I might resort to a piece of wire!

I'd be very surpised if you can't get a reliable link to 20m with just
about anything - depending on transmission power of course - even
an 1/8 wave wire (85mm)

Jo

2010\12\21@183818 by John Coppens

flavicon
face
On Tue, 21 Dec 2010 21:42:45 +0000
David <.....listsKILLspamspam@spam@edeca.net> wrote:

> What sort of range do you get and with what TX/RX?

Of course it depends on how you mount them, but if perpendicular, or
'in front' of the PC board (not parallel to), these antennas are known
as helical antennas. They are the same antennas as those 'rubber
duckies' seen on handies and such. They act as a 1/4 radiator, and
basically are fed the same way.

Joh

2010\12\21@185036 by IVP

face picon face
> ANT-433-HETH-ND>
>
> OK, taking a look at these they seem like they might be very useful.

In this pdf

http://www.radiometrix.com/files/additional/wrx2.pdf

there's a suggested winding for a short helix and some discussion

24 turns of 0.5mm wire on a 3.2mm former for 433MHz

I found it just as informative to look in datasheets for real
products as looking for technical data per se

Jo

2010\12\21@203651 by Roger, in Bangkok

face
flavicon
face
To etch your own giggleate on stripline or microstrip for impedance
calculators, etc.

RiB


On Wed, Dec 22, 2010 at 02:41, David <listsspamKILLspamedeca.net> wrote:

{Quote hidden}

>

2010\12\22@095539 by David

flavicon
face
Quoting IVP <.....joecolquittKILLspamspam.....clear.net.nz>:
> http://www.rfm.com/products/apnotes/antenna.pdf

I had read this, a fantastic primer.

> One of the 'others' is the Linx Splatch
>
> ANT-433-SP_-ND is very nearly as good as an external 1/4 wave
> whip. Range for a 434MHz transmitter said to be 7dBm is 320m
> with a whip, a little over 300m with a Splatch (using a ground plane
> of 38mm x 90mm)

I had also seen this.  However I wasn't sure what sort of propagation  pattern it had.

300m is mighty impressive, but if that is completely directional from  the board edge then perhaps less useful.

Have you tried a Splatch?  Either that or the short helical  recommended by Matt look fine, am just interested in how directional  they will be.

I am assuming from their datasheet diagram that the ground plane can  just be the whole board ground plane, but you mention 38x90mm  specifically?

Davi

2010\12\22@102435 by Olin Lathrop

face picon face
David wrote:
> 300m is mighty impressive, but if that is completely directional from
> the board edge then perhaps less useful.

I thought you said some directionality is OK.  You were after all saying a
dipole would be fine if it wasn't so large.

What pattern are you looking for?  In the plane of the board?  Perpendicular
to the board?  Something else?


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000

2010\12\22@112854 by David

flavicon
face
Quoting Olin Lathrop <EraseMEolin_piclistspam_OUTspamTakeThisOuTembedinc.com>:
> David wrote:
>> 300m is mighty impressive, but if that is completely directional from
>> the board edge then perhaps less useful.
>
> I thought you said some directionality is OK.  You were after all saying a
> dipole would be fine if it wasn't so large.
>
> What pattern are you looking for?  In the plane of the board?  Perpendicular
> to the board?  Something else?

OK, so this stems from my lack of understanding of RF in general.  The  plastic enclosure I have chosen will be mounted vertically, that is  the board will be at 90 degrees with respect to the "floor".

My assumptions after reading a few theory papers are that:

Perpendicular to the ground plane, radiation for antenna such as a  whip or helical will be away from the ground plane.  This would mean  that orientation for short distances such as 20m would not be  critical, as long as the front side of the board was facing the  receiver.

Parallel to the ground plane, radiation will be away from the board  edge and quite directional.  My assumption is that the whole board  must therefore be oriented at least approximately in the correct  direction.  i.e. if the helical or Splatch antenna mentioned in this  thread are mounted on the right edge of the PCB, that edge must be  facing the other end of the radio link.

If my reading of the propagation diagrams is correct there will be  absolutely no transmission in the opposite direction, i.e. a receiver  located to the left hand side of the board would be useless.

There is not room to mount even a small helical antenna vertically on  the board (perpendicular to the plane), as the case is less than an  inch thick.  This is why the Splatch or helical mounted flat to the  board were attractive.

I realise I could just dangle a piece of wire through the case and be  done with it, but I'd quite like to understand properly how this  works.  Also, a built-in antenna would be quite cool.

Davi

2010\12\22@130233 by Olin Lathrop

face picon face
David wrote:
>> What pattern are you looking for?  In the plane of the board?
>> Perpendicular to the board?  Something else?
>
> ... the board will be at 90 degrees with respect to the "floor".

But what radiation pattern are you looking for?  In the plane of the board?
Perpendicular to the board?  Something else?

Now that you've described the orientation of the board, you could also tell
us what radiation pattern you would like from this gizmo mounted on the
wall.  In any case, please answer the questions asked.  Supplying additional
information is fine, but it's frustrating to help when you won't provide
information specifically asked about.

> Perpendicular to the ground plane, radiation for antenna such as a
> whip or helical will be away from the ground plane.

"Away" from the ground plane is rather ambiguous here.  Please be more
careful.  If you mean perpendicular to the ground plane on the antenna side,
then that's incorrect.  Antennas that rely on a ground plane generally
radiate parallel to the ground plane, but only in the half-space of the
antenna side.  There will be some spill to the other half-space,
particularly when the ground plane is only a wavelenth or less in extent.

> Parallel to the ground plane, radiation will be away from the board
> edge and quite directional.

There you go again.  You really can't discuss this stuff while being sloppy
with words and descriptions.  Parallel to the ground plane is clear enough,
but then you suddenly bring the board into it without specifying how that is
oriented with respsect to this ground plane or the antenna.  Duh.

> My assumption is that the whole board
> must therefore be oriented at least approximately in the correct
> direction.

This is getting rediculous.  How are we supposed to know what aspect of the
"board" you consider its direction.  The plane?  Perpendicular to a face?
And how is all this oriented with respect to the antenna and its ground
plane, if your antenna even uses a ground plane?

If the board is mounted vertically (parallel to a building wall) and you
want to radiate horizontally (to every place on the same floor of the
building), then it won't help to try to make a ground plane of part of the
board.  Probably the best choice is a self-contained (not relying on ground
plane) whip oriented vertically extending above or below the board.  It
would have to contain some reactive elements to achieve resonance or the
desired impedence at substantially shorter than 1/2 wavelength.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000

2010\12\22@171431 by IVP

face picon face
> 300m is mighty impressive, but if that is completely directional from  
> the board edge then perhaps less useful.
>
> Have you tried a Splatch?  Either that or the short helical  
> recommended by Matt look fine, am just interested in how directional  
> they will be.

I have tried the Splatch. The only long-range test I had time for
before the weather turned bad (and stayed bad, rain, 30C, 100%
humidity and then Christmas week came along) was with the board
(see below) standing vertically and the receiver whip also vertical.
The range comparison was with a vertical transmitter whip

I've wound a helical and tried it briefly, with a 50x80 PCB ground
plane. I got to about 80m, walking away with the receiver. What I
want to do next is place the receiver and walk away with the transmitter,
adjusting the helix for the best range, using the LED on the receiver

> I am assuming from their datasheet diagram that the ground plane
> can  just be the whole board ground plane, but you mention 38x90  
> specifically?

I meant 38 x 85. I'm not sure of its exact importance but in the Splatch
datasheet they say "Electrical specifications and plots measured on 1.50"
x 3.30" reference ground plane", so that's what I've started with. The
board overall is 38 x 100. 38 x 15 is a bare area for the Splatch, as in
their Recommended Mounting diagram

The weather isn't too bad today. I have time to do a horizontal test
and let you know. If Linx say the Splatch approximates a 1/4 wave
then you might expect the same 3D propagation. We shall see

Jo

2010\12\23@041307 by IVP

face picon face
> 300m is mighty impressive, but if that is completely directional
> from the board edge then perhaps less useful

David, just been out for several walks

I can see no significant difference in whatever orientation the
Splatch is in, vertical or horizontal, with either the short or long
edge facing the detection direction. Which is good news for at
least one application I have in mind. The signal starts to weaken
at around 300m, and you have to go sideways here and there to
pick it up. So you might call it a reliable 200m, fairly reliable at
250m and pushing your luck at 300m

BTW, this is not just a signal strenth test. A pre-amble and
16-bit ID code is sent (as 32 bits of Manchester) which must
be received, decoded and confirmed for the receiver's LED
to flash. If you were detecting something more basic, like a
1kHz tone, the range would possibly be a bit further

Will be having a go at the helix next

Jo

More... (looser matching)
- Last day of these posts
- In 2010 , 2011 only
- Today
- New search...