Author Topic: What is the smallest antenna that still performs well?  (Read 27173 times)

WhiteHare

  • Hero Member
  • *****
  • Posts: 1300
  • Country: us
Re: What is the smallest antenna that still performs well?
« Reply #30 on: October 21, 2016, 10:01:46 AM »

WhiteHare

  • Hero Member
  • *****
  • Posts: 1300
  • Country: us
Re: What is the smallest antenna that still performs well?
« Reply #31 on: October 28, 2016, 01:24:53 PM »
Regarding the SMD resistor antenna, one caution is that section 3.3.7 High Power Settings of the datasheet says, "The Duty Cycle of transmission at +20dBm is limited to 1%, with a maximum VSWR of 3:1 at antenna port, over the standard operating range [-40;+85°C]. For any other operating condition, contact your Semtech representative."

[Edit: Also, looking now at the RFM69 library I notice that the transmit power settings are fully maxed out by default if using an RFM69HW. ]
« Last Edit: October 28, 2016, 01:56:32 PM by WhiteHare »

WhiteHare

  • Hero Member
  • *****
  • Posts: 1300
  • Country: us
Re: What is the smallest antenna that still performs well?
« Reply #32 on: November 23, 2016, 10:51:51 AM »
RE this 1Ah CR2477:
http://www.digikey.com/product-detail/en/panasonic-bsg/CR-2477-HFN/P126-ND/107129
IMHO the cost is prohibitive given the capacity.



How about the smaller 620mAh battery, at less than half the cost:
http://www.digikey.com/product-detail/en/panasonic-bsg/CR-2450-H1AN/P662-ND/2404067

These have a variable influence on the small antenna performance since the board+antenna (most we've seen in the forum) is sandwiched somewhere in between the RFM and this type of battery. Would be great to see a compare between such batteries.

I had ordered some tabbed CR2477 batteries from China, but when they arrived, they weren't tabbed.   :o  So, I took your suggestion and ordered three tabbed CR2450's from Digikey.  Shortly afterward, I got a call from Digikey saying they can't ship them by regular mail because their lithium batteries contain, uh, lithium.   ::)   So, I ordered them anyways, but now it's costing me about $10 in FedEx ground delivery charges, or about 3x what it would have been if USPS didn't prohibit shipment of lithium coincell batteries.  Seems ridiculous to me at many levels, but there you have it.  Well, whenever they do finally arrive, at least this time I expect they will actually be tabbed.
« Last Edit: November 23, 2016, 11:01:33 AM by WhiteHare »

joelucid

  • Hero Member
  • *****
  • Posts: 868
Re: What is the smallest antenna that still performs well?
« Reply #33 on: November 23, 2016, 01:20:28 PM »
I use exactly these batteries - the panasonic 2032 and 2450 with tabs. They work well. For shipments to Germany digikey offers free Fedex express shipping starting at $70 or so. So I tend to just add stuff in to qualify for the free shipping (which is otherwise outrageously expensive.)

Joe

OSBSS

  • NewMember
  • *
  • Posts: 29
  • Country: us
    • OSBSS - Open Source Building Science Sensors
Re: What is the smallest antenna that still performs well?
« Reply #34 on: February 03, 2017, 04:25:51 PM »
This is pretty neat! For projects not too far away, this SMD resistor hack somehow 'magically' works. Nice find, WhiteHare!

I found this SMD chip antenna: http://www.mouser.com/search/ProductDetail.aspx?R=0virtualkey0virtualkey0915AT43A0026E which are about $1 each. It has some mounting considerations for PCB in its datasheet (screenshots attached if link clicking is too much of a hassle). Doesn't seem too complicated to work with, but without a network analyzer, pretty much designing in the blind. Has anyone tested performance of such chip antennas and this design?

I still think wire antennas are the cheapest solution that's most reliable, but when people start thinking about enclosures and stuff, alternatives like this may help, as long as it doesn't break the bank and add too much to the design.


WhiteHare

  • Hero Member
  • *****
  • Posts: 1300
  • Country: us
Re: What is the smallest antenna that still performs well?
« Reply #35 on: February 03, 2017, 07:27:26 PM »
Cool!  Looks like that chip antenna would be worth a try.

perky

  • Hero Member
  • *****
  • Posts: 873
  • Country: gb
Re: What is the smallest antenna that still performs well?
« Reply #36 on: February 04, 2017, 05:49:06 AM »
Here's an interesting document. They are using 'crossed dipoles' on credit cards with random meandering to reduce the size. I'm a little confused as these appear to be fed from the same source, but they're labelled circularly polarized which normally requires one of the pairs to be phase delayed. Anyhoo the omnidirectional radiation plots seem too good to be true...

http://cearl.ee.psu.edu/projects/assets/Project3/Project3_3/Stochastic%20Antennas1.pdf

Mark.
« Last Edit: February 04, 2017, 05:53:56 AM by perky »

perky

  • Hero Member
  • *****
  • Posts: 873
  • Country: gb
Re: What is the smallest antenna that still performs well?
« Reply #37 on: February 04, 2017, 06:09:27 AM »
And another interesting article about using two Splatch antennae to create a composite circularly polarized antenna:
http://digitalcommons.calpoly.edu/cgi/viewcontent.cgi?article=1062&context=eesp

Mark.

ChemE

  • Sr. Member
  • ****
  • Posts: 419
  • Country: us
Re: What is the smallest antenna that still performs well?
« Reply #38 on: February 04, 2017, 07:10:36 AM »
Interesting approach to let genetic algorithms tune the bending optimizing for both size and symmetrical radiation.  Shame that is is still 43mm x 43mm in the end that is pretty gigantic compared to a mote and still gigantic compared to 2xAAA (approximately 47mm x 20mm).  I need something 10mm x 10mm for my idea though I'm not too worried if it is directional.

WhiteHare

  • Hero Member
  • *****
  • Posts: 1300
  • Country: us
Re: What is the smallest antenna that still performs well?
« Reply #39 on: February 04, 2017, 07:35:53 AM »

I found this SMD chip antenna: http://www.mouser.com/search/ProductDetail.aspx?R=0virtualkey0virtualkey0915AT43A0026E which are about $1 each. It has some mounting considerations for PCB in its datasheet (screenshots attached if link clicking is too much of a hassle).

By the way, I just now noticed that a very similar looking chip antenna is what's being used on some of the amplified NRF24L01's:




Also, something new to look forward to: "zero clearance" chip antennas that can mount directly above or below the metal layer on a PCB.  So far I've only seen it at 2.4Ghz (http://www.mouser.com/new/Johanson/johanson-2450A-mini-chip-antenna/), but maybe there exists a zero clearance chip antenna for 915mhz too?
« Last Edit: February 04, 2017, 11:15:57 AM by WhiteHare »

joelucid

  • Hero Member
  • *****
  • Posts: 868
Re: What is the smallest antenna that still performs well?
« Reply #40 on: February 04, 2017, 08:42:37 AM »
Quote
Anyhoo the omnidirectional radiation plots seem too good to be true..

50% size reduction using meandering with no impact on gain at all does indeed sound too good to be true in my mind.

Then again check this out: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.936.8142&rep=rep1&type=pdf

80% effiency at 430 MHz using a 5.6 x 2.5 cm meander dipole. Scale down and use at 2.8 x 1.25 cm at 860 MHz? Sounds like magic - it has got to do with those ants.

WhiteHare

  • Hero Member
  • *****
  • Posts: 1300
  • Country: us
Re: What is the smallest antenna that still performs well?
« Reply #41 on: March 24, 2017, 01:28:17 PM »
This is pretty neat! For projects not too far away, this SMD resistor hack somehow 'magically' works. Nice find, WhiteHare!

I found this SMD chip antenna: http://www.mouser.com/search/ProductDetail.aspx?R=0virtualkey0virtualkey0915AT43A0026E which are about $1 each. It has some mounting considerations for PCB in its datasheet (screenshots attached if link clicking is too much of a hassle). Doesn't seem too complicated to work with, but without a network analyzer, pretty much designing in the blind. Has anyone tested performance of such chip antennas and this design?

I still think wire antennas are the cheapest solution that's most reliable, but when people start thinking about enclosures and stuff, alternatives like this may help, as long as it doesn't break the bank and add too much to the design.

I was just about to buy some of these SMD chip antennas, but a closer reading of the datasheet revealed that it requires some detailed PCB layout, including a 50ohm feedline on the PCB (which I have no idea how to do). 

Has anyone tried it?

perky

  • Hero Member
  • *****
  • Posts: 873
  • Country: gb
Re: What is the smallest antenna that still performs well?
« Reply #42 on: March 24, 2017, 03:37:52 PM »
Not tried it.

A transmission line like a coax has capacitance per unit length and inductance per unit length. It's characteristic impedance depends on those values. A similar construction can be done on a PCB by tracking a trace above a solid ground plane. Inductance per unit length is set by the track width, and capacitance per unit length is set by the distance between the track and the ground plane below it, the dielectric constant of the PCB, and the width of the track.

It just so happens to get 50R impedance with a 2 layer sided FR4 PCB you need the ratio between the width and the PCB thickness to be around 2:1, so for 1.6mm thick PCB you'll need a track 3.2mm wide. 50R is far easier to do with multilayer PCBs because the separation distances between signal and ground plane are much smaller, meaning sensible width tracks.

Mark.
« Last Edit: March 24, 2017, 03:39:58 PM by perky »

WhiteHare

  • Hero Member
  • *****
  • Posts: 1300
  • Country: us
Re: What is the smallest antenna that still performs well?
« Reply #43 on: March 25, 2017, 07:45:29 AM »
Thanks!  That info might prove useful in the case of a Splatch antenna, where the feedline is assumed to be on one side and the ground plane is assumed to cover the opposite side of the PCB.  In the case of the Johanson Technology antenna, though, the DS  seems to assume that the feedline is co-planar with the ground plane--I guess so various inductors/capacitors can be bridging between the two.  Hmm..  All in all, this does seem to re-affirm that the Splatch is comparatively easy.
« Last Edit: March 25, 2017, 07:47:39 AM by WhiteHare »