Effects of a dielectric layer over a GPS patch antenna

The Rocketry Forum

Help Support The Rocketry Forum:

Voyager1

Well-Known Member
Joined
Oct 23, 2015
Messages
817
Reaction score
298
Location
Canberra, Australia
This thread was motivated by a discussion initiated within another thread. The question is whether using a heatshrink protective cover over a GPS patch antenna is detrimental to the antenna's, and subsequently, the GPS receiver's performance. I found a number of papers discussing the application of a dielectric layer over a patch antenna. The thickness of the dielectrics in these studies typically ranged from 0mm (no dielectric) to 6mm. Heatshrink might be of the order of 1mm, or less, when applied.

Plastic heatshrink is a dielectric material with a dielectric constant of perhaps 2-3, depending on its composition and the frequency. When it is employed to secure a GPS patch antenna, the heatshrink covering the patch becomes what is termed a superstrate layer. Dielectric superstrate layers are employed to modify the characteristics of some patch antennas including: the gain, bandwidth and beamwidth. These parameters are modified by increasing the thickness of the dielectric layer above the patch.

Some studies suggest that the impedance, gain and bandwidth display a periodic variation as the thickness increases. In some cases the gain and bandwidth can actually increase, before falling off with a periodic variation. It was also found that the antenna impedance varies periodically about its normal value, typically 50 Ohms (Real component). Interestingly, the resonant frequency does not change as rapidly as the other parameters, but tends to drop as the dielectric thickness increases. The on-axis beamwidth tends to narrow with the formation of sidelobes as the thickness increases.

Anyone with differing experiences and knowledge are invited to join the discussion, particularly if they can provide actual data. Which ever way this goes, we can only benefit from the outcome. I certainly hope to perform some measurements when I can.
 
Last edited:

Grog6

Well-Known Member
Joined
Jul 19, 2021
Messages
637
Reaction score
507
Location
Oak Ridge TN
The added dielectric will change the 'coupling constant' between the antenna and the 'ether' this shows up in transmitters as SWR, or standing wave ratio. Minimizing this ratio optimizes the coupling.
Adding inductance will off set the added dielectric.
 

OverTheTop

Well-Known Member
Joined
Jul 10, 2007
Messages
6,985
Reaction score
4,844
Location
Melbourne Australia
Air is a very good dielectric and heatshrink not as good, as well as having higher dielectric constant. I guess the heatshrink in very close proximity to the patch would increase the fringing fields which these antennas rely to work. Increasing the fringing would effectively add capacitance to the antenna and detune it in the direction of lower frequency. That seems to be in agreement with statement in post #1 regarding frequency dropping as dielectric thickness goes up.

The question is by how much?
 

Voyager1

Well-Known Member
Joined
Oct 23, 2015
Messages
817
Reaction score
298
Location
Canberra, Australia
I plotted some data from one of the papers I found showing resonant frequency, gain and bandwidth as a function of dielectric thickness over a patch antenna at 2.4 GHz. You can see the periodic variation in the gain and bandwidth. The resonant frequency does not significantly drop until after 2mm.
 

Attachments

  • patch_with_superstrate_layer.png
    patch_with_superstrate_layer.png
    341.1 KB · Views: 20
Last edited:

jderimig

Sponsor
TRF Sponsor
Joined
Jan 23, 2009
Messages
4,183
Reaction score
1,975
Which of the cheap nano vna's would be useful for this investigation?
 

gtg738w

FlightSketch - flightsketch.com
Joined
Dec 27, 2018
Messages
340
Reaction score
355
Any kind of dielectric around the antenna will cause an impedance change. This is a patch antenna in air (blue) and the same thing in a 2.2" cardboard airframe (yellow). In reality, anything you add to a system will almost certainly make it worse unless everything is optimized together. Even then there is little to gain for GPS since you don't really want a gain increase and any improvement in efficiency would be small if any.

2.png
 

Voyager1

Well-Known Member
Joined
Oct 23, 2015
Messages
817
Reaction score
298
Location
Canberra, Australia
Any kind of dielectric around the antenna will cause an impedance change. This is a patch antenna in air (blue) and the same thing in a 2.2" cardboard airframe (yellow). In reality, anything you add to a system will almost certainly make it worse unless everything is optimized together. Even then there is little to gain for GPS since you don't really want a gain increase and any improvement in efficiency would be small if any.

View attachment 501828
Yes, totally agree. As suggested in the other thread, just mounting the GPS module in an airframe will expose it to several potentially interfering structures.
 

RFMan

RF engineer
Joined
May 28, 2014
Messages
188
Reaction score
52
Location
Cary, NC
The heatshrink tubing polymer has both a dielectric constant (changes resonant frequency, gain, bandwidth, impedance...) and a dielectric loss tangent (changes efficiency, gain...). Black tubing can be pretty lossy at higher frequencies.

The big thing, though, is that this is a GPS patch antenna on a ceramic substrate. The patch is miniaturized because the ceramic dielectric constant is relatively high. This means that the Q is high and the bandwidth narrow. Out in the open, away from nearby objects, this is not a big deal, because the GPS signal itself is narrowband. It you introduce items into the antenna's environment that shift its resonant frequency, then the antenna's effective gain will drop, and it will not work as well.

Personally, I would look for other ways to secure the antenna. But using the thinnest clear heatshrink you can find should minimize the effect.
 

RFMan

RF engineer
Joined
May 28, 2014
Messages
188
Reaction score
52
Location
Cary, NC
Which of the cheap nano vna's would be useful for this investigation?
There are so many!

I would get one of the 3 GHz versions - the 1.5 GHz versions do not quite go high enough. I've seen them for as little as ~$75 on Amazon. Might have to spring for an adapter, also, depending on how the antenna is connectorized - assuming it even is. If it's not, well, that's a PITA.
 

OverTheTop

Well-Known Member
Joined
Jul 10, 2007
Messages
6,985
Reaction score
4,844
Location
Melbourne Australia
Get one of the later revsions of the VNA. The earlier versions used harmonics to get up in the frequency bands. I have a V2 unit (Nano VNA) and it generates the frequencies directly rather than relying on harmonics. The hardware is much better in the more contemporary units.
 

Latest posts

Top