Author Topic: GPS antenna question  (Read 3349 times)

0 Members and 1 Guest are viewing this topic.

Offline MukuTopic starter

  • Contributor
  • Posts: 21
  • Country: np
GPS antenna question
« on: August 25, 2020, 07:04:20 pm »
Hello, everyone! Hope you are all doing well in this tough times. I was looking at few gps ceramic patch antennas for my project.


This is an antenna by Molex. Now they have a detailed documentation on this. They have tested the antenna on different size ground planes and the output is below.


As you can see the 50mmx50mm ground plane has the worst return loss. Now supposed my application uses PCB the size of 50X50mm, does that mean this antenna is useless for my project?
If you are working with small sized PCB then how do you increase the size of the ground plane?
 

Online m98

  • Frequent Contributor
  • **
  • Posts: 634
  • Country: de
Re: GPS antenna question
« Reply #1 on: August 25, 2020, 11:38:08 pm »
What's your target application? What does the environment of your PCB look like? Do you have any particular requirements for positional accuracy?
In general, if this is for a generic hobby project and you're pointing the antenna in the general direction of the sky, it will work perfectly fine. You might want to use a way more compact, omnidirectional SMD antenna instead, for example: https://www.mouser.de/ProductDetail/Antenova/SR4G013?qs=4NsIomN5445K9og%252BQVhwfg%3D%3D
 

Offline cdev

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: GPS antenna question
« Reply #2 on: August 26, 2020, 09:13:17 pm »
GPS, Beidou and Glonass use slightly different frequencies within that range.

The (kind of standard) taoglas 25mm patch GPS antenna works on a 50 mm GP but it does perform much better on a larger groundplane. The size of the groundplane as they point out does vary the way it resonates, significantly. 

I use (two) antennas almost identical to that one in GPS applications that give a better idea of how well as GPS antenna is performing than most do. Both are taped to CD's because they work a lot better like that.

 Just making it a bit larger improves its performance substantially. It works the best in the center of the largest piece of flat metal you can find.

This is generally true for all ceramic patch GPS antennas.

The smaller the GPS patch is the more critical tuning is and the less likely it is to cover all three GNSS constellations too - especially GLONASS, which is an outlier..

OTOH quadrifilar helix or similar designs dont require any ground plane at all, and are substantially more flexible in placement, which in some applications, such as a handheld unit thats not always going to have its top facing up, thats very important.
« Last Edit: August 26, 2020, 09:23:26 pm by cdev »
"What the large print giveth, the small print taketh away."
 

Offline cdev

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: GPS antenna question
« Reply #3 on: August 26, 2020, 09:36:14 pm »
Photos would help. As m98 says, it depends on how accurate you need it to be. A slightly larger ground plane will also make a big difference in a passive antenna situation, but it it has its own LNA, that will matter much less for having a GPS fix but that fix will jump all over the place if its non optimally placed.. Especially when the sky view isnt so great.

With any patch antenna, it needs to be resonating at the right frequency to work too. Using a patch antenna thats designed to resonate in the middle of a flat piece of metal on the end of a rectangular box isnt going to work well, unless its custom tuned to perform there.
"What the large print giveth, the small print taketh away."
 

Offline amyk

  • Super Contributor
  • ***
  • Posts: 8525
Re: GPS antenna question
« Reply #4 on: August 27, 2020, 03:13:37 am »
Look at the GPS antennas in smartphones. They are much smaller, yet work perfectly fine. It's all down to the sensitivity of the receiver.
 

Offline thinkfat

  • Supporter
  • ****
  • Posts: 2161
  • Country: de
  • This is just a hobby I spend too much time on.
    • Matthias' Hackerstübchen
Re: GPS antenna question
« Reply #5 on: August 27, 2020, 12:08:14 pm »
Look at the GPS antennas in smartphones. They are much smaller, yet work perfectly fine. It's all down to the sensitivity of the receiver.

No they don't. In fact they're so bad that the GPS in a phone will hardly achieve a lock without being assisted by the network. Don't believe me? Fire up any GPS diagnostic software on a smartphone and check the Carrier-over-Noise C/NO. How many sats do you see with a C/NO over 30dB? That's the minimum for an autonomous cold start and C/NO must be better than 30dB for about 40 seconds continually for the ephemeris download to work out. BTW the exaggerated sensitivity figures found in the marketing materials (like -167dB for recent u-blox modules) are for tracking sats only. For acquisition, you need a lot more signal.
Everybody likes gadgets. Until they try to make them.
 

Offline cdev

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: GPS antenna question
« Reply #6 on: August 27, 2020, 03:04:10 pm »
What I think is the problem is how the resonant frequency of a ceramic patch antenna shifts based on the size of its groundplane. I am now actually set up to show this, if you want a demo, it would take me five minutes. I have a taoglas 25 mm GPS patch antenna on the standard taoglas 50 mm square groundplane. Thats the minimum size to use a non custom tuned patch. But really they are made to be placed - via a magnet, on the roof of a car or similar.

That said, antennas for PNDs ('personal navigation devices" are likely sold with their default trimming optimized for applications like car GPS's.

 Contact manufacturers! 

They likely run into this all the time. The kind of square patch antennas I have are all meant to live in the middle of a ground plane. But others are designed to work on the edge of a PCB, I'm certain.
« Last Edit: August 27, 2020, 03:13:38 pm by cdev »
"What the large print giveth, the small print taketh away."
 

Offline thinkfat

  • Supporter
  • ****
  • Posts: 2161
  • Country: de
  • This is just a hobby I spend too much time on.
    • Matthias' Hackerstübchen
Re: GPS antenna question
« Reply #7 on: August 27, 2020, 03:54:21 pm »
Even ceramic antennas for PBC mounting need a sizeable ground area to work properly.
Everybody likes gadgets. Until they try to make them.
 

Offline cdev

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: GPS antenna question
« Reply #8 on: August 27, 2020, 04:30:00 pm »
I'ts entirely possible that some car GPS's use a small QFH antenna mounted on the top. Some manufacturers make ceramic QFH and patch antennas that are very small. They contain small LNAs, too.

But they are very unlikely to support multiple GNSS systems like Glonass, etc. along with GPS. And when they are ordered that are likely a custom job thats tuned to work with that specific application. I have two PNDs. One gets used in the car, the other one I use as a GPS logger when I am hiking. Both have to be facing up to work properly. They are both very thin and couldnt possibly have any decent ground plane in there. One of them has a little lump on a top corner where the antenna lives. I'm pretty sure its a tiny rectangular (not square) patch antenna. When I go hiking (often its in the big city, because thats where I do the most walking, which is a nightmare for GPS signals) I always use a specific pocket on my backpack that was made big enough to hold a portable CD player..  which holds it with the top facing up. Then its pretty sensitive and surprisingly accurate. But otherwise the traces are useless. 

« Last Edit: August 27, 2020, 04:33:41 pm by cdev »
"What the large print giveth, the small print taketh away."
 

Offline amyk

  • Super Contributor
  • ***
  • Posts: 8525
Re: GPS antenna question
« Reply #9 on: August 28, 2020, 12:30:21 am »
Look at the GPS antennas in smartphones. They are much smaller, yet work perfectly fine. It's all down to the sensitivity of the receiver.

No they don't. In fact they're so bad that the GPS in a phone will hardly achieve a lock without being assisted by the network. Don't believe me? Fire up any GPS diagnostic software on a smartphone and check the Carrier-over-Noise C/NO. How many sats do you see with a C/NO over 30dB? That's the minimum for an autonomous cold start and C/NO must be better than 30dB for about 40 seconds continually for the ephemeris download to work out. BTW the exaggerated sensitivity figures found in the marketing materials (like -167dB for recent u-blox modules) are for tracking sats only. For acquisition, you need a lot more signal.
I have a no-name several-years-old generic Mediatek phone get a lock within ~2 minutes standing indoors next to a window from a cold start. Not even a SIM card or WiFi connection to use. It works better outdoors but I consider that adequate.
 

Offline cdev

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: GPS antenna question
« Reply #10 on: August 28, 2020, 03:10:22 pm »
They grab their ephemeris from the network even if there is no sim card. So it can still send an e911 location info if you make an emergency call which doesnt need a sim card. The reason a standalone, non networked GPS takes so long is it needs to download a table of expected orbital data into memory to work properly. That takes around 15 minutes.

People in North Korea using the Chinese cell network learned (this info is probably outdated) that if they kept their Chinese cell phone without its batteries in it, and then just put the batteries in and quickly called relatives in South Korea and then hung up right away, and removed the batteries, the old system the MSS had often could not locate people to be taken away to concentration camps. A few years ago they got a much more modern system to track down illegal cellphone users that works I suspect on TDOA, and its accurate to just a few meters. In this country where people are literally starving the government had stopping illegal international calling as their top priority, so they bought a system built its rumored by Rhode and Schwarz, which must have been sold to them by some 3rd party dealer, because of sanctions. Now all their cell phones, probably just like other countries, send their GPS locations, probably right away, due to AGPS. So they have multiple, redundant systems for locating illegal users of foreign telephone networks in their border areas.

----

A cell phone has access to the network so they now use AGPS, which basically means fetch the ephemeris over the network enabling an almost instant start.

If you dont pay for a data plan it still gets the ephemeris data and your GPS's location data is available to everybody else but you.

Will it fall back to downloading the ephemeris from the satellites when its not in range of a cell network? I don't know but I kind of doubt it. So dont count on using a cell phone GPS is you are a hiker, becuse as soon as you are over the hill from a cell phone site the GPS may stop working, (if the phone cant fall back to using the actual satellite ephemeris) either right away or in a day or two when the ephemeris data gets stale.

There should be a law requiring that GPSs natural functioning not be disabled for use by users by greedy cell phone companies.


-----

Yesterday I was surfing for some other reason and I noticed that Molex now sells compact helical GPS antennas. I dont know if they are QFH but it seems that they might be, if so they may be more likely to operate properly independently of any ground plane.

Sarantel used to make very good - very copact (tiny) ones but I remember hearing they are no longer in business. They are worth hunting down and using if you can find one.  But even if you can find one now it might be hard to find them in the future.

Similarly, if this is for a hobby use maybe you could find an old Vaisala aerosonde (QFH) GPS antenna. If you really want an GPS antenna that will work in almost any orientation, thats what I would pick. They are a full sized QFH for GPS which means two interwoven helical loops with a height or around 8 cm which fit in a cylinder with a diameter of around 3 cm.
« Last Edit: August 28, 2020, 09:39:27 pm by cdev »
"What the large print giveth, the small print taketh away."
 
The following users thanked this post: Muku

Offline MukuTopic starter

  • Contributor
  • Posts: 21
  • Country: np
Re: GPS antenna question
« Reply #11 on: August 28, 2020, 09:13:56 pm »
Thanks for reply cdev. I learned alot about GPS in general.

I thought of QFH antenna but I am making a tracker for my bike so which will mostly be used outdoors facing the sky. QFH are also little expensive. I am thinking of using one of those GNSS modules with attached patch antenna. There's one made by company called telit and they also have small form factor. Lets see how that will perform. I will let you know

Thanks
 

Offline cdev

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: GPS antenna question
« Reply #12 on: August 28, 2020, 09:24:01 pm »
A small patch antenna will probably work fine. You could also if you want, make a QFH antenna but if you do so make sure to use a blocking cap because a QFH being (two) loops will look like a short to any GPS that supplies power to an antenna. (Also protect the GPS's antenna input from ESD). A helical antenna is very easy to make and would work. I would just use a cheap puck antenna from ebay, whatever is acceptable cosmetically with the aforementioned tradeoffs.

Thanks for reply cdev. I learned alot about GPS in general.

I thought of QFH antenna but I am making a tracker for my bike so which will mostly be used outdoors facing the sky. QFH are also little expensive. I am thinking of using one of those GNSS modules with attached patch antenna. There's one made by company called telit and they also have small form factor. Lets see how that will perform. I will let you know

Thanks
« Last Edit: August 28, 2020, 09:45:16 pm by cdev »
"What the large print giveth, the small print taketh away."
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf