Looking for a GPS module

[8086]

I am trying to get a good mix of price/physical size/ease of use.

So far I have been looking at the Maestro A2035-H

I wonder if any of you know of any smaller/cheaper modules?

Or perhaps a solution that isn't already put together? GPS chipset + amp + antenna? Could it be done cheaper than the module?

[bobdagangster]

http://www.hobbyking.com/hobbyking/store/__31135__NEO_6M_GPS_Module.html
I've used a few of these modules and they really work great compared to the Sparkfun Venus I was using before.
But they arent that much cheaper.

[Erwin Ried]

I don't think you can find something cheaper than the one in hobbyking, I have one similar from ebay GTPA010 used in this "pendrive" http://www.thingiverse.com/thing:26887 very easy to use (wires to a 3V batt, plus some other headers, supports both usb and serial interface), not very good reception for indoor things, but acceptable.

[Psi]

U-blox make good quality GPS chipsets so that NEO-6m should be pretty good.

You can also run the u-center PC software and reconfigure all the messages and options.

[jpb]

I'm happy with the Adafruit from proto-pic

http://proto-pic.co.uk/ultimate-gps-breakout-66-channel-w-10-hz-updates-mtk3339-chipset/

It is rather more expensive though, but I wanted a good 1pps signal for timing with the option of an external antenna. It is also easy to connect to a Raberry Pi (see the Adafruit site) though I've not yet done this. Annoyingly though it only supplies 3.3V to the antenna even though the input voltage is 5V (which is converted to 3.3V with an on-board regulator) - this means it won't work with the 5V only antenna I bought on e-bay. It works very well with standard modern puck antennas though.

[ddavidebor]

Ublox is very good and very popular in china

[mrflibble]

+1 for Ublox NEO-6M. Also, is this for navigation or timing?

[kripton2035]

don't think you can go cheaper than this one :
http://www.ebay.com/itm/U-blox-PCI-5S-GPS-PCI-E-B39-Mini-Wireless-Card-GPS-Module-/350753812611?pt=US_Internal_Network_Cards&hash=item51aa8e7083

the useful pinouts are in the attached jpg

I never used this module, so I dont know how it works in real field.

[cyr]

Are size and price the only criteria? What is the application?

[8086]

Thanks for the replies so far, folks

A little more info:

It's for an animal tracking project, but it's going to ultimately be aimed at mass-market so the cost must be as low as possible.

I'm concerned with price first and foremost, then physical size, then power consumption.

Is it nontrivial to put together a GPS chipset/amp/antenna yourself? I feel like I could potentially make a smaller solution, or perhaps even just fit it to the form factor more effectively than using a premade module. Anyone have experience with this sort of thing?

[ivan747]

Is it nontrivial to put together a GPS chipset/amp/antenna yourself?

To my limited knowledge, it requires lots of trigonometry to calculate the GPS position, you might need to use less common parts and that can potentially make your solution more expensive than a GPS module or a chip designed for GPS. Let's say you need a $5 DSP and $7 worth of amplifier parts, add to that the cost of a suitable impedance matched board, it could get expensive.

Something you can do is get the chip used in the GPS module. These chips should come with an application note detailing on their implementation. I tried to find one on Broadcom, but I cannot find a link to the datasheet, so they probably want you to email them .

[Psi]

From a purely legal standpoint i'm pretty sure you can't just build your own gps chipset.

My knowledge of this maybe quite old but i think there's a GPS receiver license that you pay for in the price of every GPS module when you buy it.

There are also software features a GPS chipset needs to have to get a license.
For example, all GPS receivers should stop working above X height and/or above X speed.
This is to stop them being used to guide missiles by anyone other than the US military.


In reality i'm sure there are dodgy GPS chips that don't have a licenses and probably don't implement such restrictions.

[WarSim]

PSI
I though there was restrictions also.  But was proved wrong.   
I and others have experimented with consumer GPS over the years. 
Only one could not track at over 300 MPH and over 10,000 ft.
The one that couldn't would lockup occasionally but would reset and try again. 
The only problems seemed to be with display limitations. 
EG:  The speed and alt displays overflowed and showed the least significant digits.

Lat long would be right but only lagged by the required amount for commercial products. 
 
The tested unit included Magellan and Garmin.   

[Psi]

Interesting.

[ElectroIrradiator]

Is it nontrivial to put together a GPS chipset/amp/antenna yourself?

If you mean cloning the actual functionality of the GPS receiver/decoder IC, then no. Not doable in any practical sense. For instance GPS receivers have to correct for Earth's gravitational field by applying Einstein's Theory of General Relativity (no kidding), so most people would find the math to be fairly tough going. It might be possible to get away with just using Euclidean geometry though, if you are not too interested in the absolute positional accuracy of the receiver.

There are also some really tricky timing bits in there as well, not to mention the RF bits wouldn't be easy to fabricate, at least not without a fair amount of experience.

If you have the chipset, then with some PCB fabrication experience you definitely can make the PCB assembly yourself, including the antenna and any controlling MCU you may need. Not completely trivial, but the app notes for the GPS chips frequently explain all the RF details you absolutely need to follow, including the antenna design. So you can often 'steal' the RF design from one chip to use with another, if your chosen GPS receiver chip doesn't provide the needed detail.

Something like this, but with added PCB track antenna, MCU and whatever else you need.

[komet]

Is it nontrivial to put together a GPS chipset/amp/antenna yourself? I feel like I could potentially make a smaller solution, or perhaps even just fit it to the form factor more effectively than using a premade module. Anyone have experience with this sort of thing?

Have a look at this then

http://lea.hamradio.si/~s53mv/navsats/digital.html

The gentleman in question is clearly mad, in the best possible way. - There's also an FPGA based design somewhere, but the above is just a better read.

[ptricks]

It's for an animal tracking project, but it's going to ultimately be aimed at mass-market so the cost must be as low as possible.

The first thing you need to look at isn't price but performance. The problem with animal tracking is animals tend to prefer cover, under trees, brush, areas that are not usually GPS friendly, it is the reason animal tracking in the past has used RF at low, sub 100Mhz frequencies. 

Off the shelf modules are not going to be ideal for animal tracking because they are designed with different criteria involved, mainly functionality in a human world where GPS data is easily obtained.
The best results will be from approaching a manufacturer with what you need and they can tailor the design better.

Most animal trackers using GPS are cell phone hybrids, using the towers for the accurate clock source and for data transfer. Tagg and Garmin have products for pet tracking .

http://www.lotek.com/gps.htm

[Psi]

A gps module with helical antenna would work better for random orientations.
The ceramic antennas tend to fail if signal is marginal and you tilt them away from the sky.

[ve7xen]

I've used the SkyLab SKG13C (available on alibaba) module in a couple projects. You add the antenna and antenna power supply, but that's basically all you need to do yourself. Cheap and easy to use. I believe it's based on a MediaTek MT33xx, but it must be an output-only version or something as it doesn't respond to commands for that chip. Seems pretty sensitive.

[8086]

Thanks for replies.

I realise I need to consider performance, but cost is more of an issue.

When I was saying to put together chipset + amp + ant I meant using an already available chip.

I have a new plan. Basically this plus this.

Thoughts? Should I get a small amp in there too or is there not really any point?

[ptricks]

Thoughts? Should I get a small amp in there too or is there not really any point?

I would start with the fewest components possible and see how that works out . Testing it in the different orientations and environments is going to be a big part of the work.
You might also want to look at commercial versions , maybe find some cheap broken ones on ebay to get some ideas from.
What type of animals are considering for tracking ?

[8086]

Thoughts? Should I get a small amp in there too or is there not really any point?

I would start with the fewest components possible and see how that works out . Testing it in the different orientations and environments is going to be a big part of the work.
You might also want to look at commercial versions , maybe find some cheap broken ones on ebay to get some ideas from.
What type of animals are considering for tracking ?

Yeah, I'll give it a go without the amp and see how it goes then.

Cats.