A progress report: the additional batteries are in place. I replaced all twelve of the batteries with the black/gray Imedions (they are my favorites).
The GPS module was laid into the backshell. I tested the TX output of the module, and its voltage was fluctuating in that way that says “data is here!” I soldered the wires for power, RX, TX, and GND to the Rpi2, fired the beast up, and dumped the serial port:
cat /dev/ttyAMA0 gpscat /dev/ttyAMA0
Plenty of output appeared, but no satellite fix. So, today I’ll take the unit out to a clear spot and see if I can get a fix. They say, for this unit, the “first fix” (populating the almanac) can take 1-3 hours. Stay tuned…
Did you stay tuned? The open area shown in the figures allowed the GPS unit to “see” the sky well enough such that the cgps program screen indicated a SNR (signal to noise) greater than zero for 11 satellites! Unfortunately, after two and a half hours, there was still no first fix. I played around in the open sky area for about five and a half hours total (that’s how long those 12 AA NiMH batteries lasted) and then went home, a little discouraged.
I noted that during the entire exercise, cgps never indicated an SNR (signal to noise) of more than about 29. This, I found from internet lore, was pretty marginal for a first fix. More data is downloaded on a first fix, hence a better signal is needed. Others with the same brand GPS device were seeing 35-40 SNR on their screens, so what was up with mine?
I decided to look at the antenna. The antenna was physically close to both an NiMH battery, a large HDMI ribbon cable, and the GPS board itself. I moved the antenna away from the battery, and retried. Still – no dice. I moved it away from everything, and hung it over the edge of the enclosure, outside of the box altogether. Now cgps rang up a SNR of 35-40, depending on satellite. Wow! The antenna placement is *very* critical. Immediately after hanging the GPS active antenna on the side of the enclosure (well, within about 30 seconds), I had a fix.
Then I wondered, “If the first fix requires a better signal, will the GPS device work in regular mode when bolted inside of the box?” Yup, once the first fix is obtained, the thing works inside of the box with only four or five satellites showing, while the Raspy-Riscy is inside of my house. Inside of the house, I see SNRs of only 20-25, but it still works (as long as the first fix is already in the bag). So, get a first fix, bolt the device into the enclosure, and forget about it. Right? Well, reading a more in-depth brochure from the manufacturer suggested that this will not work. The device, when it boots, uses a cold, warm, or hot boot and satellite management algorithm. Cold boot is basically a “first fix scenario” – while hot boot is within four hours, and takes basic info from RAM memory (no problem) – and warm boot is somewhere in-between. I think the “in-between” mode uses some, but not all, of the previous GPS info. After four days, no matter what, it’ll do another cold boot. Drat!
I am having a dickens of a time finding a place inside of the enclosure, where it’ll do a first fix. This is something I did not anticipate! Stay tuned, while I sort it out.
Meanwhile, some internet cruising let me in on some other neat tricks to do with the GPS device. GPS gets time signals from satellites, so you can set your PC to the (somewhat) correct time by using the GPS. This means that on a boat, for instance, where you do not have internet access, you can get an approximate time with most any GPS. More expensive GPS devices allow for better time if they have a special time output pin (mine doesn’t).
So, a couple things I’ve put into the notebook. 1) – Having a bunch of satellites in view means nothing, if your GPS + antenna arrangement cannot pull in SNRs in the range of 30-40. 2) – The antenna placement inside of any GPS powered device is *very* critical. 3) – You get sunburned if you stay out in the “open area with a good sky view” for too long.
It’s also been mentioned that excessive interference signals (from, for example, a processor board) can cause degradation of the SNR.
I had to move some power wires from the vicinity of the active antenna. They were about an inch away from the antenna, so I routed them to the other side of the box. This is what solved the problem mentioned three paragraphs back, “I am having a dickens of a time“:
The power wires were affixed to the lid of the enclosure. Opening the box yielded good signal, and closing the box reduced the signal, in accordance with the proximity of the wires. At least, that’s what seemed to be the problem. I am not the world’s foremost expert on 2.5 Ghz GPS active antennas, so take all this as the empirical evidence of a hobbyist. Worked for me, though.
In the figure (above) can be seen a snapshot of the cgps utility showing that the GPS can work with marginal signals if the “first fix” has already been accomplished. This is *after* moving the active antenna, and the power wires, but was taken with the Raspy-Riscy in a lower level interior room. Next to a window, or on the upper floor, I get 35+ SNR.
To be continued …
Note: the author does not have a recent, applicable background in circuit building, or battery related issues, so this is presented as the work of a hobbyist, and is not meant for duplication by others. Readers should look elsewhere for design advice and info.
Note: This author and site is not affiliated with the Raspberry Pi in any way. For information about those projects visit http://www.raspberrypi.org. “Raspberry Pi” is a trademark of the Raspberry Pi Foundation.
Imedion and PowerEx are trademarks of Maha Energy Corp. This author and site has no affiliation with them.