Carloop Pro GPS time to signal lock

[lucas]

Hello,

I got the Carloop pro basic application to work and connected via USB I can see the data (see below).

The "HASGPS" was something I added just to see if GPS was enabled on the device. After being plugged in with clear line of site to the sky with no clouds for 20 minutes there is still no lat/lng is outputted.

Is there anything I can do to speed up the time to lock? The timestamp begins to display after 20-30 seconds, which makes me think that there is some signal being received. It did work briefly the other night, but I cannot get it to lock on since then.

I've tried moving the device so the side with the label is facing up and also facing down. I've tried it with the battery + usb plugged in and also with just the usb plugged in. I can't seem to get the GPS coordinates to output via USB.

Thanks

Battery voltage:     4.356387 CAN messages:            0 HASGPS 1 GPS  36549 chars: ********** *********** 10/04/2016 23:17:15
Battery voltage:     4.350586 CAN messages:            0 HASGPS 1 GPS  36768 chars: ********** *********** 10/04/2016 23:17:16
Battery voltage:     4.367988 CAN messages:            0 HASGPS 1 GPS  36987 chars: ********** *********** 10/04/2016 23:17:17
Battery voltage:     4.362187 CAN messages:            0 HASGPS 1 GPS  37206 chars: ********** *********** 10/04/2016 23:17:18
Battery voltage:     4.362187 CAN messages:            0 HASGPS 1 GPS  37425 chars: ********** *********** 10/04/2016 23:17:19
Battery voltage:     4.356387 CAN messages:            0 HASGPS 1 GPS  37517 chars: ********** *********** 10/04/2016 23:17:20
Battery voltage:     4.362187 CAN messages:            0 HASGPS 1 GPS  37863 chars: ********** *********** 10/04/2016 23:17:21
Battery voltage:     4.356387 CAN messages:            0 HASGPS 1 GPS  38082 chars: ********** *********** 10/04/2016 23:17:22
Battery voltage:     4.362187 CAN messages:            0 HASGPS 1 GPS  38301 chars: ********** *********** 10/04/2016 23:17:23
Battery voltage:     4.356387 CAN messages:            0 HASGPS 1 GPS  38528 chars: ********** *********** 10/04/2016 23:17:24

[lucas]

Quick update, it just locked, but it seems to take a full 5 minutes of running before outputting a GPS coordinate. Is there anything I can do to speed this time up?

Thanks

Battery voltage:     4.188164 CAN messages:            0 HASGPS 1 GPS   4734 chars: ********** *********** 00/00/2000 00:00:00
Battery voltage:     4.199765 CAN messages:            0 HASGPS 1 GPS   4912 chars: ********** *********** 00/00/2000 00:00:00
Battery voltage:     4.199765 CAN messages:            0 HASGPS 1 GPS   5117 chars: ********** *********** 00/00/2000 23:22:14
Battery voltage:     4.193964 CAN messages:            0 HASGPS 1 GPS   5322 chars: ********** *********** 00/00/2000 23:22:15
Battery voltage:     4.193964 CAN messages:            0 HASGPS 1 GPS   5533 chars: ********** *********** 10/04/2016 23:22:16
Battery voltage:     4.205566 CAN messages:            0 HASGPS 1 GPS   5744 chars: ********** *********** 10/04/2016 23:22:17
Battery voltage:     4.205566 CAN messages:            0 HASGPS 1 GPS   5955 chars: ********** *********** 10/04/2016 23:22:18
...
Battery voltage:     4.350586 CAN messages:            0 HASGPS 1 GPS  80890 chars: ********** *********** 10/04/2016 23:27:01
Battery voltage:     4.356387 CAN messages:            0 HASGPS 1 GPS  81196 chars: ********** *********** 10/04/2016 23:27:02
Battery voltage:     4.367988 CAN messages:            0 HASGPS 1 GPS  81502 chars: ********** *********** 10/04/2016 23:27:03
Battery voltage:     4.362187 CAN messages:            0 HASGPS 1 GPS  81808 chars: ********** *********** 10/04/2016 23:27:04
Battery voltage:     4.356387 CAN messages:            0 HASGPS 1 GPS  82225 chars:   37.778935 -122.47364310/04/2016 23:27:05
Battery voltage:     4.356387 CAN messages:            0 HASGPS 1 GPS  82642 chars:   37.778931 -122.47363510/04/2016 23:27:06
Battery voltage:     4.367988 CAN messages:            0 HASGPS 1 GPS  83047 chars:   37.778931 -122.47364310/04/2016 23:27:07
Battery voltage:     4.362187 CAN messages:            0 HASGPS 1 GPS  83452 chars:   37.778946 -122.47364310/04/2016 23:27:08
Battery voltage:     4.356387 CAN messages:            0 HASGPS 1 GPS  83857 chars:   37.778950 -122.47364310/04/2016 23:27:09

[alanm]

Hey @lucas,
Thanks for the feedback and helping us test the GPS in even more real world scenarios. We are working on potential improvements to the firmware to help with this issue. Either me or @jvanier will reply with an update.

By the way, after the GPS found the satellite fix once, did it take as long for subsequent fixes in your experience?

[lucas]

After waiting through the 5 minute period to get the initial fix, the subsequent fixes after restarting the device were acquired in ~20 seconds. Much faster!

Was it slow to get a fix because the device wasn't powered on in over a day?

[ScruffR]

Late to the party, but in order to get a proper fix, the GPS receiver needs to know the current positions of all satellites to be able to calculate its own position from the data it gets of the individual sats in sight.
That position almanach is downloaded via low bandwidth transmittion in the sat data stream and hence takes a long time to download. Once that almanach is stored and not outdated subsequent fixes go quick.

A faster way to get a fix after having the device depowered for longer is to download the almanach via a higher bandwidth connection (cellular on Electron or WiFi on Photon) - at least this is how it can be done with the Particle AssetTracker.

[jvanier]

Super informative. Let us know if you have more information. Thanks!

[TylerB]

Hey ScruffR!

Thanks for the info. I took a look at the AssetTracker code and couldn't see anywhere the almanac was being downloaded over a faster connection. I may have simply missed it.

Could you possibly point me in the right direction for that?

Thanks!

[cyclin_al1]

Another possibility is that some car windshields have a micro metal film over the surface to reduce glare and keep some of the solar heat out of the car. This could be the cause of some shielding of the GPS signal. To test the theory, try getting a GPS lock with the GPS antenna outside of the car and see if the lock time improves significantly.

[ScruffR]

I'll have to dig up that respective post where one of the Particle gurus had described who he's done that.
I don't think it's in the stock AssetTracker stock repo but rather in a repo called Fancy AssetTracker (or so)

---

Yup, that was the name. It was a project of Dave Middlecamp
https://github.com/dmiddlecamp/fancy-asset-tracker

But as it seems, the respective ephemeris-loader sample isn't in there.
Maybe you can ping/bump @Dave on the Particle forum in the respective thread

Adafruit GPS Photon problem

[TylerB]

Using an external antenna mounted to the roof significantly improves performance, but I am really hoping to keep the unit self contained, if possible.

I may end up having to go with an external antenna. Does anyone know of one that plugs right in to the 6 pin JST on the carloop board?

[TylerB]

Thanks for the info! I'll dive in to this and see what I can figure out!

[kogs]

I just received my Carloop Pro in the mail today, and after running it for several hours - inside, on the front porch, and finally for a couple hours in the back yard with a clear view of the sky - I'm unable to get a GPS fix. My results are basically the same as Lucas' first post, but I don't even get a time signal from the GPS unit. Here's an example of the output:

It's a revision 2.3.

Thanks!
-Matt

[alanm]

Hey @kogs,
I shipped a new GPS and cable your way. I suspect it is a bad cable but keep us posted on this community post so everyone can benefit from what you learn.

Alan

[Affarram]

Im having the same issue as @kogs. I let it run for a pretty long time and unplugged it a few times to see if that would help. I finally decided to read the data directly from the gps unit using Serial1
if (Serial1.available()) {
Serial.write(Serial1.read());
}
This is what my output looks like...
$GPRMC,035129.00,V,,,,,,,151216,,,N*71
$GPVTG,,,,,,,,,N*30
$GPGGA,035129.00,,,,,0,00,99.99,,,,,,*6A
$GPGSA,A,1,,,,,,,,,,,,,99.99,99.99,99.99*30
$GPGSV,1,1,02,03,,,21,32,,,20*78
$GPGLL,,,,,035129.00,V,N*46

As far as i can tell by looking at the datasheet it is not acquiring a signal from enough satellites to determine my location. Is there something I can try to get it to work better?

[jmb]

Sadly I seem to be having the same problem described above. It ran all afternoon today with clear access to the sky, but no GPS fix.

[jmb]

Help? Who do I contact for a replacement? I've tried everything. I think I got a dud GPS module. It's quite frustrating as the applications I had in mind are all critical on using position data! So I'm a bit stuck...

[alanm]

Hey @jmb that's a bummer! You can send us your GPS module and we will send you a different one. We have had really bad results with this GPS version and we are going to be announcing our next GPS version shortly. Our initial tests are really good but we want to make enough units to provide anyone that wants to upgrade. Please send me an e-mail at alan@carloop.io so I can send you instructions on replacement of the current version.

[pawelmanu]

Hi Team, what's the proceeding if I have rev 2.3 and I also have problems to lock GPS signal?
Any tricks? Or should I replace the module?

[cyclin_al1]

@pawelmanu,

Have you not gotten a GPS lock ever, or is it difficult to get a lock?
Try it out somewhere with a clear view of the sky, with no signal shields.
Unfortunately, a metal car roof or a metal solar film on a windshield make good signal shields...

In case there is a hardware issue, @alanm and @jvanier should be aware you are having difficulty.

[pawelmanu]

Is there some option to test GPS module directly without using Electron board? I am not getting GPS fix following open-sky prerequisite (by using carloop_minimal example). Maybe there is an error for GPS handling once device is in Europe