Inconsistent performance on ZED-F9P breakouts

GPS/GNSS Development Boards
1

I bought two ZED-F9P SMA Breakout boards (https://www.sparkfun.com/products/16481) along with two SMA cables (https://www.sparkfun.com/products/12861) and two helical antennas (https://www.sparkfun.com/products/17383).

I have them sitting next to one another on a desk as shown in the attached image.

7F274F12-63F0-4000-B8EB-245143167ACE.jpeg
7F274F12-63F0-4000-B8EB-245143167ACE.jpeg4032×3024 2.7 MB

One gets a GNSS fix quite quickly - on first boot up it was less than 1 min - and it sees 4-9 satellites on all of the constellations (GPS, GLONASS, Galileo, and BeiDou). The position is relatively stable, shifting 10 or so feet around from time to time.

The other took over 5 min to get a lock, sees 3 GPS satellites, 4 GLONASS satellites, and none from the other constellations. The position also varies by hundreds of feet.

I’ve swapped the SMA cables and antennas between the two with no change in performance.

Am I missing something? Is this expected variability? Or might I have a defective unit?

Thanks.

2

Try taking them outside and repeat your test…

You will get highly variable results indoors…

Best wishes,

Paul

3

Also, those antennas work best when mounted vertically (SMA connection at the bottom).

Have fun!

Paul

4

Paul,

Thanks. I moved outside and I still have the same issue. One can’t see any Galileo or BeiDou satellites and has significant position error. The other behaves properly.

Any additional ideas?

5

Are both antennas vertical? Try separating them too - so one is not shadowing the other.

Best wishes,

Paul

6

Sorry, yes. I also reoriented them both vertically and they’re now 9 in apart mounted on top of a cardboard box.

7

It’s probably also worth mentioning that it’s consistent which unit works well and which one doesn’t. I can swap antennas, antenna cables, USB cables, power one but not the other, move them around etc. The one that doesn’t behave as I expect is always the same.

8

OK. Please have a close look at the solder joint on the central pin of the SMA connector on both boards. Do you see a difference? (I’m wondering if you have a semi-dry / intermittent joint on the ‘poor’ one?)

Best wishes,

Paul

9

Or any stray solder on any of the three small components between the SMA connector and the ZED package?

10

I don’t see any obvious difference on the SMA solder joint nor on the passives next to it.

Over time, the positions seem to have converged, but there’s still a marked difference in the satellites I can see. I tweaked my code for more fidelity on the GSV message. The GPS and QZSS signals it sees on all the bands are +/- 1 from one another (12 L1, 7 L2, 5 QZSS) . However, on Galileo, I see 10/9 sats on one (the different bands) and 1/1 sats on the other. GLONASS is 8/7 vs 4/6, and BeiDou is 7/3 vs 3/3.

11

I haven’t actually configured anything yet. I guess I made the assumption they came with a standard configuration. They look brand new and they came out of sealed plastic. Could there be some configuration difference between them?

12

u-center has a spectrum analyzer feature that may shed more light on this.

Are both boards running the same firmware? UBX-MON-VER will provide more details.

Did you buy the boards direct from SparkFun? We can arrange an exchange if required. Let us know.

Best wishes,

Paul

13

Ok, thanks. Yes, I got them both from SparkFun just a couple of weeks ago. This was the first chance I’ve had to plug them in.

Both firmwares are the same. No u-center for me, I don’t have a Windows machine anywhere. But that seems like a handy feature for future me. Maybe I’ll mess with a VM or Wine.

I’ll go through and swap things around again just to make sure I haven’t made a mistake and then I’ll reach out for a replacement.

Thanks for double checking me and for the quick responses. I’ve been getting stuff from SparkFun for 15+ years and you never disappoint.