Trying to figure out the cause of my Yaw error

So, I have about 94 flights in, and this is only the second time I received this error… and I was able to fly back and land with no issue, but I did one thing I dont usually do. If you look at the log

you will see that there is a steel roofed barn in the image. I walked quite close to this after taking off a good distance from it. The drone never got near it, but the controller did. (the snow was very deep so I was going where we had cleared it, which was close to the barn).

Any thoughts?

you lose the gps signal at the same time as the yaw error…
could this be the cause of it?
it has to use its vps where there is no gps signal…

1 Like

I see the warning, but its odd that with 19 sats it could be weak. Plus I fly over that spot regularly with no issues. At that altitude (approx 45 feet) would the VPS still even work? Anyway, thanks for the suggestion, its still unclear to me what happened though.

@Northwood…Yaw Errors are usually compass related. But there can be other factors involved. I don’t think the barn had anything to do with your issue and it was not long in durartion. The reason you received the weak GPS warning was due to the NAV Health. Not the number of satellites acquired. This is part of the internal calculations that the aircraft makes. The Yaw error was short in duration. You may consider a cold calibration of the IMU.

Notice the Nav Health at the start of the Yaw Error below:

@bob…The aircraft uses the barometer to maintain altitude without GPS.

1 Like

good point…
sorry for the misleading information.

@bob…Not really misleading information, it does have limited effect. Notice the VPS is inactive before the Yaw errors at altitude. It is functioning properly.

ok @Fly_Dawg
just thought…
vps would not be working at that height any way would it?
dji says inspire vps only works below 30 feet.

Thanks all. If you take a look at my YT channel, I have another flight there from last weekend which was error free, but will show you how much snow we have. Once that clears, I am heading out to the middle of one of my fields and going to re-calibrate, going to get the wife to assist, so that even the iPhone is nowhere near the aircraft. (I wont go as far as getting naked, but will leave my belt with her! I dont wear watches or other jewellery) so that my calibration is as clean as earthly possible. Right now we still have 2 feet of snow to deal with. Cheers… Flight 92 - Early Spring Pond Visit -no audio - YouTube

1 Like

It shouldn’t, but I have seen a lot of strange things with the VPS. If there is a gimbal guard below the sensors it might send the aircraft up until the 500 meter mark. This has happened. I never use it anyway. I find it of very little use, quite honestly.

yep only any good for indoor flying…
or on the inspire it lowers the landing gear when you get close to the ground - useful.

Futher to the great advice I got above, some history and then a question again… I calibrated once before flight #1 with my Spark, it was on June 1, 2018. Since then, all has been well except for one or two glitches, the second last one started me wondering. Yesterday, I decided to walk out into the middle of my front field, which is pretty large, and no trees anywhere near, so i could, for the first time, check out a quickshot mode, specifically the Helix. So it seemed to go well, there ws not much wind and the flight behaviour was as expected. Once the quickshot was complete however, the drone suddenly went into ATTI mode and I got a “poor GPS” warning. I immediately flew the Spark back to my location and hand landed. So, next chance I have (I was out of time and battery by then) I am going to do the calibration, and now for the question after this long lead in (sorry)… As I have heard conflicting stories…

I use a TilePro velcro-ed to the bottom of my battery on every flight. I am now assuming, from what I have been reading, that I should actually calibrate with it attached.

What say you experts?

Since many trackers contain metallic metal parts, they could negatively affect the Spark’s compass if mounted too closely to it. If your tracker does negatively affect the compass, calibrating the compass with the tracker attached will resolve that issue since it’ll essentially make the tracker part of the aircraft. After that point, you’d need to either keep the tracker attached or ensure it’s in the same location if you remove an reattach it.

Not sure if your tracker is negatively affecting the compass? To check, sit the Spark on the ground, power it on (without the tracker attached), wait until it’s ready to fly, and then set the tracker in the location where it will be attached. If you see #3 or #5 from this thread, then your tracker definitely needs to be attached to the Spark before calibrating the compass.

Thanks for the clarification. What about when calibrating the IMU?

The IMU is not negatively affected by metal objects. When calibrating the IMU, you should:

  • Wait for the Spark to cool down before starting the calibration and do it indoors in a cool location. Since the IMU needs to warm up to the temperature at which it was last calibrated before you can fly, you don’t want to calibrate it while the Spark is warm/hot.

  • Do the calibration on a level surface (any table will be fine).

  • Ensure the Spark is not disturbed while sitting still in each step of the calibration process.

Thanks Mike, well, I just finished the calibration, as you suggested, cool environment, 100% level table top on the middle floor of the three floor house I am in. Before I started, the Go app indicated I was in Atti mode, and no satellites were visible. After the IMU calibration, whilst the Spark was powered up in the same spot I performed the IMU cal, I then had 15 sats… wow, ready to fly in GPS mode!! So, I am wondering what I will find when I get the compass done also,feeling like it should be rock solid. (well, as best as can be expected!) Could it have been way off for a while I am pondering.

As for your compass suggestions re the tile tracker, I took advantage of the opportunity to test with and without the tile, whilst watching the sensors… no change. So I expect the tile does not interfere. its about 70% plastic and some sort of white metal edge. Might even be aluminum. (which would be better!)

1 Like