G force readings backwards... How to prevent?

Help on issues you run into with LapTimer; in case you have a question on how to use LapTimer, use the forum "Using LapTimer" instead
AEsco48
10 or more Posts ★
10 or more Posts ★
Posts: 18
Joined: Mon Mar 17, 2014 1:06 am

G force readings backwards... How to prevent?

Post by AEsco48 » Tue Apr 22, 2014 3:15 pm

Hello,

A few weeks ago I used HLT for the 1st time and I guess at some point in my setting the app up and then laying the phone down inside my center consul (external GPS) the orientation got flipped... On a left hand turn it shows the dot on the G circle the the left...

How do I prevent this from happening. Getting the phone into the center consul cradle I made requires it to tilt around a few times before finally resting.

http://youtu.be/sGax-JdhIXg

Thx!
User avatar
Harry
Site Admin
Site Admin
Posts: 9581
Joined: Sun Sep 12, 2010 10:32 am
Location: Siegum, Germany
Contact:

Re: G force readings backwards... How to prevent?

Post by Harry » Tue Apr 22, 2014 3:21 pm

To prevent this from happening, please make sure the phone is not mounted flat. Furthermore it must not be rotated at all while already driving. LapTimer blocks any rotation (i.e. re-orientation) while driving to not flip when driving though corners hard...

Good news is you can repair your acceleration values by re-calculating the values from GPS. Check the lap's Lap Details view for the Calculate button...

- Harry
Image Image Image Image
User avatar
Harry
Site Admin
Site Admin
Posts: 9581
Joined: Sun Sep 12, 2010 10:32 am
Location: Siegum, Germany
Contact:

Re: G force readings backwards... How to prevent?

Post by Harry » Tue Apr 22, 2014 3:24 pm

P.S. Looking into your video, please make sure you have not ticked the "Reverse Lateral Acceleration" Expert Setting. It should be used in case the iPhone points to the back of the car only.
Image Image Image Image
AEsco48
10 or more Posts ★
10 or more Posts ★
Posts: 18
Joined: Mon Mar 17, 2014 1:06 am

Re: G force readings backwards... How to prevent?

Post by AEsco48 » Tue Apr 22, 2014 3:35 pm

Ok great! How much of a tilt should it have? I can modify the cradle...

Actually the phone was flat and the top of the iPhone was pointing towards the back of the car and "Reverse Lateral Accl" was not switched on. I mounted it this way due to the way the phone fits in the center console with the power connector coming out of the bottom of the phone. I think I can mount it with a tilt and with the top of the phone pointed forwards. But in case I cant, if I tilt the phone and select "Reverse Lateral Acceleration" it will solve the lateral accel flip but not the front/back flip, right?

Thx!
User avatar
Harry
Site Admin
Site Admin
Posts: 9581
Joined: Sun Sep 12, 2010 10:32 am
Location: Siegum, Germany
Contact:

Re: G force readings backwards... How to prevent?

Post by Harry » Tue Apr 22, 2014 4:20 pm

For an iPhone nearly flat, your video looks remarkable good ;-) There is no absolute value I can name for tilt. 0 tilt will not work, everything else work in theory. The less tilt you have, the bigger are inaccuracies introduced from the Z axis to all other axis. I'd suggest to have it at least 30 degree up. "Reverse Lateral Acceleration" works when the iPhone points to the back with an reasonable tilt.

- Harry
Image Image Image Image
gplracerx
20 or more Posts ★★★
20 or more Posts ★★★
Posts: 836
Joined: Thu May 03, 2012 5:26 am
Location: Kingsport, TN USA

Re: G force readings backwards... How to prevent?

Post by gplracerx » Tue Apr 22, 2014 9:50 pm

The less tilt you have, the bigger are inaccuracies introduced from the Z axis to all other axis.
There are ways around that. If you used quaternions or Direction Cosine Matrices, the device orientation wouldn't matter, as long as it stayed fixed relative to the vehicle axes. If DCM can be implemented on an Arduino for real time drone flight control where you have to deal with cross winds and banking when turning, it should, in principle, be trivial to do on an iPhone for a car. Or just redefine the Z axis as the device axis closest to being parallel to the gravity vector when stopped.

Using a DCM would also be useful for roll correction on a banked track. I'm betting there's a big difference between the GPS lateral acceleration and the accelerometer readings when going around the Karussell's on the Nordschliefe. Lapping at a 70 MPH average at Bristol Motor Speedway and defining turning left as positive lateral acceleration, I was seeing apparent negative lateral acceleration from the banking on the straighter parts of the oval because I wasn't going fast enough. I'll have to try harder this year.

I understand that this is not a priority for the majority of your customers. I'm just saying that there are solutions.

Off Topic: I played around with the iPhone magnetometer. It's useless for navigation because, for some reason, there's too much hysteresis. There was a 25 degree difference when turning counter clockwise vs. turning clockwise. Probably not coincidentally, that's the stated error in the magnetometer heading.
AEsco48
10 or more Posts ★
10 or more Posts ★
Posts: 18
Joined: Mon Mar 17, 2014 1:06 am

Re: G force readings backwards... How to prevent?

Post by AEsco48 » Tue Apr 22, 2014 9:54 pm

Harry wrote:For an iPhone nearly flat, your video looks remarkable good ;-) There is no absolute value I can name for tilt. 0 tilt will not work, everything else work in theory. The less tilt you have, the bigger are inaccuracies introduced from the Z axis to all other axis. I'd suggest to have it at least 30 degree up. "Reverse Lateral Acceleration" works when the iPhone points to the back with an reasonable tilt.

- Harry

Great, thx! I am guessing it does not matter if its the top or bottom of the phone that is tilted up?
38F/62R
20 or more Posts ★★★
20 or more Posts ★★★
Posts: 20
Joined: Wed Jul 31, 2013 10:29 pm

Re: G force readings backwards... How to prevent?

Post by 38F/62R » Thu Apr 24, 2014 11:59 pm

AEsco48 wrote:Hello,

A few weeks ago I used HLT for the 1st time and I guess at some point in my setting the app up and then laying the phone down inside my center consul (external GPS) the orientation got flipped... On a left hand turn it shows the dot on the G circle the the left...

How do I prevent this from happening. Getting the phone into the center consul cradle I made requires it to tilt around a few times before finally resting.

http://youtu.be/sGax-JdhIXg

Thx!
Aesco48, nice video. Which version of phone do you use and how do you mount it? I mount my phone to the roll bar so I can see my hands but there is so much vibration as I go to higher rpm's that the video is not much good.
jwmelvin
20 or more Posts ★★★
20 or more Posts ★★★
Posts: 53
Joined: Fri Oct 18, 2013 1:28 pm

Re: G force readings backwards... How to prevent?

Post by jwmelvin » Fri Apr 25, 2014 12:52 am

38F/62R wrote: Aesco48, nice video. Which version of phone do you use and how do you mount it? I mount my phone to the roll bar so I can see my hands but there is so much vibration as I go to higher rpm's that the video is not much good.
I use a Ram mount on my harness bar. It gets some vibration and lots of wind noise, as you can see here:
http://youtu.be/MaTBBwu1JV0
gplracerx
20 or more Posts ★★★
20 or more Posts ★★★
Posts: 836
Joined: Thu May 03, 2012 5:26 am
Location: Kingsport, TN USA

Re: G force readings backwards... How to prevent?

Post by gplracerx » Fri Apr 25, 2014 2:29 pm

38F/62R,

It sounds like the phone is vibrating in the cradle. Have you tried putting some padding in the system to dampen the vibrations? I've used adhesive backed moleskin foot pads on the cradle itself. You could also try putting a layer of closed cell rubber foam between the clamp and the bar.
Post Reply