Noisy Lat and Long G data

Post Reply
Grabla66
Posts: 66
Joined: Tue Oct 25, 2016 7:01 am
Location: Fordshire
Contact:

Noisy Lat and Long G data

Post by Grabla66 »

Is this normal, for the G force to jump around as much as this?
Or can the data be smoothed?
I'm using the 10Hz GPS module
Attachments
noisey g force.png
noisey g force.png (118.18 KiB) Viewed 2483 times

RICE RACING
Posts: 214
Joined: Mon Feb 24, 2014 1:08 am

Re: Noisy Lat and Long G data

Post by RICE RACING »

Yes It's pretty normal.

Seen it on the unit you have as well as the LR AG50 job, which I have bought all of them and run in various cars.

The accel sensors are logged at 100Hz on both and are different to the GPS component of the unit you have.
They are fairly accurate < I base that off using them and comparing to my Race Logic VBOX3i unit which is running a RL IMU (read mega expensive!). I did a test once where I was consulting back to Jason English (no longer at LR) and validating some of their A10 units performance. The GPS is a bit how you going but the accel part of it is good especially given the price point.

If I could be stuffed I can put up some RaceLogic data showing high dynamics non filtered derived Lat and Long G from velocity V's the IMU's fitted accel sensors and as long as its flat land test with no bumps or potholes etc or GPS interference like a tree etc, then at 100Hz with no filtering you can overlay the two of them perfectly.

Keep in mind the accel data as recorded by a physical sensor, so when run on bumpy tracks or any type of camber or pitch change will give corresponding changes to the actual true acceleration figure when its derived from velocity changes.

Typically spikes you will see are from just those kinds of impacts against the sensor itself, and yes you can filter it, but you will also then loose out some information like when to tune traction control and cut severity is too high, this on an unfiltered signal will always show up.

Hope this helps you.
http://www.riceracing.com.au
Real turbo road cars fast > reliable > durable
Water Injection Specialist

RICE RACING
Posts: 214
Joined: Mon Feb 24, 2014 1:08 am

Re: Noisy Lat and Long G data

Post by RICE RACING »

Here is the pictures to show what I mean,
air born at top of 4th gear (bump in run way), both correlate (derived accel from velocity changes) and IMU accel sensor, only shown long g to keep it simple, as same effects happen in other axis.
Also did a report for you showing averages between 5kph intervals as well as on the speed line (non filtered value).
Attachments
yo3.jpg
yo3.jpg (608.96 KiB) Viewed 2456 times
yo2.jpg
yo2.jpg (726.12 KiB) Viewed 2456 times
yo.jpg
yo.jpg (736.96 KiB) Viewed 2456 times
http://www.riceracing.com.au
Real turbo road cars fast > reliable > durable
Water Injection Specialist

RICE RACING
Posts: 214
Joined: Mon Feb 24, 2014 1:08 am

Re: Noisy Lat and Long G data

Post by RICE RACING »

more
Attachments
yo4.jpg
yo4.jpg (607.82 KiB) Viewed 2456 times
yo6.jpg
yo6.jpg (567.06 KiB) Viewed 2456 times
yo5.jpg
yo5.jpg (587.24 KiB) Viewed 2456 times
http://www.riceracing.com.au
Real turbo road cars fast > reliable > durable
Water Injection Specialist

RICE RACING
Posts: 214
Joined: Mon Feb 24, 2014 1:08 am

Re: Noisy Lat and Long G data

Post by RICE RACING »

last one
Attachments
yo7.jpg
yo7.jpg (605.17 KiB) Viewed 2456 times
http://www.riceracing.com.au
Real turbo road cars fast > reliable > durable
Water Injection Specialist

Grabla66
Posts: 66
Joined: Tue Oct 25, 2016 7:01 am
Location: Fordshire
Contact:

Re: Noisy Lat and Long G data

Post by Grabla66 »

Yes, very useful thanks. My GPS module is cable tied to the chassis directly above the front dampers, I hadn't realised that bumps of course have an effect. I wonder if I could mount it on something more compliant.

RICE RACING
Posts: 214
Joined: Mon Feb 24, 2014 1:08 am

Re: Noisy Lat and Long G data

Post by RICE RACING »

Grabla66 wrote:
Mon Apr 29, 2019 10:34 am
Yes, very useful thanks. My GPS module is cable tied to the chassis directly above the front dampers, I hadn't realised that bumps of course have an effect. I wonder if I could mount it on something more compliant.
In the example I gave fyi, the 100Hz GPS (derived accel data) antenna on roof, the IMU (100Hz accel data) in center of car, velcro/dual lock type fixing.
You don't really want the IMU dampened as such cause its reflecting everything that is happening to the chassis (bumps included).

Can use the filter in software or a maths channel to clean it up as you see fit, just keep in mind on 100Hz you will loose the instant detail of things like trq reductions for example on traction control tuning.
http://www.riceracing.com.au
Real turbo road cars fast > reliable > durable
Water Injection Specialist

Post Reply