g-sensor not calibrated

Any discussion on using LapTimer. Please use this forum in case you need guidance on how to use LapTimer or perform a certain operation
KiSwiSje
20 or more Posts ★★★
20 or more Posts ★★★
Posts: 39
Joined: Sat Jul 14, 2012 5:25 pm

g-sensor not calibrated

Post by KiSwiSje »

i think the g-sensor is not calibrated correctly anymore
in v15 - when you tapped video - it always wanted confirmation for the display orientation

in v16 this is different - maybe the landscape view of the menus is interfering

remember - a motorcycle is at an angle when parked
but for the sake of the videos it is mounted horizontal - maybe use that as a default setting?
or have a calibration menu to use once per trackday or something?

no response needed - just passing the info so you can fix this in a next version
User avatar
Harry
Site Admin
Site Admin
Posts: 9985
Joined: Sun Sep 12, 2010 10:32 am
Location: Siegum, Germany
Contact:

Re: g-sensor not calibrated

Post by Harry »

Yes, v16 is different. V15 had two situations it required a re-calibration: when navigating to Video, and when coming back. The reason was it had Video landscape and everything else portrait. LT 16 allows several views to be operated in landscape and portrait - and some in portrait only. So when it is rotated, LapTimer will request a re-calibration. In case you come from a landscape view (like Timer), and switch to Video, it does not need a recalibration any more as it is calibrated landscape already.

- Harry
Image Image Image Image
KiSwiSje
20 or more Posts ★★★
20 or more Posts ★★★
Posts: 39
Joined: Sat Jul 14, 2012 5:25 pm

Re: g-sensor not calibrated

Post by KiSwiSje »

something went completely wrong as the g-force overlay is totally off to one side - as if it was calibrated on the side stand
i can provide the link to youtube when i get it uploaded - if you wish
User avatar
Harry
Site Admin
Site Admin
Posts: 9985
Joined: Sun Sep 12, 2010 10:32 am
Location: Siegum, Germany
Contact:

g-sensor not calibrated

Post by Harry »

In case you see calibration is wrong while in video view, just tap the calibration circle to initiate a new calibration. For existing laps, you may recalculate acceleration from GPS (see Lap Details).

Harry
Image Image Image Image
tch_popeye
Fewer than 10 Posts
Fewer than 10 Posts
Posts: 3
Joined: Wed Aug 22, 2012 12:47 pm

Re: g-sensor not calibrated

Post by tch_popeye »

I'm wondering about the lateral g calculation...?

If driving around a turn that is completely flat, then a GPS calculation could be expected to determine the lateral g's.

However, if the turn is cambered, the GPS-based calculation will be incorrect. Consider, that if the turn is steeply banked, that at the correct speed one's tires will not see any lateral component of load whatsoever. From my point of view, I would like to know the lateral force/effort actually exerted by the tires, before and after different types of upgrades - and I think that this cannot be accomplished with GPS data unless the tilt of the road surface is included.

Comment?

And thanks!
lunat1ck
20 or more Posts ★★★
20 or more Posts ★★★
Posts: 42
Joined: Thu Apr 19, 2012 4:01 pm
Location: New York, NY, USA
Contact:

Re: g-sensor not calibrated

Post by lunat1ck »

You are entirely correct that the lateral G-force calculated from GPS positions, or measured by the iPhone accelerometer, does NOT represent the force exerted by the tires. That value is simply the lateral G-force acceleration of the car, and you simply shouldn't interpret it any other way.

I suspect you would have to invest in a much more expensive data collection system, one that includes remote sensors on the chassis to measure it's movements, before you would be able to make this calculation, because it requires knowing the relative forces applied by each tire on the car, the various angles of the suspension components under pressure, AND the camber of the road as well.

Another way of thinking about this is that the G-force calculation you get from GPS or the accelerometer are the "net" G-forces the car is experiencing. if you're looking to understand how a change in suspension, tires, etc alters the handling of your car, you can still get an objective measurement by comparing these net G-forces before and after the change. While you won't be calculating the absolute force difference you get from changing from a treadware 300 to a treadware 200 tire, for example, is that really important to know? Isn't the overall net change in handling of the car what you're after anyway?

I did some searching, since I'm an engineer, and trying to learn more about race car data acquisition and analysis for my own benefit, and I can't find anything that discusses the G-forces of the tires specifically, separate from the overall G-force limit of the handling of the car. For example:

http://www.modified.com/news/0708_sccp_ ... ewall.html

That article shows how a number of cars were setup to maximum cornering G-forces, but the only number they give is the "net" G-force the cars were able to handle. I don't think it's practical to measure the tire's G-forces specifically, again unless you invest in a $10-20K data acquisition system, and then do a lot of complex calculations that include the geometry of the road. I doubt we'll ever see that kind of functionality in an iPhone app, but maybe Harry will impress us all (again :-P).
tch_popeye
Fewer than 10 Posts
Fewer than 10 Posts
Posts: 3
Joined: Wed Aug 22, 2012 12:47 pm

Re: g-sensor not calibrated

Post by tch_popeye »

Thanks for your input, lunat1ck.

My sloppy choice of words had you give a more considered response than was actually necessary. I'm also an engineer, which is why I'm interested in this topic. To clarify my earlier post, I'm not actually looking for the force exerted at each wheel, lateral or otherwise, but am interested only in the aggregate (net) result. However, I WOULD like to separate the contributions to lateral grip, i.e. how much of a given turn was resisted by the net friction grip of the tires (under influence of the suspension, of course), as opposed to also including the component of normal-acting force which was contributed by camber of the road surface.

You're correct, too, that I'll get a general indication with simple GPS-based calculations of relative improvements from an upgrade to the car. But to make the most sense of that data, I'll need to ensure that I repeat all measurements for comparison on the same piece of track/road/parking lot/whatever. This is because I cannot separate the camber-supplied cornering force with only GPS data, for a meaningful comparison between dissimilar stretches of asphalt.

But I'm nitpicking. I like to try and determine the best option during my research phase, and back away from it slowly to save money, all the while being aware of what each degree of improvement will cost. It helps me find my preferred balance of value-for-money.

By the way - the link you posted was great. Thanks so much for your post.

-Trev
User avatar
Harry
Site Admin
Site Admin
Posts: 9985
Joined: Sun Sep 12, 2010 10:32 am
Location: Siegum, Germany
Contact:

Re: g-sensor not calibrated

Post by Harry »

tch_popeye wrote:I'm wondering about the lateral g calculation...?

If driving around a turn that is completely flat, then a GPS calculation could be expected to determine the lateral g's.

However, if the turn is cambered, the GPS-based calculation will be incorrect. Consider, that if the turn is steeply banked, that at the correct speed one's tires will not see any lateral component of load whatsoever. From my point of view, I would like to know the lateral force/effort actually exerted by the tires, before and after different types of upgrades - and I think that this cannot be accomplished with GPS data unless the tilt of the road surface is included.
Neither GPS calculated nor measured acceleration is 'correct' if the track is not flat. On the later, just think about a motorbike - here you get lateral acceleration of 0 as long as calibration has been done on flat surface. But to comment on a less extreme level, if the track is not flat, part of the lateral acceleration will be measured as down force and will miss in the lateral component.

The GPS based calculation is nice as long as the radius between two fixes LapTimer assumes matches reality. What LT does when calculating is checking direction driven in each GPS plot, deriving a corner radius driven, applying speed to the formula and calculating the centrifugal force.

- Harry
Image Image Image Image
tch_popeye
Fewer than 10 Posts
Fewer than 10 Posts
Posts: 3
Joined: Wed Aug 22, 2012 12:47 pm

Re: g-sensor not calibrated

Post by tch_popeye »

Thanks Harry.

I get the point. In my naivete, I had hoped for a simpler quantification of performance. Imagine lapping a track, and taking each of 14 turns at the very limit of adhesion, suspension well settled, everything (much too) perfect. In my mind, I had hoped to be able to see 14 briefly sustained peaks of very similar lateral g, and thus conclude the exercise. In reality, I will see 14 very different levels of lateral acceleration, whether from GPS-calculated data, or even from accelerometers, due to the camber of each turn among other things.

I suppose it's these nuances that keeps it interesting for inquiring minds... ;)

Regards,

-Trevor
Mgbond
Fewer than 10 Posts
Fewer than 10 Posts
Posts: 9
Joined: Thu Feb 06, 2020 11:54 pm

Re: g-sensor not calibrated

Post by Mgbond »

Hi, I used the app for the first time and it was very good.
The only issue is that on the overlay the G meter was showing the G forces the wrong way round. So if I was going round a left bend the G meter was over to the left instead of the right. Any ideas?
Post Reply