MoTeC Global Forum

I do have some questions on the 57205 "Single axis G Force Sensor" as well.
I couldn't find any of these characteristics:

- accuracy of measurement,
- noise density and intensity

Can you pass me these? Or who is the manufacturer of the sensor, so that I can contact him and ask.

Greetings
Julian

CAPMAGIC wrote:I do have some questions on the 57205 "Single axis G Force Sensor" as well.
I couldn't find any of these characteristics:

- accuracy of measurement,
- noise density and intensity

Can you pass me these? Or who is the manufacturer of the sensor, so that I can contact him and ask.

Greetings
Julian

No idea what you are doing exactly.
But I have found:
1) That the noise of the chassis on the road is way more than the accelerometer's noise.
In fact idling the noise in a 3-axis accelerometer is ~10dB higher than with the engine off. Rolling it is at least another 10-20 dB higher.
2) Any shifting in the chassis will move the apparent G-Force around. Braking can pitch the car more degrees than one would imagine, and chassis roll is also a lot higher than I would have guessed. To account for the dynamic chassis attitude you would need to know the roll pitch and yaw angles. And similar issues in steady-state - when going up or down hill...
3) An IMU (Accelerometers, rate gyros, and magnetometers -all * 3-axis's) might be better suited it you are doing anything where you need more than 1-axis of acceleration. I am using a bit of high end unit, but MoTeC has also configured the eLean unit which is primarily for motorcycles. This is not really an IMU (just rate gyros and accelerometers * 3-axis), but spits out raw values for the accelerometers and rate-gyros. I have not tried one yet, but plan to do so as they are relatively cheap.

Bottom line is that you will likely be dominated by chassis noise and roll pitch and yaw attitude that can appear to cross couple into the single axis depending on how far you "peel the onion". Having more axis and rates can be helpful if it is not for a drag car.

If you PM me your email - I will send the eLean datasheet, or contact Ivo Boniolo at ivo.boniolo@e-shock.it

Hi Holmz,

I am doing research on dynamic cornerweight for my diploma thesis. For that I have installed one of the above mentioned single axis gforce sensors at each corner of the car. To estimate calculation error I need further specifications of the sensor.

I'm not quite certain if the technical terms I used are translated correctly into english. I used dict.leo.org to translate. What I am looking for is in german called :

Messgenauigkeit - accuracy of measurement
Rauschabstand, Rauschdichte - noise density
Empfindlichkeit, Rauschintensitaet - noise intensity

*edit*
Another specifications I need to know:
Grenzfrequenz - cut-off-frequency, limit frequency
Frequenzbereich - band, frequency range
*edit end*

Thanks for your help.

Greetings
Julian

CAPMAGIC wrote:Hi Holmz,

I am doing research on dynamic cornerweight for my diploma thesis. For that I have installed one of the above mentioned single axis gforce sensors at each corner of the car. To estimate calculation error I need further specifications of the sensor.

I'm not quite certain if the technical terms I used are translated correctly into english. I used dict.leo.org to translate. What I am looking for is in german called :

Messgenauigkeit - accuracy of measurement
Rauschabstand, Rauschdichte - noise density
Empfindlichkeit, Rauschintensitaet - noise intensity

*edit*
Another specifications I need to know:
Grenzfrequenz - cut-off-frequency, limit frequency
Frequenzbereich - band, frequency range
*edit end*

Thanks for your help.

Greetings
Julian

Then you will need to get those numbers from whoever is supplying you the accelerometers.


Could you also do dynamic corner weight with damper position or with roll and pitch measurements?

The accelerometers will only give you the transient changes.
In a long turn will it be easy to see anything happening in the accelerometer data?

I am using damper position as well. Doing the maths creates damper velocity and acceleration by derivating.

Damper position is configured in relation to cars body. So that calculated damper velocity and acceleration are only relative values. By subtracting relative damper acceleration from vertical acceleration of the body I get my absolute acceleration of unsprung mass.

By knowing the sprung and unsprung masses of the car I calculate the dynamic change in cornerweight.

Julian,

I know you have probably already set up your measurement equipment, but you could also measure wheel load directly. Depending on your suspension configuration, you could put a load cell at the top of your damper mount or put a strain gauge on your pushrod/pullrods (I didn't know if this might be a formula-style vehicle or not).

Just one more idea.

PaulGM wrote:Julian,

I know you have probably already set up your measurement equipment, but you could also measure wheel load directly. Depending on your suspension configuration, you could put a load cell at the top of your damper mount or put a strain gauge on your pushrod/pullrods (I didn't know if this might be a formula-style vehicle or not).

Just one more idea.

Good point Paul.

Julian wrote:...
Damper position is configured in relation to cars body. So that calculated damper velocity and acceleration are only relative values. By subtracting relative damper acceleration from vertical acceleration of the body I get my absolute acceleration of unsprung mass.
...

Julian-

Do you need to have chassis height relative to the road, or the road surface modelled/known as well?
Is your frame of reference:
1) the car body?
2) The road?
3) The earth?

Probably you would want to run some modelling simulations of it all before you go with a bunch of sensors.
You could do one corner and verify that your models and the measurements make sense, before getting the other corners set up.

Sounds like an interesting thesis.