Estimation of the damping coefficient of a suspenion

[serbring]

Hi all,

I have a suspended seat with a scissor mechanism like the following:

This seat is composed of coil spring and a hydraulic damper. My aim is to develop a multibody model of that seat. Actually, my model is quite complete, only the parameters of damper are missing (i.e. the damping coefficient with respect to its piston speed). I do not have any technical details about it, so I need to estimate it. I am thinking to use a system identification approach, that means, applying a known excitation to the seat, measure the displacement of the seat cushion and tune the model parameter so that the measured displacement will be reproduced. I do not have any rig so that I can easily apply a known excitation. I was thinking to excite the seat by applying an impulse (falling mass at a certain height) to the seat cushion. What do you think about this approach? Do you know an easier approach to get the damping curve of the damper?

Thanks,

best regards

 

[anorlunda]

The damper manufacturer must supply that information. Otherwise, no seat manufacturer would know which size damper to buy.

 

[serbring]

Hi Anorlunda,

you are on right, unfortunately I do not work for any seat manufacturer. I have contacted both the seat and the damper manufacturer and both have not supplied any information. So I was thinking to use an experimental mode.

PS
I am sorry for the late reply, I trusted on the email alert :|

 

[Chestermiller]

If you know the spring constant, you can estimate the damper coefficient by displacing the seat, releasing it, and then measuring the subsequent displacement vs time.

 

[serbring]

Yep! you're on right. Easy and safe. This made me think, that this type of dampers has two different coefficients, one in compression and the other in rebound, right?

 

[Chestermiller]

Yep! you're on right. Easy and safe. This made me think, that this type of dampers has two different coefficients, one in compression and the other in rebound, right?

I don't know about that. Damping coefficients are usually assumed to be the same for both directions.

 

[anorlunda]

How about you just experiment. 60 seconds of searchiing found this item, an adustable force air damper.
https://www.alibaba.com/product-det...l?spm=a2700.7724838.2017115.49.57c31d936fIF7q

They are inexpensive. How about just buying one, then use the adjustment to tune it to your needs?

 

[serbring]

How about you just experiment. 60 seconds of searchiing found this item, an adustable force air damper.
https://www.alibaba.com/product-det...l?spm=a2700.7724838.2017115.49.57c31d936fIF7q

They are inexpensive. How about just buying one, then use the adjustment to tune it to your needs?

Which kind of test you are thinking about? I need to find the parameters of the damper that is already installed in a seat suspension

 

[anorlunda]

Which kind of test you are thinking about? I need to find the parameters of the damper that is already installed in a seat suspension

Ah, I see. Are you able to make a deflection, then release it while accurately measuring the position versus time recovery? That is similar to excitation that you suggested in #1, except that instead of a periodic excitation it would be a step function. Success depends on the instrumentation available.

Come to think of it, a video recording analyzed frame-by-frame would be a good way to measure deflection versus time.

Edit: The whole test and analysis might be easier if you removed the damper from the chair.

 

[cjl]

Yep! you're on right. Easy and safe. This made me think, that this type of dampers has two different coefficients, one in compression and the other in rebound, right?

I don't know about shocks for seats, but the shocks used for high end sports car suspensions frequently have 3 separate characteristics - high speed compression, low speed compression, and rebound. The simple model used in physics problems frequently assumes that there's a single damping coefficient, but this is frequently not the case in reality.

 

[serbring]

Ah, I see. Are you able to make a deflection, then release it while accurately measuring the position versus time recovery? That is similar to excitation that you suggested in #1, except that instead of a periodic excitation it would be a step function. Success depends on the instrumentation available.

Come to think of it, a video recording analyzed frame-by-frame would be a good way to measure deflection versus time.

Edit: The whole test and analysis might be easier if you removed the damper from the chair.

I know that the best is testing only the damper, but I do not have any tool to apply a known load. Any suggestion is appreciated. My idea is to install very precise draw-wire displacement sensors. I do also have a high speed camera.