Force plate zero offset

I have collected data from a force plate with the use of matlab. But initially, the signals of 6 channles were set slightly below or above the zero.

I want to know how to set it zero offset with the use of matlab code.

[Andrew Newell]

How do you know how much the offset is?

[Sangwoo]

I could figure out the offset after I had plotted the signals.

Answer 1

[Jiro Doke]

Typically, you would do a reference measurement (with no load on the force plate). Take that measurement and compute the average (mean) of it. Then after taking your actual measurements, subtract the average you computed before with the no-load condition.

[Sangwoo]

Thank you for quick response.

I could figure out how to do it literally. But I am not sure how to do it practically.

I have created the code as following.

If you don't mind, could you show me how to do it with the code??

AI=analoginput('dtol',0);

set(AI,'InputType','SingleEnded');

chan=addchannel(AI,1:6);

set(chan,'InputRange',[-10 10]);

duration=10;

set(AI,'SampleRate',250);

actualrate=get(AI,'SampleRate');

set(AI,'SamplesPerTrigger',duration*actualrate);

set(AI,'TriggerChannel',chan(3));

set(AI,'TriggerType','Immediate');

start(AI)

wait(AI,duration+5);

data=getdata(AI);

plot(data)

xlabel('Time')

ylabel('signal(volts)')

[Jiro Doke]

Well, do you get a plot when you run the above code?

If so, I would just run the code without any load on the forceplate, take the average of "data", and save that average no-load data. Then run the code again with the load and subtract your "average data" from "data".

[Sangwoo]

Let me give it a try. Again, thank you very much!!

As a matter of fact, I am a student specializing in biomechanics. I want to be a good programmer such as you to be a good biomechanist!!

[Sangwoo]

I got the average code. But I had to run the code without load and then run the code with load after subtracting the average from the data with load.

I am wondering how to run the code without load and with load at the same time.

[Jiro Doke]

I'm not sure that's physically possible. You have to physically have nothing on the forceplate for the "no load" condition, and then do the experiment. The way I used to do the experiment, was to record the no load condition at the beginning of the test and store the data. Then after doing all the experiments, I postprocess by subtracting the no-load data from all of my test results. Another way is to run the test with no load for the first few seconds. Then when you post process, take the mean of the first few seconds and subtract it from the rest of the experimental data.