Force Sensor port connections confusion

I have constructed 2 SIMSCAPE models of a 3 DOF system. One is with a force sensor connected in parallel with a 3 degree of freedom mass/spring/damper system & the other is in series. The connection schemes in SIMSCAPE is a bit confusing; nonetheless, I have attached an image of how I interpret the 2 configurations in a conventional mechanical scheme.

In the case with the series connection there is a node that remains free. Hence, I would not expect them to contribute to the response of the system. However, if I employ the zpk or tf commands listed below I find a singular REAL pole locate @ c1/b1. So my interpretation of the connections must not be correct. Can anyone provide a correct schematic interpretation of my 2 models? My parallel connection schematic I believe is correct.

I have included both .slx models along with the necessary input data to allow you to experiment with the results. Let me know if anything is missing & I will provide it.

zpk(SIMULINKbode.values)

tf(SIMULINKbode.values)

Thanks for any assistance

Answer 1

[Ed Marquez]

Hi Jeff,

this is definitely an interesting application - and thanks for attaching these nice models. I think the doc link below can provide some helpful information for your measurements in Simscape:

https://www.mathworks.com/help/physmod/simscape/ug/basic-principles-of-modeling-physical-networks.html#bq89sba-3

I would suggest sticking to the force measurement in series. Simscape works with through and accross variables and the product of these two equals power in that domain (P = V*I in electrical, P = F*V in mechanical translation, or P = T*w in mechanical rotation). The through variables are always measured in series, and the across variable measured in parallel. In your models, if you want to place a gauge in parallel with the elements, then you should be using an ideal motion sensor (for x and v).

The controls part of the system (zpk and tf) heavily depends on the measurements being correct, so I would start by making sure that the response measurements yield good numbers and then re-visit the controls part after that. I hope this help and good luck!

[Jeff]

Thanks Ed, I am still investigating this issue. What you have stated seems to correlate with my observations. Connecting these sensor ports whether in series or parallel do yield vastly different results. Unfortunately, the documentation I have seen does not discuss this. I will review what you have referenced.

[Jeff]

Hi Ed,

Your direction did help. I was able to construct an accurate mathematical model based on the schematic below. The results concur with the force sensor in parallel. However, there was a perplexing discrepancy which I have posted @ https://www.mathworks.com/matlabcentral/answers/340167-reconciling-zpk-output

If you have a working knowledge of the zpk command maybe you can provide some insight on the discrepancy.