Kistler to Vicon analog input NOISE

[Bernard]

Hi all,

We recently revamped out lab with Vicon cameras (Vantage V5), Nexus v2.12.1, two Kistler forceplates (9281EA). We are having trouble synchronizing the plates. Now I note this old post https://biomch-l.isbweb.org/forum/bi...forms-to-vicon.

Analog method
The force trace into Vicon is extremely noisy (+/- 20N), and I was wondering what reasons could be contributing to this? We connected our plates to the Kistler provided AD converter (which I assumes powers the plates), which outputs it into the Vicon lockbox, which goes into our computer.

Digital method
The force trace into Vicon is very good, We connect our plates to the Kistler provided AD converter (which I assumes powers the plates), which goes into our computer. The Vicon lockbox is also linked to the kistler ADC for syncing.

Link to one sample vicon data: https://filetransfer.io/data-package/4GyPRbRv#link (expire in 11/02/2022)

Questions:
1) What could contribute to the noisiness to our analog force trace? We followed the book in connecting, or could there be a mistake? I have included images as attachment.
2) Vicon recommend the digital method. With the modern equipment these days, I wanted to understand the pros and cons of each method.

Images






Many thanks for anyone's expert input.

Regards,
Bernard

[Edmund Cramp]

You could look at the data that you recorded into the C3D file with the free MOKKA viewer which can display the data in the analog channels - just looking at the force plate signals can be quite helpful because the file appears to contain the raw analog data "Force.Fz2 (N)" and the processed digital data "Force.Fz1 (N)" with a slight timing difference. You can verify the force plate data timing by coating a golf ball with retro-reflective tape and recording it dropped on the force plate to confirm that the force data timing matches the golf ball impact (when the Z trajectory reverses).
A little baseline noise isn't pretty but it can be quite helpful when you just see low level noise because if there are any force plate mounting problems you would be able to see large foot strike noise as the subject walks towards the plate before they step on it. Looking at the noise in the recorded C3D analog signals that you attached it looked fine but much lower than 20N - that's probably just a scaling issue?
I always found that getting force plates set up and connected to a motion data collection system was never simple but never too difficult - seeing problems like this allows you to get them fixed and guarantees your data quality once everything looks good.
AnalogDigital.JPG

[Jacob Wells]

Out of curiosity what happens when you zero the force plates in the software?

We recently added some Vero's to our t-series set up which introduced analogue noise to our "trigger" which we use for marking the data. Fortunately the level of noise was low enough that it didn't impact the spike caused when activating the trigger. I spoke with Vicon about this and their recommendation was to invest in Ethernet cables with metal connections.

The noise doesn't seem to have any impact on our Bertec force plates.