WLS-9163 not connecting wirelessly on rotating machinery

[smathews]

Hello All,

I am using a wireless data acquisition system consisting of a WLS-9163 wireless data acquisition unit with a NI-9237 card inserted for gathering strain data from a machine. The WLS unit is powered with a battery power source via a remotely controlled switch to facilitate a low power state and conserve battery power. The wireless data acquisition system required is chained to a rotating 250mm diameter machine shaft which rotates at 840rpm, or 14Hz.

During wireless system testing I can remotely switch power to the WLS unit and connect to the unit with windows wireless network connection which shows excellent signal strength. I then run a vi to gather strain data from the machine, and then power the machinery to 840rpm and the WLS-9163 continues to transmit data.

But here is the problem: if I remotely switch power off to the WLS unit while the machinery is running as is desired to conserve power, which disconnects the windows wireless network connection, then remotely re-power the WLS to connect again, windows can not connect to the WLS. The only way to connect again is to the WLS is to wait until the machinery stops or slows to RPM.

Unfortunately, during machine testing, the machine can not be stopped or slowed from 840rpm during its normal operation, and due to the length of the machinery testing to be done, there will be a long period where the WLS unit will be powered down to conserve battery power until the WLS unit can be powered on and the testing can begin.

So, does anyone have any ideas why I am not able to re-connect?

[smathews]

Bump

[Vimal_F]

Hello Smathews,

Thank you for bringing this issue to the forums. While I look further into this issue I would like to know a few more things.

1. What firmware is currently installed on your NI WLS-9163 device?

http://www.ni.com/pdf/manuals/373535a.pdf

Here is the latest 2.0 update if you do not have it installed:

http://joule.ni.com/nidu/cds/view/p/id/1615/lang/en

You can view and install firmware using Measurement and Automation Explorer.

2. Can you please expand a bit more of what you mean by Windows cannot connect to the NI WLS-9163 device? Can you discover the device in Measurement and Automation Explorer? Or, is the device visible in Measurement and Automation Explorer but, you cannot connect to it?

Thank you,

Vimal Fernandez

ELP Engineer

Applications Engineering

[Jesse_D]

Good morning smathews, 

Were you able to get your system working?

This sounds like system that other engineers may want to set up and it would be good to know that it has successfully been created. 

[smathews]

Hello Vimal and Jesse,

Jesse, no the system is still not working as I plan to use it... yet..

Vimal, in answer to your question 1, I have the latest firmware, and as the following email states the interesting result of doing a firmware upgrade for my application. In reponse to question 2, due to the layered nature of windows wireless and ni's max, you can not connect with the wireless device in max if windows wireless can not connect with it first, you get error in max...

As mentioned, the following is a more thorough email sent to an ni tech:

I am using a wireless data acquisition system consisting of a WLS-9163 wireless data acquisition unit, and a NI-9237 card inserted for gathering strain data from a machine. The WLS unit is powered with a battery power source via a remotely controlled switch to facilitate a low power state and conserve battery power. The wireless data acquisition system required is chained to a rotating 250mm diameter lathe chuck and set the lathe to rotate at 840rpm, or 14Hz to simulate final testing I am aiming for on a 275mm diameter rotating shaft.

During wireless system testing I can remotely switch power to the WLS unit and connect to the unit with windows wireless network connection which shows excellent signal strength. I then run a vi to gather strain data from the machine, and then power the machinery to 840rpm and the WLS-9163 continues to transmit data.

But here is the problem: if I remotely switch power off to the WLS unit while the machinery is running as is desired to conserve power, which disconnects the windows wireless network connection, then remotely re-power the WLS to connect again, windows can not re-connect to the WLS. The only way to connect again is to the WLS is to wait until the lathe stops or slows to around 240rpm. This showed me the error is possibly occurring on the ack handshake signal of the wifi connection.

Unfortunately, during final machine testing on the rotating shaft, the machine can not be stopped or slowed from its operational speed of 750rpm, and due to the length of the machinery testing to be done, there will be a long period where the WLS unit will be powered down to conserve battery power until the WLS unit can be powered on and the testing can begin.

I first updated the WLS with firmware version 2.0.1, which allowed me to re-connect at a higher 350rpm, but not at the lathe’s next highest setting of 470rpm.

Wondering if a doppler type shifting was occurring on the ack handshake signal of the wifi connection I added an antenna extension to the WLS and placed the antenna in the centre of the lathe to stop the back and forth movement, but this did not work so ruled out this idea. Next, I tried a larger antenna, but as is assumed, this made no difference as when connected my signal strength is excellent.

　

Sooo, bump

[Jesse_D]

Hi smathews, 

Could your double check my math? (14Hz)(.138m radius)(2*3.1415 rad/rot)= 12 m/s, the speed our antenna is traveling at

and using the doppler effect for electromagnetic waves and the minimum frequency per the 9163 specifications: df = (+/-)(12 m/s)/(3E8 m/s)*(2.412GHz) = (+/-)96.5Hz, the doppler effect on the signal

The spacing of each channel in the 802.11b and 802.11g is 25MHz which creates an error of 4.4E-6 of the range of a given channel and 37E-9 for the center frequency. According to page 29 of the 802 specifications, the center frequency can tolerate an error of up to 25E-6. Right off the bat I don't see doppler being a problem but I need to do some more research before I can put a lot of confidence in this assertion. The entire situation, especially the dependence on speed, sounds like some sort of interference.  

I wonder if as a quick work around to this situation you could use a slip ring to supply power to 9163 and just never worry about powering down. 

[Jesse_D]

Hi again smathews, 

Is your machine electric? The noise from the AC motor could be interfering with the wireless.