Just what are we talking about when we say wireless? How wireless are wireless sensors? Do they require any cables? Are the wireless sensors battery powered and transmit to a software package?

Don, it all comes down to your application. There are various combinations of sensors out there. Wired, wireless, tethered ect... One that I have worked with and (Vendor warning ) is http://www.techkor.com/industrial/wireless.htm

Don

an example of two versions of wireless devices can be found at this website:

http://www.rlwinc.com/products.aspx

RLW produces both the wireless, battery powered sensors and a concentrator that connects through wires to sensors but is wireless back to the network.
The S2NAP acquires data, performs signal processing and send the results ot a network host as an XML document.

We have had good results with these devices

Steve Shores

I think the S2NAP uses 802.11b (effectively a WLAN), but I've seen Bluetooth and even some in the unlicensed bands like 300 MHz, 433 MHz, and 2.4 GHz (essentially cordless-phone frequency ranges). I've heard about Zigbee trying to make inroads; it's a wireless mesh networking protocol that promotes itself for things such as slow data transmission, such as might be used in monitoring applications.

For wires? If it's battery-powered (some are) and its data transmission is wireless, it's 100% free of wires, much like a cellphone sitting on a machine "talking" to a central hub that might process the data.

The way I see it, there's two ways you could have a system like that--1) Wireless distributed data acquisition with a central processing unit, or 2) Distributed "smart" measurement devices that house the entire intelligence locally and don't necessarily need the 'mother ship' to communicate with. That's just my observation, though.

Jeff

Some of the wireless systems use motes to form self organizing networks. Batteries are needed.

Besides wireless sensors get rid of senseless wires.

We are using wireless sensors from CSI. They are battery operated and require a yearly maintenance contract to keep them in calibration and good batteries in them. We mounted the Transmitters on the crane and we CAN (are able to) receive the data back in the shop in real time. In practice though we will go out to the crane and take a reading from the receiver on demand because there are many problems to overcome when using wireless transducers on things like a crane. Bad readings are something to be considered because the crane doesn't always run at the same speed.

This is a work in progress and MIGHT get abandoned. We do think that in a full time operation that the results would be very good.
Gary

Just wondering the reason it's being abandoned.. Are you just unhappy with the quality (or at least reliability) of the results? Immature product?

Not necessiarily getting abandonded... Please notice MIGHT in all cap letters. The reasons are many but to narrow it down a bit.
1. the crane is a very slow device. We run out of rail before we can get a viable reading.
2. variable speed during data collection offers erroneous readings.
3. not enough head room for hoist drum readings

The product works well in steady state applications. A crane is anything but.
Gary

Ok--I did note the fact that it wasn't totally scrapped yet, but was just wondering the basis if it did go away--product quality, square-peg/round-hole fit for the application, etc. Thanks.

You might find this article interesting:

http://www.dliengineering.com/downloads/DiagEvolution.pdf

In my opinion one of the largest concerns with wireless sensors (vs. wireless systems) is that the sensors cannot be queried - therefore you can;t control when a test begins or ends, nor can you only take tests under proper conditions. This is what you are seeing with the crane application. A more intelligent system can wait until the appropriate time to collect data...

The wiring at the machine level is typically not the issue. It is more important to have wireless communication from the machine to the monitoring/analysis software. (The big saving is in wiring cost from the machine that many times can be as high as $200 per foot.)

New solutions are being made available that challenges the traditional walk around techniques in both cost and simplicity. There are systems today that can be installed and commissioned including the sensors, wireless communication monitoring/analysis software starting as little as $1,000 to $1,500 for the first machine going as low as $750 per machine at quantities above 300! (Each machine having 10 vibration sensors and 4 temperature sensors)

I think the “walk around” technology will disappear within the next couple of years as the cost of wireless systems go down. It will simply be too expensive to have people walking around every thirty days when you can monitor every 30 seconds for the same cost...

Eric

Eric,

I would like to hear more about your low cost system.

Walt
w_f_strong [at] msn [dot] com

I don't know if I agree with the idea of the walk around routes disappearing. Allot can be said for the visual inspection process (eyes, ears, smell etc) that is accomplished during the readings. I do like the idea of wireless sensors for some applications though.

I can't imagine anyone placing 100% confidence in 'automated' diagnostics. Nothing will replace human judgment (even if it's justified). What could be the best result is more effective, focused walk-arounds, triggered by these wireless sensors separating the useful result from what appears to be a sea of data.

Walk around data collection will never disappear. Nothing replaces you own senses at a machine. Your ears, eyes and touch tell you a lot about a machine, even with out a data collector.

I can only hope that the paper machines I monitor will be covered with sensors but that is a long shot. I just love the months of July and August.

We are currently setting up a remote system at the paper mill I cover. For now the remote system will only covers the plant most critical/costly equipment (supply and exhaust fans, TAD/Yankee drives, Fan Pumps). In all we will monitor around 96 points + speed and temp for each machine. Our software (all web based no VPN, window explorer or your choice of browser) will allow for me to monitor these critical machines when I am away from the mill, so nice.

The mill has already see the value behind Azima's equipment. There will be more expansions to come.

Too many wireless sensors can be hard too manage. Each sensor has to have its own IP address.

If you wire to a 8 or 16 channel device that is wireless to a network then it only takes one IP. Much easier to manage.

I don't think that the technology is the problem. Costs are coming down and capabilities of wireless systems are improving all the time.

The process industries are slow to react to new technologies. The walk arounds will never be replaced but it does not make sence to have people walk around and take vibration measurements once every 30 days when you can get them once every 30 seconds for the same cost!

The new wireless technology makes it possible for maintenance people to concentrate on the bad actors rather than spending time walking around taking monthly measurements. Companies like Azima (www.azimainc.com), DLI Watchman (www.dliengineering.com) and Prosmart from ITT Industries (www.ittprosmart.com) makes this a reality.

Some of the the new technologies provides web based interface, overall vibration and 10 band alarms per vibration axis(checking the FFT ever 5 seconds!),FFT's on demand, automatic FFTs on alarm or on schedule.

These systems are so cost effective that they can be used on all types of rotating equipment. We are talking less than 20% of the cost of the most cost effective traditional hardwired systems from CSI and Entek and less than 5% of Bentley Nevadas. As an example an 16 roll paper machine with 32, 3-axis, vibration sensors cost less than $20,000 installed and ready to go! That is less than the installation cost of traditional systems!