Will Wireless enter the plant through the maintenance shack?
There's been an enormous amount of verbiage and ink spillage over wireless sensor networks for the past six or seven years. The single biggest application, however, is the release of trapped data in existing HART transmitters. Most of this data is diagnostic and maintenance data, and it makes sense to me that the asset management folks (read maintenance people) will be foremost in wanting this data...and WirelessHART adaptors are expected to be inexpensive enough to be bought on maintenance funds, not necessarily as capital purchase items. What do other folk think? Will wireless sneak into the plant through the maintenance door, or will there be lots of capital projects to retrofit existing HART transmitters?
Walt Boyes
Editor in Chief, Control and ControlGlobal.com
- Login or register to post comments
- 2266 reads
- Permalink
- 8 comments
Wireless enables a new approach to diagnostics
Most conversations I've seen about using wireless technology for diagnostics view wireless as a new path for data. I'd like to see the conversation turned on it's head. Does wireless allow us to take a new approach to diagnostics? With wireless technology, any device can access information from any other device. Can this "no boundaries" access to information allow us to do new and fundamentally more valuable things?
A simple flow loop example. The flow valve travel sensor shows the valve position is closed and there is sufficient seat pressure to stop flow. The flow meter shows positive flow downstream of the valve. Taken individually, both are normal and proper readings that have no diagnostic value. Taken together they show that either fluid is leaking past the valve seat, or the flow meter has a zero offset. Either of these conditions represent a genuine process problem and the diagnostic has real monetary value.
I submit that wireless technology will allow us to get fundamentally more valuable insight by allowing data from multiple devices to be used together to draw conclusions not possible using information from a single device.
Thoughts?
TomW
I think you already have this & it's for sale
Any of the Gensym (?)-based "expert system" operator advisory packages - "Nexxus" comes to mind - "can" this sort of diagnostic practically out-of-the-box, for example, "if pump stopped there should be no flow". Whether you can "enhance" this with more in-depth instrument diagnostics is interesting - but I haven't found the utilization % of such operator advisory packages highly motivating. That is, even using 4-20 mA conventional data and "situational knowledge" input from experienced operators in the creation of complex "rules", sites with which I'm familiar have not reported any smashing success. Even operators with limited experience say the "intelligence" coming from the expert system, warning of some abnormal condition, is still too rudimentary i.e. "I already knew that". Therefore they end up like the chattering alarm: when they finally detect a condition that could "save the day" (and pay for the high cost to implement) you wonder if anyone will be looking . . .
Like advanced control, these schemes need champions / shepherds / custodians (usually the same engineer) to "tune them up" & maintain them in hopes of higher utilization. In my mind, it has much in common with all the Asset Management related initiatives we dream of benefiting from smart, integrated devices.
John Rezabek
ISP Lima LLC
Just be creative !!!
What I've been seeing so far is that every plant implementing WL technologies has its own story. Sometimes through maintenance, then again sometimes through a project.
Bottom line is that WirelessHART is by now a proven technology : it works and it works good. It is standardised (HCF) and a lot of products are now available of different vendors.
So : if you get familiar with the technology (it is simple !)and are open minded, the possibilities to get more out of your plant for less are only limited by one's creativity....
Diagnostics & Maintenance Data
We use some of the HART variables for diagnostics (as an example:-comparing AVP to IVP to see if a valve needs to be scheduled for maintenance). The AVP value is displayed on the operator screen right next to the IVP. So, the operator puts in a maintenance request (if the valve can be taken out of service) OR it goes on schedule for the next turn around for repair.
Transmitters are another issue. Quite often there is not a lot a technician can do except to replace the electronics module (hockey puck). So, sending out a lot of maintenance or diagnostics data to a maintenance station does not offer any better idea/solution. All it does is create a whole bunch of nuisance alarms (or alerts in this case).
Setting up travel alerts, high scale, low scale alerts and things like that only adds to more nuisance because process conditions can change often. So, depending on the user, it may give the appearance that a lot of diagnostics data is not being used (or no one is taking ownership).
Walt, I would wonder why -
Walt, I would wonder why - if the maintenance department wanted to see this data, why haven't they installed muxes by now? I was surprised how cheap they were, and the pricing I was getting for radio antennas to connect to transmitters in hazardous areas was 2-3X the cost of the "old school" way.
I hear of some very sophisticated plants where there's challenges getting anyone to take ownership of the diagnostic data they have (whether by HART, FF, or Mux).
John Rezabek
ISP Lima LLC
Cost evaluation of Wireless technologies
John,
When you are comparing costs of the Wireless way of unleashing the hidden diagnostics of existing intelligent devices, are you comparing only the cost of the hardware itself ? Or did you compare the total cost of the project to mux the digital info off of classical 4-20 mA wiring to a separate work station with an asset management system ?
Cost of engineering, re-doing the drawings, installation, retesting the loops, stopping the plant, ....
Just wondering .... ?
Cost evaluation
I would be curious to know that also. I always thought the installed cost was much lower, and has been held up as a primary catalyst for deploying wireless devices.
Total installed cost with a MUX I would think would be much higher.
Cost Estimates
As far as drawings go, if I'm updating a loop drawing to show a Mux co-termination, wouldn't I do the same for an antenna co-terminated at the device? Some of the Mux solutions drop in in place of existing TB's, so you can have cases where the drawing impact is next to nothing.
Don't I have to design / document my wireless "gathering points" and their connection into the system?
The antennas are going out in classified areas, so they have to be designed for that area classification, and certified to "not" compromise the rating of the old instrument to which I'm affixing it. If I'm not doing this during a shutdown, I need hot work permits and trip bypasses to get the work done. This is not insurmountable, but adds further to the wireless cost disadvantage.
It has been more than a year ago, but when I asked for a budgetary price, I found I could buy a 24-point mux from P+F for the price of about 3 antennas. Maybe the price has dropped, but I can do a lot of terminations (same # as wireless solution) and drawing updates at that rate, and still be money ahead.
In fact the system I was inquiring about was an ESD system (Triconex) heavily laden with legacy HART-smart devices. It would require process outage for all practical purposes - even though separate voting transmitters are on separate cards, operations simply wouldn't want the stress and the mayhem (staring at multiple "votes to trip" even if they were byapsses). But this "cost" is really a "sunken" one, not borne by this project - we shut down periodically anyhow. And muxes can be implemented incrementally as well.
In my mind, the wireless solution has these advantages:
1) Management thinks it's sexy. I am routinely being asked, "Is this a good application for wireless?!?" (pant, pant) Until we look at installed cost.
2) It can be installed incrementally and less disruptively
3) It keeps the be-soiled instrument techs out of the DCS room and preserves the old work practices where systems people don't have to interact with the field device people. Maybe not a good thing?
John Rezabek
ISP Lima LLC