Can anyone give any info regarding using 4-20mA sensors? or loop sensors

From reading on my own it seems you can do different ways.
1) using a 4-20mA sensor direct to the PLC
( process logic control ?? like a honeywell system? )
2) using a standard accelerometer, wired to a vibration transmitter that outputs a 4-20mA signal?

I know nothing about this.. the 2 ways above is what I understand at the moment and I am sure there are errors in what I understand.

Idea is to have the signal displayed to operations 24-7 thru PLC. Noticeable changes can warrant call ins or shut down, depending on levels and alarm values set.

Current interest is for a few oil circulating sleeve bearing fans.

This
http://www.hardyinstruments.com/products/index.php?id=72 seems to be a compact sensor transmitter that can have a connector added to plug 2120 into, Is this how it appears?

Then I have read info on other transmitters that are separate, accellerometer is used and it sends the 4-20mA to the PLC.

I would like to understand the differences and reasons one is used over the other.

I see where Wilcoxin has the IT transmitter, allowing to send 4-20mA to PLC but also has a connector to hook up equipment such as 2120 meter for route collection.

anyone else using 4-20mA to PLC and/or for route based meter connection?

Thanks
Mike

This message has been edited. Last edited by: MikeH,

Mike,

You are correct in the two basic sensor arangements. You have several choices including:
Accel, vel, disp, or high frequency acceleration
Detection: rms, peak or true peak
Filters
Full scale amplitude
Buffered signal output

Some venders include: CTC, IMI, SKF, and Wilcoxon in no particular.

Walt

2 wire loop powered sensors with dynamic output for analysis:
http://www.reliabilitydirect.com/transmitterproducts/cmcp420vt.htm
http://www.stiweb.com/proddata/cmcp420vt.htm
3 wire 24 Vdc powered transmitters for interface to any sensor:
http://www.stiweb.com/proddata/cmcp500_vibration_transmitters.htm
3 wire 24 Vdc transmitters with monitoring/protection functionality:
http://www.stiweb.com/proddata/cmcp500_vibration_monitors.htm
Typical integration:
http://www.stiweb.com/proddata/cmcp500sample.htm

Personally, I believe people place far too much faith in readings from 4 - 20 mA vibration sensors and have a false sense of security about what a steady level really indicates.

Why do I say this? Consider an overall velocity reading. If the machine is slightly out of balance, the predominant peak in the vibration signature will be at rotation rate. The predominant peak in a bandwidth controls the overall level. In fact, until another peak is more than half as large as the predominant peak, it will contribute almost nothing to the overal level. Even if a developing problem causes another large peak at a different freqeuncy, the overall level will only increase by about 30% when the second peak reaches the same level as the first peak!

What this means in the real world is that developing problems are masked. In the case of the above machine, a bearing could go from perfect to serious deterioration and the overall level, controlled by the 1x peak, will show almost no change at all.

Monitoring acceleration, velocity or displacement will focus on different parts of the vibration signature, as will the various filtering options, but this really means is that you have to decide what's important to monitor, to the exclusion of everything else. Monitor velocity if balance is the biggest concern. Monitor acceleration if you're concerned about bearings. But be very much aware that you will be ignoring other potential failures! An unchanging overall vibration level doesn't mean the machine condition is steady.

Jon
Spintelligent Labs

quote:

Originally posted by Jon Chandler:
Personally, I believe people place far too much faith in readings from 4 - 20 mA vibration sensors and have a false sense of security about what a steady level really indicates.
Why do I say this? Consider an overall velocity reading. If the machine is slightly out of balance, the predominant peak in the vibration signature will be at rotation rate. The predominant peak in a bandwidth controls the overall level. In fact, until another peak is more than half as large as the predominant peak, it will contribute almost nothing to the overal level. Even if a developing problem causes another large peak at a different freqeuncy, the overall level will only increase by about 30% when the second peak reaches the same level as the first peak!
Jon Spintelligent Labs

Jon
I AGREE !

I was told "they" were going to put something on the fan with an alarm wired to shut the fan down if vibration levels got too high.
I suggested temperature since this is oil circulated sleeve bearings on a large fan.

A bearing could be wiped out in this particular application and overall or 1x could be changed little with shaft eating its way down thru the sleeve, especially since it is large and heavy, ooseness may not show greatly until severe looseness is present.
But.. I was told this is what they were going to do. What can I do ? just get the info needed and present it.

Thanks for all the input.

ps. This equipment will remain on a routine 4- 6 week route survey,for reasons I dont know they want a vibration alarm feature wired in that will alarm then shutdown if vibration gets excessive.I asked what they wanted to monitor thru the 4-20mA, and as you mentioned told them if we monitor accelleration it will give you idea if bearings are dry but may not warm of imbalance issues.

I tried to think of what may suit best and suggested velocity for running speed or alignment issues and temperature for bearing condition issues.

I am waiting for directions.

Mike

Hello Mike,
I face the same issue here. The DCS is responsible for the emergency vibration shut down. The vibration signal is integrated through a homemade system and is seriously unreliable. After few failed system (4-20mA sensor), the control engineer has finally ask our help. We propose three different scenarios and the engineer choose the less effective one! He uses an IMI transmitter with a 100-mV/g acc that feed 4-20 mA to the DCS. The IMI hardware is rugged and effective but I feel that it may involve too many workers from different background, which can lead to system malfunction. I’d rather buy an online system dedicated to machinery monitoring and safety then the overall value can always be directed to the DCS through an Internet connection or 4-20 mA hard wired. An online system is also a very effective predictive tools that maintenance may uses as the Pdm’s foundation. The tough part is to get all the different trade people to work together as team and see what is best for the plan’s productivity.

Best regard, Marcel

Mike,

Since the fan has sleeve bearings, velocity is the best bet. This should detect the most likely failures.

I guess I should say that overall alarms can warn of high vibration levels - they'll let you know when the damage is done, even if showing no changes until that damage has occurred. These systems should be considered a safety system rather than a predictive maintenance system.

Jon
Spintelligent Labs

Mike
We have several machines with 4-20 mA probes
on them. They work great especially with fluid film bearings.

Our experience shows temperature sensors on the bearings are as important as vibration sensors.
Also suggest you look very hard at alarm and shut down levels.

I have experience with IMI and Metrix Sensors.
Hardware seems to be about the same But IMI has
by far the best Product Support.