Vibration Monitoring

A vibration sensor is a transductor that measures amplitude, frequency and acceleration. When coupled to a signal processing device and appropiate software you get a vibration monitoring system. You can place the transductor practically anywhere you want on the machinery or prime mover. There´s a lot systems and vendors out there. This a good place to start: www.reliableplant.com

The one I am needing is supposed to be inside the motor, I suppose measuring vibration on the bearings. Does the motor vendor have to build the motor with the probes inside the motor? What are some probe manufacturers that make these types of probes? I need some sort of 4 - 20 mA signal to take back to a PLC (AB PLC).

It´s easily installed by a plant technician. The transducers have a magnetic attacment and are typically small, 3-4 cm2. Better to look up some vendors and catalogs and start a discussion on a complete system.

Vibration sensors are not usually mounted inside the motor. They are mounted close to the bearings which usually means the sensor will be outside the motor (on its surface). There are two types that I know of: magnetic or bolt-on. The magnetic type just attaches to the surface by magnetic force. This usually means it's only temporarily attached. The bolt-on type requires a threaded hole in the bearing housing where the sensor is mounted. This is the more-or-less permanent type (until you need to replace the motor, in which case you remove it for attachment to the replacement motor).

We use holthausen elektronik's ESW-compact (hol600) in some of our equipment. They're very easy to use and they have 4...20mA output. You can't mount them inside the motor, sorry, and they're a bit big (about 3 inches in diameter, maybe 4 inches in length). Very reliable too. I haven't had one go bad in 13 years.

What is the KW rating of motor? what is the type of bearing?

The standard vibration monitoring used for motors with roller bearings are the contact type screwd onto the bearing housing( http://www.spminstrument.com/methods/evam/ is a popular one). You can see on the website measurement can be a mm/s reading or they have special transmitters for giving continuous signals for monitoring motor condition.

If motor is with sleeve bearing, one has to use non-contact type shaft deflection measurement technique (Bentley Nevada is a popular one). Measurement output is only in analog signal like 4-20mA. This monitoring system is very expensive and only feasible for motors >2MW for critical applications.

Both the above methods can be offered by most motor vendors as already fitted accessories. Hope this helps!

In mmy short experence I did not see these type of sensors mounted inside motor frame. One normally take vibration readings at the footing of motors because vibration is ultimately transmits to foots or mountings.

If the aplication concerns a large electrical motor(> 5m diameter or >10.000kW), I suggest using a system that combines the measurement of vibration, temperature rise and air gap.

This can be obtained wit equipment from Vybro SystM, a Canadian company located in Longueil, Quebec.

You may contact mr Marc Bissonette and tell him that I have recomended the use of this system.

Kind regards

Emmanuel Britto F.

Kistler Instrument Corp might have the sensors you need.

sales.us@kistler.com

Saludos!!!

IAN

First some definitions: Motor = electric motor. I think you are discussing an engine, such as your Avatar. There are 2 basic types of vibration monitoring devices.

Type One: Velocity & acceleration types which measure forces of the case itself moving (G-forces) and are using velocity and acceleration by measuring displacement (how far did the housing move) versus time.

Type Two: The other type is a proximity probe, termed a gap or eddy or prox probe. This measures movement of a shaft without touching the shaft.

To teach displacement v. time (velocity or acceleration type measurements >> Type One) I tell students to stand up and move their head to and fro 10 cm (displacement) but do it only 6 times per minute (frequency or time). Very slow. Do they have a headache? No. Now, move your head the same 10 cm 100 times per minute. Headache? Yes ! Because now we have g forces and reversals where you brain is smashing up against your skull and you can't stand it . . . . but in each example the amount of movement is still 10 cm. So to measure this we use a velocity probe or accelerometer physically stuck on the outside of a housing in an area of interest. Normally the area is inline with the center line of the shaft in X, Y, Z planes. On small engines, I put one inline with the crankshaft in the front and rear Horz (X), front and rear Vert (Y), and one axial above the flywheel housing (Z). On large engines > 3 MW I'll put another set of X & Y in the middle and also another Z on the non drive end. These are attached by threading to the case, or glued with epoxy. On the outside of the engine. Google: "Velocity transducers" and "Accelerometers". The output of these probes is in mm/sec (inches/sec) for velocity probes with 8 mm/sec or 0.3 ips 0-peak to be a industry standard velocity limit. The accelerometer outputs are in G forces and this style of case displacement is normal for higher frequencies.

The non-contacting proximity probe (Google "Proximity probes", "Eddy probes" or "Gap Probes") looks like a pencil with the 'eraser' tip being the part that shoots an eddy current into a shaft and measures how much that shaft moves. These must be mounted inside your engine and threaded through a bearing area, through the bearing material and the probe tip must be very close, but not touching, the shaft you are measuring. There is no exposed shaft in an engine where you can screw these probes into a housing and 'see' the shaft, so they would need to be fitted through the bearings. I have been and engine consultant for 30+ years and I have never utilised proximity probes inside an engine except for laboratory experiments. Prox probes are common in other equipments with shafts such as electric motors, gear boxes or pumps, but not engines.

Good information. Thanks.

The motor is an electric motor as you stated. It is a 40hp TEFC 1800 RPM Motor. What I didn't realize is the fact that was mentioned previously regarding ball and sleave bearings. I can't use a prox probe inside the motor because there is a gap between the balls in the bearing and it will measure that as 'vibration'. That being said I have convinced the client of this and are going with some sort of vibration system mounted on pads on the outside of the motor housing (probably either Metrix or Bentley Nevada). I do have another question, though. The bentley nevada system, from what I can gather, requires separate data acquisition, where the metrix system is just a 4 - 20mA signal I can send back to my PLC without the use of separate DA, correct? Thanks for all your help on this subject!

You are exactly right that with ball bearings, there is no movement of the shaft when they are warm (tight when hot, a tiny tiny tiny bit loose when cold), so proximity probes will move with the case and always read zero. You need 'case displacement' mounted on the housing near the ball bearings.

Always use the terms "Electric Motor" or "Reciprocating Engine Driver" so we don't get confused, especially since your Avatar looks like a small Cat engine !! 'Motor' to people not familiar with engines can be diesel 'motor', or electric motor.

I haven't bought a Bently system since the old 3500 series system but typically Bently always tries to sell more black boxes. It is how I tell Don Bently "I'm making your yacht payment today" when I buy his equipment. Now GE owns Bently so it is really expensive stuff; good, but expensive.

There are many many manufactueres of acceleration transducers (ball bearings can produce high frequency so ask the supplier if an accelerometer is indicated or a velocity transducer) so just Google for these. Metrix is a good brand. National Instruments has virtual instruments where you use whatever transducer you require and a PC to build a virtual instrument panel with shutoffs and alarms. Then you don't even need a PLC, but, as you know a PLC is it's own brain box and works independently of Bill Gates and Microsoft. With the virtual instruments, you are at the mercy of Bill Gates, so if your PC goes 'kaput' when MS operating system locks up, you are 'down' for protection until your IT guys fix it. But the virtual systems are VERY VERY cheap and quite functional if you keep Bill Gates happy.

There are a number of manufacturing maintenance trade magazines out there. I fail to recall the names of any. Their advertisers sell the exact type of systems you are looking for. I suggest knocking on a few doors and see if any of their maintenance departments have the names of these magazines. It might take a few tries because they usually just throw them away or they don't get forwarded to the maintenance department. From there it's easy.

Avatar = Cat C9 Turbocharged about 300 BHP if my poor old eyes can focus on that little picture and my memory is correct ??????

It is a Cat engine. I work with Cat engines and gensets and wanted a picture depicting something I work with....it has nothing to do with my question about vibration monitoring.

Whats up with your crazy baby picture?!

This answer assumes vibrations of the shaft ONLY

Shaft vibrations can take 3 forms (I will refer to vibrations

in 3 axes using circular (z, r, therta) coordinates:

1. Vibration of the shaft from the normal shaft centerline (Vibrations

in the r direction). This is the most common, and eddy current probes

are used to detect this vibrations. I suggest you Google Bently Nevada or

go to:

www.gepower.com/prod_serv/products/oc/en/bently_nevada.htm

2. Thrust Vibrations (Vibrations in the z axis direction). One uses

thrust bearings which have a dynamic response. Google Thrust Bearings.

3. Torsional Vibrations (Vibrations in the theta direction). This is the

most difficult measurement, and the interpretation of the data is also

the most difficult. One often uses short range r.f. telemetry

with a reasonable (1-2 kHz) data bandwidth and a full strain gauge bridge sensor.

Reply to this with your e-mail address, and I can send you further information.

Or, Google Torsional Vibrations.

Came across a good vibration meter. Hope it helps too.

http://www.industrialindia.com/catalogs/Portable_Digital_Vibration_Meter/Portable-Digital-Vibration-Meter-HTC-VB8205.pdf

hi there

my self sivaram.....

how actually the pick up works...........?

as u know we use electromagnetic pick up (poximity) to mesure the vibrations(shaft)

as u give the supply(12v) the magnetic pick up will be excited

and it will produse a electromagnetic field initialy( when the m\c is standstill u can see there will be no voltage drope r oscilations of readings) as the m\c startrotating it astually cuts the electro magnetic flux and hence an emf will be generated and it will be quated to some value of vibrations

and hence the switch will be activated when it reacehs and extream valu r set value and terminates the machine from services hence provided in view of safety...