Proximity Probes and Other Sensors

so today we're gonna be covering proximity probes and other vibration sensors vibration transducers basically convert motion into millivolts for displacement it's mils of displacement to millivolts of electricity generally we have two types of bearings you have your roller element type bearing and your fluid film bearing and we monitor these generally quite differently because of the physics of each one for the roller element bearing you have a very physical link between the shaft and the casing in the roller elements and the shaft does not move very much relative to the bearing or the bearing housing because of that very direct physical link so using a proximity probe mounted to the bearing face or to the bearing cap would not show very much motion because it is very limited by the available clearance between the elements of the bearing however it does transfer energy very well from the motion of the shaft into the casing of the machine and as long as you mount a case mounted sensor near that bearing cap you can get good transmission of that vibration data to that sensor however on a fluid film bearing the fluid acts as a cushion and absorbs the the motion or energy of the shaft motion and does not transfer it as well to the casing so therefore using a proximity probe that is mounted to the bearing face or to the bearing cap looking at the shaft you do get a view of the motion of the shaft relative to where it's mounted on very large mass rotors or where the mass of the rotating component is a higher ratio to the casing mass you can get pertinent data from mounting a case mounted sensor on the exterior of the bearing quite often you see that with nuclear steam turbines they will have a velar mounted in the same plane as the proximity probe and so with that we can get the motion of the shaft relative to the mounting point the motion of the machine case relative to a point in free space and by summing the two we can get the motion of the shaft relative to a point in free space so you can get a lot more information we also see we're mounting transducers on the outside of a fluid film bearing are good for detecting things like foundation issues also where there are outside influences that may shake the casing on one side I was at they were taking down the stacks at a power plant to go to a scrubber type system and they put Villars in a number of places on the machine and on the foundation to measure and compensate for the dropping of those stacks and it basically just created an earthquake type effect and you could see that stack drop and you're everything we kind of go haywire for a bit and then settle back down and by having that motion of the casing to correlate with the proximity probes we were able to tell if we wiped a bearing or anything due to that impact so we're gonna start with proximity probes proximity probes measure displacement generally used with fluid film bearings it is capable of going down to low frequency actually zero Hertz cuz things that aren't moving can be measured in displays are not rotating can be measured for displacement things like case expansion thrust differential expansion things like that that do not require rotation gets down to the zero Hertz and it's generally reported here in the states in thousandths of an inch or mils and in other parts of the world they use microns it is generally built up of three components the proximity the extension cable and the probe these are series specific so the five and eight millimeter proximity probes the probe cable and proximity are all color-coded you can see they have the the blue here and you have a blue label on the proximity at the end I think I've got one that's got a purple a purple set which had believe is the 25 millimeter but by looking at it you can visually confirm that they are the correct ones for that series they do need to be matched as an electrical length so the proximity for the eight millimeter probes come in either 5 meter or 9 meter without any counter modification and that means that the total length of the extension cable and the probe electrically not necessarily physically sums up to that 5 meter or 9 meter so if you have a five meter proximity you want either a 4 meter X inch cable and a 1 meter probe or a four and a half meter extension cable and a half meter probe same thing goes for your 9 meter as long as the total of the extension cable and the probe add up to the total length of the proximity you should be okay now how do they work the proximity produces a oscillating signal at a radio frequency and as that passes through the probe tip which is basically wound as an antenna it generates a field around the tip and as a conductive material moves into that field it generates eddy currents in the surface of the material and that creates a loading on the signal and by reading that loading and unloading we are able to measure the displacement so the very fast quickly line here is the radio frequency oscillation from the proximity as the material moves in and out of the range of the proximity probe as it moves further away we have less our full signal as it moves into the the field it absorbs that energy as a eddy current so again here we're showing it as ad modulated signal so we removed the the radio frequency component of it and we are looking at just the waveform that is left as it moves in and out of the field so here the probe is far away from the shaft material with a 8 millimeter proximity you're generally gonna see something in the 19 to 21 volts negative if there is no material in the field and as you get closer oops wrong way and the shaft gets right up against the probe tip you'll see it go down to about a negative 0. 65 there is a dead band of about 10 thousandths of an inch from the probe tip before it starts to change so generally have to get about 10 thousandths away from that probe tip before you start seeing the voltage go more negative from that 0. 65 volts so once we extract that demodulated waveform we measure generally peak to peak for your peak to peak vibration so measuring from the lowest point to the highest point we get your peaked peak we can also do peak RMS other values can be calculated from that waveform and the DC gap voltage is the average value of that waveform so the DC gap gives us basically the distance the shaft is from the probe tip and the AC component is the vibration movement we do both radial and axial measurements with proximity probes quite often everybody's familiar with the XY vibration component of using procs probes on a fluid film bearing one thing that many people don't pay a lot of attention to is that DC component can also be used to measure the position of the shaft in the clearance of the bearing in system one we plotted as a shaft center line plot and if it is a horizontal machine and lift all is off and the machine is not rotating the bearing the shaft should be sitting at the the very bottom of the bearing you can save that as a reference point and put in the clearances of the bearing in the shaft center line plot and be able to see that you do get lift when you turn on lift oil if you have a restriction to the oil flow and you don't get lift it can definitely save a bearing from being spun up dry you also can see as the machine comes up speed you see the oil wedge build and push the the rotor to the side and if you do have a machine incidence and it does why preparing you can actually go in and look at that shaft centerline plot and confirm that the the bearing was wiped we also measure axial movement looking at thrust differential expansion rotor expansion and case expansion case expansion it's not a very common use for proximity probes usually it's a greater distance than the proximity probes can properly read so the often that's used with an LVDT these signals can also be read as vibration I have run into cases where there was axial vibration and a machine and we just jumpered the thrust probes over to a radial vibration channel pair and we're able to see how much axial vibration they were getting and help them diagnose what was going on proximity probes can be landed up to four times on different instruments without getting into serious loading accelerometers bolometer z' generally if they get connected to more than one monitoring source it does load the signal and degrade the quality but proximity probes can be depending on the electrical components can be connected up to four times we also use proximity probes as a key phaser this gives us a once per evolution marker point and having that marker point recorded and the angle at which the probe is installed and the angles at which the X&Y probes are installed is necessary for doing proper diagnostics and also for doing proper balancing we usually use a notch because a notch is the easiest one to deal with you have the greatest chance of gapping the probe against the surface that is appropriate without having to get in there and actually look for the physical knotch being not lined up with the proximity probe I have had numerous cases where just pure bad luck they did get the probe lined up with the notch gapped it but they gapped it at the bottom of the notch and as soon as they started the Machine up it broke the tip off the probe and it was required replacement with a projection you do have to make sure that you roll the the rotor to the point that the projection is under the tip of the probe and if the projection is something like a key in a key way for larger shafts you do need to make sure that the corners are rounded down on that outer surface so that when you cap that probe in the center of the key as soon as that starts to turn those square edges of the key are physically further out from the centerline of the shaft then that Center mark of the key is and you can either get into having a double pulse or actually physically breaking the tip again for thrust we recommend that you install the probes within 12 inches of the thrust collar best practice is to have it actually on the thrust collar but some cases that is just physically not possible but as long as you get it within 12 inches generally we don't run into a problem however if you put it further away say on the other end of the shaft you can get into thermal expansion issues and if you're fighting how the the rotors thermally expanding versus actual thrust movement it really makes it hard to comprehend what's actually going on with the thrust I've got a kick out a display mode to go through this section so just a second one of the things that I quite often get asked is about the gapping of proximity probes customers will call up and go hey you know I've got this machine that you're the probes or gap perfectly at negative 10 volts right in the center of the linear range I've got this other machine over here that's capped at negative 6 volts is that throwing my vibration reading off and the answer to that is no not unless your vibration reading is massive so here we're looking at a 10 mil peaked peak which is for a 8 millimeter probe about 2 volts at 200 millivolts per mil so at a negative 10 volt average we only have a 1 volt less negative and 1 volt more negative swing so we have a great deal of room to both the upper and lower ok limits the upper limit for 8 millimeter proximity probe is about negative 1.

6 and the lower limit is about negative 16 so as long as I do not cross over into that not okay range the probe will read accurately as long as the system is matched length and for material so here if my okay limit is at negative 1. 6 and we only have a 1 volt difference to the centre line as long as I am over 2 points over negative 2. 6 I'm not going to get into that okay limit so you can have a very wide range of gap setting and not run into issue however we do recommend that you set it in the middle of the range to give you the greatest amount of clearance and available room possible machines with quite large clearances here when you gap that probe they're generally gapped or come in from the top so your shaft sitting down at the bottom you've got your lift oil you have your oil wedge all of those are going to lift that shaft up so even if you gap it and perfect negative ten volts when the machine isn't running it's not going to be a perfect negative ten volts when the machine is running I want to switch back to the presentation view now we do recommend that you run a calibration on your proximity probes once a year we also realize that you guys are in the process of or in the business of making money and to make money you got to have the machines running and you're not gonna take them down every year just to check the proximity probes so we do recommend that you check them anytime that you have access to the probes to check them you take the probe out of the machine you put it into a spindle micrometer cradle and you basically set your spin or micrometer to zero you but your probe up against the spindle micrometer target and lock it down the target does need to match the material of the shaft and then you have a meter connected to the proximity looking at the gap voltage or the DC voltage and you record that as you roll the spend drop spindle micrometer out so every ten thousand are every 10 mils you will record the gap voltage and our guys use a Excel spreadsheet that's set up to graph and give you a nice linear system if everything is functioning as expected and we also have it where it calculates the average scale factor for a eight millimeter probe should be in the 200 millivolt range per mil but the specification allows for 188 millivolts per mil to 212 millivolts per mil most of the time I've seen them in the 195 to 205 range which is within spec you can get very strong changes in your signal if you have the improper cable length so if you have a five meter proximity and you have a eight meter extension cable and a one meter probe for a nine meter cable length you would get a long setup and that will read much less vibration than is actually occurring if you have a short system where you say have a 9 meter proximity and you have a 8 meter extension cable and 1/2 meter probe for a total system length of 8 and 1/2 meters you will get more vibration being reported than is actually occurring so something very important to check and the longer the system the more important it becomes we have a mod for the 8 millimetre probes to allow up to a 13 meter length and we had a customer that used that mod and they ended up with a 12 and a half meter system so they got a 12 meter extension cable and a half meter probe they should have installed a 1 meter probe and there was a 30 plus percent difference from actual vibration to what was being reported and since it was short it was reporting 30% higher vibration then the machine was actually showing also your target material has a big effect on the linearity and the scale of the signal our standard proximity is calibrated for 4140 stainless steel any other alloy of stainless any other material in a plating material anything like that you need to change the calibration of the proximity I think the last time I looked we had somewhere around 240 different calibration mods that could be applied to proximity pumps are probably one of the most common places that I see where they change the shaft material on a regular basis due to them trying to fight corrosion Sood anytime you place the shaft make sure that you confirm what the shaft material is and get a proximity to match it can throw the signal off a good bit by scale and it in certain alloys it actually causes kind of an S type shape to the the curve so it's not even linear so you can't just say oh well you know it's a hundred and sixty-five millivolts per mil so making sure that the proximity is calibrated for the shaft material is very important some of the common pitfalls crosstalk if you get two probe tips too close together the radio fields can interfere with each other and that causes fault signals if you're measuring a small shaft and having the two probes 90 degrees apart or if you're have a key phaser in the same area you may need to shift one probe actually along the rotor to just prevent the two from interfering with each other generally you want to have at least two to two-and-a-half times the probe tip diameter between the center lines of the probes so if you're talking about a eight millimeter proximity you want at least 20 millimeters between center lines another issue you run into is what we call sideview this is where that radio field picks up material from the side on installations we recommend that you have a relief cut near the shaft if the casing comes near to the shaft that is that two-and-a-half times the diameter or the tip and you can actually get to a negative 10 volt gap setting by side view so one of the things that you need to do is make sure that the probe actually gets down to the shaft and is looking at the shaft not looking at casing material one of the ways you can do that is to use a plastic copper or bronze rod insert into the hole see where the shaft is measure that on the the probe put yourself a little mark and then roll it in to that to no more than that depth also monitoring the DC component of the signal from the proximity with a meter rolling it down to a negative two volts you can't get to negative 2 volts with side view so if you get a negative 2 volts you know that you're looking at the shaft and then you can back it out to your negative ten one other thing that I've seen people do is run the probe all the way down till it touches the shaft feeling for that extra tightness as they roll it down and that can lead to cracking of the tip and which damages the antennae allows oil ingression things like that that just destroy the the probe so again we recommend that you just roll it down to a negative two volts and then back it out to your proper gap setting you also need to make sure that you have an appropriate target size if the radio field is wrapping around the sides it will not read accurately so we have a minimum shaft diameter and also a minimum flat target size you do also need to make sure that that flat target is available all the way around the circumference I had a customer that was retrofitting thrust probes on her machine they found a nice flat machined surface and they built their bracket but what they didn't think about was that machined surface was over the bolt holes for the coupling and as soon as they rotated a couple degrees a bolt hole passed under the what was supposed to be a thrust reading thus invalidating the thrust so they had to relocate it to a different spot it was just one of those didn't think about the machine actually rotated moments proximity probes for rail vibration generally get installed in two methods one is an internal and the other is an external mounting internal mounting is usually inside the casing and uses some sort of brackets to hold the probe to the bearing cap or the bearing face having the cable tagged down every 12 to 18 inches does keep the cable from flopping around getting physical wear and also picking up stray current and inducing a electrical signal that is not vibration and giving an erroneous signal so making sure that those are tagged down can be very important generally it comes out through some sort of oil seal and into a junction box if you are running to running into an issue where you have oil wicking through the cable say you have a positive pressure casing if you start getting oil in your junction box we have a fluid lock cable that has silicone injected every 6 to 12 inches to act as an oil dam to keep that oil from wicking back up the insulation in the cable so if you have that problem switching to the fluid lock cable should eliminate that for you couple issues considerations for doing the the brackets you don't want it to be more than six inches from the face that are from the bearing you want to use at least two bolts to keep the the bracket from rotating you also want to make sure that the bracket will not resonate or vibrate due to any condition machine doing a resident study designer review of the bracket just to make sure that it's not gonna cause you any issues and have to open a machine up to either replace or reinforce a bracket that can get a bit more tricky if you have redundant probes and/or have the shaft absolute setup where you may have accelerometers on the same bracket as the proximity probe and you just don't want to have that flexing or moving around so the next method is the external mounting where the probe is basically mounted through the housing using a housing so mounted through the casing using a housing generally it's a 3/4 pipe thread that is cut into the machine case this portion here is the lower housing this is a stinger that goes through the probe in the stinger is replaceable it's just a reverse mount probe and you've got about two inches of thread in here to gap the probe and if we have the measurement from the outside of the machine case to the shaft we can pre-cut and thread that stinger for you or if you want we can supply you with a long stinger that can be cut and then internal threads tapped into the the bottom of the stinger to mount the reverse mount probe we offer two varieties are two styles the 31,000 32,000 is a composite housing it is API 670 compliant and then we have the explosion-proof in twenty one thousand which is aluminum and the 24 701 which is a stainless steel fueled wiring we with standard wiring we have a thousand foot limit that is with 18 gauge wire in a non hazardous environment and looking at a frequency range in the 0 to 10,000 Hertz because all of these things affect how the signal degrades through the cable you can go over a thousand feet but we do need to do some calculations with the frequency hazardous area certain classification using things like barriers can actually increase the available usable cable length and going to a better capacitance cable can also extend that range so we've we've got a calculator so if you need to go greater than a thousand feet from the proximity to the monitoring system get in touch with us and we can help you sort out what kind of cable you need to source to to do that so here again I mentioned at the first that we had the the blue bits here we've got the purple bits so this is a different set up and so I believe this is either the eleven or twenty five-millimeter I can't remember what purple is so proximity robe oops and cable would all need to match we offer a five millimeter 8 millimeter 11 millimeter 25 millimeter and 50 millimeter tip sizes we also have it available in a ceramic capped for encrusted for corrosive environments we have the narrow side view version which we discussed in the pitfalls we offer a radiation resistant variant we also have the fluid lock available and we also have a underwater series of probes that have greater water ingression resistance one of the questions that comes up quite often is the 3300 versus the 3300 XL if you see the the gold connector on the cable that is 3300 XL the 3300 s had a silver connector the 3300 8 millimeter is compatible with the 3300 XL systems so if you have a 3300 probe a 3300 XL extension cable and a 3300 proximity the system will work just fine so the the components are interchangeable but the 3300 XL has the the gold plated connectors that have click lock with the 3300 you had to go finger tight on the connector and then another quarter turn with pliers the 3300 Excel has the click lock so it has a ratcheting self lock to keep it from loosening overtime the 3300 had a center core the insulator a shield which was the the negative and then just a outer jacket the 3300 XL increases that protection to having a dual jacket another metal mesh shield and then an outer jacket this helps with abrasion resistance and RF so the 3300 xl's are much less susceptible to noise and damage all right we're going to stop there for any questions on proximity systems okay here's the top one five millimeter and eight millimeter probes shouldn't be gap at - nine volts they can entirely be gaped at - nine volts as long as you are in the air it depends on what your full scale were expected full scale ranges but generally most machines aren't going over zero to ten scale range so as long as you're in that negative 2.