I've never used ultrasonics before but would like to know the following.
Is there a device on the market compatable with RBMware 2120 & 2130
Is a spectrum collected and trended like VA
If so can fault sets be run over the spectrum as in VA
A client has up to 30 ceiling mounted axial flow fans providing ventilation in an underground mine on a transport road. If data can be collected from a distance (no obstruction to traffic) ultrasonics might solve the problem. Can you see any pitfalls in this solution
Your advice and opinions on this matter are appreciated.

I'm not sure, but I believe CSI will sell you an unltrsonic gun. If I'm not mistaken, and they haven't changed their ways, their product is the only thing that is compatible with RBM Ware anyhow.
If you're going to try from a distance, you may have to purchase a parabolic dish for aiming too.
I have used them from a distance, but it is hard to get anything for sure without a dish from 100' or more.

Dave

That's right Dave . CSI had their 7100 SonicScan,working with RBMWare (Sonic View).
If you are using the SonicScan envelope detector output, you can perform an FFT by using a 2120. That's all I know about this product (beside that I many times was working "inside" the scan.)

G'day Jonesy,
what are your anticipated failures modes?
Ultrasound can do so much and there are various ways of being "compatible" with rbmware if it is really really necessary.
I would like to learn more about the application before going in much deeper.
best regards,
Tom Murphy
ASNT Level 2 Ultrasound Trainer

RRS is exactly right. CSI makes an ultrasonic called the CSI7000 & 7100 that can hook-up to RBM ware. Since CSI is closed archectecture software no other unit can can work with the CSI. However, if you use any airborne ultrasound instrument and use the headphone output your results will be virtually the same.
I have used UE & SDT, to name a couple with most of the CSI units as well as the COMMTEST and most other vibration instruments.
For your particular application you need to be close to the fans. The parabolics do not have a very tight focal point they are rather wide. You may be able to put the laser pointer on the motor but this doesn't mean you are picking up the bearing.
How high are the ceilings? What are you particularly looking to accomplish matters greatly when using ultrasound.
If you have already have an ultrasonic instrument such as the UE or SDT ultrasound instrument and already have waveform analysis. Try them first and let us know your results.
The attachment below is an old file for using a the UE 2000 with the CSI 2110.

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

UP2000_to_Vibra_Collector_instruction.doc (497 KB, 22 downloads) UE2000 to CSI Instrument

quote:

The parabolics do not have a very tight focal point they are rather wide.

I'm not sure what you mean by "tight", but I'll guarantee they are tighter than the rubber snout that comes with most of the air borne instruments.
He's probably talking about a 20' ceiling or so, and being off the haul road to the side, which is probably 40' wide. So 45 to 50' a very rough conservative guess. I'm not even sure what they are mining

D

Tight, being that if you are 20ft away and aiming at a motor 18-20inches round and 25-30 inches long, they will not "focus" on the drive-end or non-drive end bearings. You will hear an "area" that may be too wide to assume which end of the motor you are hearing. If the parabolic has a laser pointer its a little better but still you would be looking at a spot size of 18-24" in diameter or larger. The cross hairs or a laser pointer is just a good reference but does not represent the exact point unless you are very close to the target.
Think about the laser thermometers and the spectral spot size of those units. When I used to sell UE products we also sold Raytek instruments and some customers thought by putting the laser dot on a 2 inch pipe 30 feet away that the temperature they were reading represented the temperature of the pipe but in reality it was an area closer to 2-3 feet in diameter that they were measuring.
The rubber cones you refer to especially with the CSI instrument is not a very good fit. Compared to the UE rubber cone. The UE rubber cone will pull the sensor out of the gun before it will let loose of the sensor, thats tight.
The CSI parabolic is pretty good, although it only has one sensor aimed back towards the dish, the sensitivity at 20 feet would not be as good as the UE parabolic that has seven minature piezoelectric crystals aimed back towards the dish.
You may want to consider the SDT and/or UE units with the help of a "Long Range Horn". The difference of focus is about the same at 20ft., if the bearing is particularly what you are listening to, at 20ft the parabolics and/or the Long Range Horns will only give you the overall sound of the motor. But, if the bearings are in a catastrophic failure or near catastrophic failure you will probably hear a lot of noise but you will not know just what your hearing. The SDT parabolic is about 10-12 inches in diameter and at 20 ft the focal point is even a little larger than the CSI & UE larger dishes. The UE and CSI are 18-20 inches in diameter. So if it is in fact the bearings are what you are looking to hear, you are most likely going to hear a catastrophic failure or near catastrophic failure you need to be a lot closer.
Can you stand below the fans and touch the fan housings without getting ran over by the traffic. If you can use a HOME DEPOT or LOWES telescoping pole. I have modified these poles to hold the contact probes in the past to get a reading.
I would suggest first calling these companies and renting a parabolic and ultrasound instrument first and trying it. Remember as I mentioned you can hook-up your CSI to one of these units as well use the headphone with a "Y" splitter. I will loan you one if need it.

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

Jim,

You are a real pro. Plenty of opportunity and not one sales pitch. Just helpful information.

I hope all vendors are taking notes.

Thanks,

Jonesy,

There are many types of axial flow fans, many of which do not have exposed bearing bousings. You did not specify the actual shortest distance to each fan. If the distance is about 25-ft or less, then I would first consider using an accelerometer with magnet base on a pole, assuming safe contact can be made.

I have detected faulty bearings in axial flow fans on a power transformer using ultrasound. I was about 30-ft away and safely aimed my parabolic ultrasound receiver through the chain link fence. The parabolic receivers that I have made for SDT have no difficulty detecting two sources (say bearings) that are spaced 12" apart over a 50-ft distance. My suggestion is to prove your measurement and fault detection concept before being too concerned about interface to CSI RBMware. Why start to "Trend" something before knowing that it going to be effective? There are probably more differences in performance of ultrasound products than there are for vibration instruments, so buyer beware.

Walt

quote:

Originally posted by Walt Strong:
Jonesy,

You did not specify the actual shortest distance to each fan. If the distance is about 25-ft or less, then I would first consider using an accelerometer with magnet base on a pole, assuming safe contact can be made.

I have detected faulty bearings in axial flow fans on a power transformer using ultrasound. I was about 30-ft away and safely aimed my parabolic ultrasound receiver through the chain link fence. The parabolic receivers that I have made for SDT have no difficulty detecting two sources (say bearings) that are spaced 12" apart over a 50-ft distance.

There are probably more differences in performance of ultrasound products than there are for vibration instruments, so buyer beware.

Walt

Jim,

First of all I agree with Danny. For a salseman, you did offer good advice without a pitch.
But then, I have to agree with what Walt has said too, I used a parabolic dish type receiver with a SDT and a UE that worked well in differentiating between two sources that were close together (100 HP Reliance motor, probably 24" apart) from about 50' off.
I also agree with Walt about the differences in guns too

Jonsey, give us some more info. How wide is the haulroad? How high is the ceiling? What is the closest you can get to the fans?

D

I am vendor neutral. I do not sell equipment.
.......Its a different situation it seems to me. Are we not talking oranges to apples. This man wants to scan something in the overhead, possibly at an angle, most likely a covered housing (obstructed). This is an application may be worth trying but until he can furnish more info, from the mines I have been in, its a long possibly angular shot with too many possibilities for accuracy. I know that I wouldn't want to send someone into the overhead to change out a bearing and be wrong or too soon?
I worked for SDT as the salesmanager, (2001-2004). I drew-up the Long Range Horn and I made the original SDT Lube Adaptor. I own an SDT-170M & 170MD equipment. The current SDT parabolic small dish is unlike the dish that Walt had mande for SDT. Walt has excellent experience in the use of airborne ultrasound, vibration and mechanics than I have. His ultrasound experience predates certainly mine, and he is an engineer. I am a technician with experience. I once told someone that Walt was hearing air leaks when I was still trying to hear the valves on my old 55' Ford Fairlane 292 v8 (in 1969) using a a screwdriver and I'am 56 yrs old.
I love Walt and his contributions to this forum and hopes he never stops feeding us information especially with his experience. He's too smart and wise to get into a shooting match over "ultrasound", me too, I know when to "fold" them and I will fold to him. He did in fact make a parabolic for SDT I found to be very effective, larger dish, and better focal point. I have used both SDT dishes and I prefer the larger 18-20" dish, that is both the Clear Plastic and the Stainless Steel one. You see these dishes are set-up at the factory for optimum range. The UE dish has an adjustment that should only be adjusted by the factory for optimum range or focus. I trust that Walt had his at optimum range and yours too DAN was probably optimum for your application. I have had these dishes banged around from being shipped all over the country many having to be readjusted. Most are shipped in cardboard boxes with little protection. SDT to their credit now, with the smaller compact minutature "little tiny, tiny dish", is packed in a foamed-lined equipment box (just joking).
But as Walt clearly stated their is "more difference in performance of ultrasound products than there are for vibration instruments, so buyer beware".
I still say "Rent" it, try it, and if it works, Glory Haleluah Man! We would all love to hear about it.
Hope to hear from you guys while we are all at the RMC Conference in Daytona Beach. I wish you all could be there in Daytona. I will be there like some others "TWITEERING" from the site. My user name is "jhall.ultrasoundtec", hopefully I will have the whole twittering thing fiqured out by then. You can "follow" the RCM09 at "http://twitter.com/RCM09" to see what is going on.
Hey Guys, remember the questions you ask and what you do with airborne ultrasound is how we all learn about this interesting technology. I can't tell you how many technicians there are that have never heard a 5 psi air leak at 50-ft. Keep the forum's alive ask your questions....God Bless.
Got to go "24" is on tonight.

Jim,

Thanks for the heads-up on "24". My main point on the application question is that ultrasound may not be a slam-dunk solution. As you said, we know very little about the fan type, location-distance, and the environmental conditions. By all means one r more ultrasound products should be tried. I don't rely on trending, and I don't generally use spectrum analysis with ultrasound. The sound amplitude and audible sound quality should be adequate to detect most bearing faults and much more.

Walt

Now I'm a little confused, some say you can focus others say you cant???

Like I said I've never used this technology before and I'll try to answer some questions in no particular order

Tom. Anticapated failures/faults would be the usual stuff for fans. Imbalance, bearings, looseness, motor faults etc.

Walt. A co worker of mine actually has the client so I'm only going off our conversations. The roof is 4 metres high and I'm guessing you may have to stand another 4 to the side so you dont get run over. Co workers opinion is its unsafe and disruptive to production/traffic for the extension pole option.

The client wants to use RBMware for job creation, costing KPI's etc. We also do VA for them as well, so it seemed logical to run the ultrasonics with RBM as well.

Do some people trend and others not? I only assumed thats how it was done.

Thanks for the feed back fella's I'll take some more if you've got any.

Jonesy.

Jonesy,

At a client that had the UE 10000 and their software (software was a pia), we set up a route, and collected data and trended.
We actually found some items from increase in db, and concluded the system worked. However, if you can get a transducer on it, and use your vibe box, you'll get a lot more useful (read that identifiable) information, imo.
The best we could do with spectra looked something like 120k f-max with 400 lines.

D

Jonesy,

Focusing ultrasound is not an issue, but background ultrasound could be. If you want to detect all of the "fan-motor" type of faults that you listed, then ultrasound is not the best measure. You need to try ultrasound before committing to a routine measurement program and doing "costing KPI's"!

Have you considered measuring vibrations with a remote permanent accelerometer on each fan or ride/walk under each fan and make a quick measurement during a shift change or other production slack time?

Walt

I was in the field yesterday and so only today am I getting caught up on the discussions threading from this topic. I'm going to attempt to add my few cents without going off on a vendor rant. My intent is to help explain when and why to use a parabolic sensor VS an extended distance or long range sensor, or other configured airborne ultrasound sensor. Walt... you already know this stuff. We still find your parabolic dishes in the marketplace and you made them for us more than 8-10 years ago. That speaks to the robustness of your design and the fact that they worked.

The parabolic shape of the dish is designed to ONLY receive ultrasound pressure waves that are parallel to it's axis. This is true of our parabolic sensor, and its true of the parabolas you see on the sidelines of football games used to hear the audibles from referees. This means that reflected ultrasound pressure waves, or ultrasound pressure waves coming from sources other than what you are aiming at, will not register on the sensor. Why? Because they are not parallel to the sensor's axis, and the parabolic shape reflects them away. If you open this image you can see what I mean. Back to the football example, without the parabola microphone we would hear much more than just the clear voice of the on field official (who's mouth is much smaller than the prescribed 18-24" focal point suggested by an earlier post). In fact, we wouldn't be able to distinguish the referees voice over the other noises, unless a parabola was used.



There was an earlier comparison made between a parabolic sensor and a spot radiometer. I'm not convinced this is a good comparison as the spot diameter of the radiometer grows proportionally over distance and is a product of the instrument's optics. Drawing parallels between these two isntruments is inaccurate and confusing because a spot radiometer is more effective the closer it is to the surface it's measuring. The opposite is true for an ultrasound parabolic sensor, which was never intended to be used for close up applications. As I said, the parabolic is influenced by parallel ultrasound waves, but to have ultrasound waves that are parallel to the parabolics dish's axis, the ultrasound source must be far enough away from the sensor. So the parabolic dish is mainly effective at great distances and should be advantageously replaced by other airborne sensors for close applications. It is difficult to assign an exact number that defines when to use the paraoblic and when to use a different sensor. Why? Because while ultrasound reception may be theoretically calculable, emissions and propagation of ultrasound waves exist in reality; and reality has too many variables. However, to offer a guideline, it is generally accepted within SDT’s technical department that the parabolic is NOT very efficient when used within a proximity that is less than 2 metres. The Parabolic Dish becomes more efficient between 2-8 metres, and is very good beyond 8 metres. These are not scientific observations, but are the generally accepted guidelines we use, teach, and follow.

The full benefit of the parabolic dish and parallel waves is that it can pinpoint with great precision the source spot at far off distances. SDT has determined that the source spot that the parabolic dish can identify is 5cm @ 15m.

For Jonesy's application, I agree with Jim; rent, borrow, or call in a vendor so you can try this application on your ceiling fans. Don't invest in equipment unless you know the tool will fit your needs (all of your needs, including software issues if you decide to trend).

Since you will have no data available for these fans initially, your going to have to implement comparative trending. Once you've established that the parabolic dish will indeed provide you with the focused data you require, the next step is to collect dB readings from each of the 30 installations and compare them to each other. If 28 of them are within a value range and 2 of them are well outside that range (8-10 db outside) then it can be presumed that 2 fans are "different" than the other 28 and potentially in or approaching failure. Remember, ultrasound waves are produced by friction, rubbing, grinding, metal to metal contact.

One other thing has to be closely followed during the data capture and that is distance between the parabolic and the source. Since ultrasound pressure waves incur some attenuation as the transport through the medium it will be relatively important that all 30 measurements are collected from similar distances. If you are standing 6 metres from fan #1, then try to stand 6 metres from fan #2, #3, #4, etc... I hope this is possible. I'm not concerned if you are standing 6 metres for one, then 6.5 metres for the next one. That's not a big deal. But you don't want to be 6 metres for one, and then 12 metres for the next one.

If its impossible to be within a consistent distance for each measurement point then a calculation can be made to compensate, using the inverse square rule taught in Level 1 ultrasound theory. This rule compensates for varying distances when collecting airborne data. Better to try and keep your distances constant instead of complicating matters with extra math. You may even want to put a mark on the floor to indicate where to stand for the next time you do your scan.

I hope, if nothing else, that the parabolic sensor is a better understood tool.

Thank you Allan for that explanation. It was very understandable. I knew they worked well for me, now I understand why

Thanks again.

Dave

The inverse square rule works fairly well for sound propagation in the audible frequency range, but it is not too good for ultrasound frequencies (about 40-kHz) for distances from source beyond several meters. The ultrasound level will drop more than 6-dB per doubling of distance (hemispherical spreading) because of excess atmospheric attenuation that varies somewhat with humidity. I suggest that anyone planning to measure ultrasound levels beyond 2-meters do a walk-away test and measure the dB level every doubling of distance. Wind can signficantly affect ultrasound measurements outdoors, especially in cross-wind and up-wind directions. It is very easy to experiment with ultrasound, and it is highly recommended!

Walt

You're welcome Dave. Glad my excessively long post had clarity.

Walt, I'd like to understand the science behind your statement about the inverse square rule not being applicable to 40kHz. I do think its a good idea to experiment with ultrasound. There's so much to learn and share.

Bringing focus back to the original post, testing ceiling fans in underground mines, I feel this application should prove a perfect opportunity to test and experiment with ultrasound. Two of the main variables brought to attention by Walt Strong (wind and humidity) should remain constant in an underground mine which can only make for more meaningful data collection.

Good luck Jonesy.

We haven't heard from Jonesy in a while, so he must be Down-Under Down-Under. If the fan is an open propeller fan, then there will be air flow, so both "wind" currents and background ultrasound will be present. It may be possible to detect a significant bearing fault (fan/motor) but unbalance detection is unlikely with distant airborne ultrasound measurements.

The propagation of sound is well described in most Acoustics books. A source that spreads sound in a spherical pattern will have an amplitude decay of 3-dB per doubling of distance; also called the inverse square law because sound level is proportional to distance squared. A sound source located near a reflecting plane will radiate twice the sound power, hence the sound pressure decay is 6-dB per doubling of distance. Atmospheric absorption further attenuates sound.

Walt