Ok, now I am really getting a headache. In a earlier topic we discussed velocity meters versus accelerometers with slow moving equipment. It was said, and I believe it, that velocity meters are better on slow moving stuff. I work in a paper mill so we are talking dryers and such.

Well, I got to go to the level one training with TA and the instructor says use accels because with peakvue you get a earlier indicator of problems.

Now, I am really confused. Should we take readings with both? Who is right?

Aub

After many years working in a papermill, probably might have even been owned by the same company who owns yours, I started out with velocity transducers but changed over after about a year or so to accelerometers for the sake of demodulation and then Peakvue. Never really had any complaints about the accel. Still use them, even though I no longer work for a papermill full time.

Dryers give good data when using the 793 Wilcoxon. Peakvue doesn't work too well from what I understand when using a velocity transducer. But never would I use BOTH in a papermill. My suggestion is accelerometer 100 millvolt.

And yes, a little training can be dangerous, especially from someone who thinks they know more than really do, after receiving this "little training". Seems like they know just enough to be dangerous, as I always say. I recall a guy who went to a training once and came back and everything that had a vibration, the cause was resonance.

Only my opinion and I could be totally wrong.

Ralph you are so right. A little traning is very dangerous in the wrong hands. I have been running a very successful vibration program at a papermill for 10 years. I have stuck with the velocity sensors because it has been successful for us. I have not lost a dryer or felt roll bearings due to bearing failture in 10 years. When picking a vibration tranducer you need to know what you are looking for in a spectrun. This takes years to understand this. I use the velocity sensor because of speed and assessablity. I do have peakvue points set up on all my dryers just in case I have any question about vibration readings. Another tool for us. I'm staying with the velocity tranducer.

quote:

I'm staying with the velocity tranducer

That what makes life so interesting!! The world is full of different opinions and we all can choose what makes life easiest for us (to a certain degree anyway). I thank God for my freedom of choice and Will!

Aubrey,
PeakVue works well on most dryer cans. I have seen one case where we had problems with it, but you should give it a try.

What mill are you at?

David Eason

D THOM

With all due repect - running a vibration program for 10 years at a papermill is admirable. I also ran a vibration program at a papermill. I wish everyone would take pride in thier job like I see you do - we would be more competive if that were the case.
With that thought understood -
Now when I conduct equipment reliability trainings, my first objective is to develope an open mind when learning.
I hear all the time that its been working good the way we been doing it.
With that way of thinking we will not reach our potenial.
I wonder where vibration analysis would be today with that attitude?
May be I would try a small route with an accelerometer.
I have known Ralph for 12 years and I learn't the hard way not to at least consider what he recommends.
Mark

Mark

The great thing about vibration analysts is that if you lock 5 of them in a room, they will come out with 8 opinions! If they all agree on something, be suspicious.

Seriously, practically any method of monitoring used will detect problems; if you select one and stick with it, you'll be successful. If you change directions every time the wind blows, you'll never get good at any method, nor will you build up enough background data to detect the changes that show something is going on.

Jon
Spintelligent Labs

I ran a little train once; still caught H_l_ when it ran off the track. And as you guess: they wouldn't let me blow the whistle either.

Sam,
Lionel or American Flyer?

Walt

I guess I got enough training to understand that I was confused. :-)

Thanks guys

David,
I am at the big mill on the river in Savannah. :-)

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

quote:

Originally posted by Aubrey:
Now, I am really confused. Should we take readings with both? Who is right?
Aub

****After reading my own post, I got a headache for being so long winded, therefore I have edited my post.****

I prefer accelerometers for route based data collection on tissue and board machines.
I prefer a 100mV/g accel.

*peakvue - its a great tool I rely on.

*Acceleration Waveforms with Velocity Spectra.

I have added peakvue pts on board machine felt rolls and within 6 months caught 3 felt roll bearing with a brass cage design. The cage fingers separating the rollers had worn "away". NOTHING was separating the spherical bearing rollers and peakvue picked up the impacts as the rollers impacted one another, a few had turned sideways. Route data showed no issues.

Peakvue is a wonderful tool that I for one would hate to be without.

Mike

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

quote:

I work in a papermill, dryer cans rotate 1 rpm and I use accelerometers

Hi Mike,

I know you meant 1 Hz here, huh?

Glad to see you trying different things. Always room for improvement, no matter how long we have been doing this stuff, right? I have said it time and time again, this is a never ending learning process we are doing.

quote:

Originally posted by Ralph Stewart:

quote:

I work in a papermill, dryer cans rotate 1 rpm and I use accelerometers

Hi Mike,

I know you meant 1 Hz here, huh?

Yes ! I did, thanks for correcting.

Mike

Aubrey,
Be sure to re-read MikeH's "long winded" post. My experiences with PeakVue run close to everything he said.

I've never been to Savanah, so I'm not sure which mill is by the river. Look in your address book and see if I'm in there; we may work for the same company.

David Eason

David,

I cannot find you in my company's directory.

aubrey.green at ipaper.com

Aubrey,

Find a known problem and try some experimenting.
I once worked on a paper machine that was wired with axial mounted velocity probes from Bently Nevada. These probes worked fairly well, but there were too many things slipping through the cracks. Like you I was originally taught velocity is best for low frequency stuff. However, I found through trial and error that is not always the case. The dryer cans I had turned in the neighbor hood of 100 to 150 RMP. The response for the velocity transducers dropped off drastically below 10 Hz. The bearing signals were often buried in the noise floor down around 6 to 14 Hz. I began using an accelerometer and Peakvue. Cracked inner races were the issue that most often went undetected. Using a 500 HP filter solved virtually all my inner race problems. Several times I found several cracked races using this method where the velocity transducers would show nothing. That made a believer out of me. Low frequency data collection can be tricky. You must understand the nature of the defect and the limitations of the equipment.

As for learning/trying new things.... if you are using CSI's 2120/2130, I strongly urge you to get the headphone adapter and a set of headphones and start "listening" to your vibration signal, especially on slow turning rolls. It is a bit expensive but well worth it, in my opinion. I had a pair of David Clark headphones (identical to what CSI sells) that came with another instrument, so I just bought the adapter from CSI. I also bought a cable from them, but it was not heavy enough and didn't last long... definitely not worth $159. All you need is the 2-pin Lemo connector and an $9 headphone extension cable from Radio Shack (if you are handy with a soldering iron and heat shrink).

The advantage of this approach -- as opposed to "ultrasonics" -- is you hear the raw signal from the accelerometer anytime your meter is powered up (not just when actually collecting a point). It has a switchable 2kHz and 8kHz filter, but I seldom use it. You will see things that don't show up in your 2kHz route spectrum, which tells you to investigate further.

Also, many faults look much worse in the spectrum or waveform than they actually sound. I find that I am less "alarmed" about things I see in the spectrum than I used to be. Other faults "sound" worse than they look in the spectrum which causes me to look more closely than I otherwise would have. Overall, using the headphones, I just have a lot more confidence in my "calls" -- or "no-calls" -- than I did a few months ago.

The only downside is, your ears sweat like crazy, especially in this hot, humid weather we are having right now. I'm sure there are other more comfortable headphones you could find, but the David Clarks also function as hearing protection and are set up for use with a hard hat.

I have to give the thumbs up to the headphones also.
I began using them just to experiment, now I wouldnt want to do without them.
One example is being able to hear carbon seals squeaking in the dryer sections and know that was the reason your peakvue went nuts with spikes and its not a bearing.I can also hear the siphon tubes rubbing.

Both siphon tubes rubbing and carbon seals can cause the peakvue data to change or show spikes, being able to hear this noise is a plus.

Mike

Using head phones will also teach you how not to place your transducer on equipment.

I know plenty of people who'll say those who use head phones are not analyst. If you have a tool that helps use it. I always slept better when I used the phones on my routes.

This message has been edited. Last edited by: Waylon M,

Very neat to hear the positive comments about "listening" to stuff. I did a lot of this, mainly on nuclear / thermal / process heat exchangers about 30 to 35 years ago. It was almost astonishing to experience what comes through, how good the quality of the audible can be, what the ear can do, and how useful it all is.
Looking forward: if someone could come up with an affordable, hand held laser vibrometer, you could "point and measure" (velocity directly)and also "point and listen" (to the same velocity signal). Listening to velocity (given adequate HF response) has a bit of an edge on acceleration. It is actually velocity that the ear responds to. So, if the sound has a mixture of frequencies and/or broader band, a velocity signal would give a truer replication of what the ear would hear, whereas acceleration has a bias for higher frequencies.
Maybe, one day....