In my brief time learning vibration I was told to take vibration readings on the bottom of the bearing on the drive side of our paper machine and on the top of the bearing on the tending side of the paper machine. We only take axial readings. I would assume the load zone (60 inch dryers) would be on the bottom on both sides. I could understand that the smaller rollers could be "held up" by the felts and maybe the load zone would be on top of those bearings.

Does someone have an explanation why we do it this way?

Thanks,

Aubrey

Hmmmm!!!!
This may be another one of those "ole wives tales".

I collected data on 60 inch dryers for YEARS and always took the data at the 2:00 position or 10:00 position, depending on the easiest to reach and never heard of the "top on one side and the bottom on the other".
I also take the data on the head of the cap bolts (if they were visiable and were on the dryer cans, but then again this probable was another ole wives tale ). If a little in doubt of some data I was getting that indicated a possible defect problem, I would take data at other positions to see what I got there, but again never heard that "rule". Not to say it is not correct.

Where it is possible, I take dryer bearings at the bottom in the axial direction on both drive side and tending side. Sometimes it is not possible to acquire data on both sides due to heat, steam leaks, etc. Obviously the best thing to do at this point is to try to convince the owners that they need permanently mounted sensors, etc. Most of the time that does not go far. Sometimes you can use different methods to look at high frequency data( Acceleration, Enveloped Acceleration, Peakvue) to determine which bearing is bad, but there is no idiot-proof setup for all machines.

So, sometimes the velocity vibration transmits so easily from the drive side to tending side that you can find almost all defects on the tending side. The problem with some dryer designs is that there is not a big difference in frequencies between drive side defects and tending side defects. In the case of close defect frequencies, you can take an additional vertical reading on the tending side. If the defect is on the tending side it will show up well in the vertical direction, but if it is on the drive side it will be much weaker.

Maybe that was the original intent of taking the data "on top" of the tending side?

By the way, for smaller rolls the direction of the load zone can be indeterminate depending on felt and web tension. A typical dryer felt may have between 5 to 15 pounds per linear inch of tension. If you have a 2000 pound roll( I think I've seen your machine before) and the tension is at the low range, gravity wins; but if the tension is at the high end, the felt wins.

-David McCall

quote:

So, sometimes the velocity vibration transmits so easily from the drive side to tending side that you can find almost all defects on the tending side.

I have always thought of this procedure as a "Russian roulette" type of data collection. Not to say some can not be found, but "fear" from missing a defect before it reaches the total destruction mode, like ruining the journal when an inner ring cracks and slips for months with out my knowing it.

Of course this is only my fearful opinion and I could be totally wrong.

Hi Ralph,
I totally agree. I would not recommend taking data on only one side of the machine. Thats why I used the words "almost all defects" and "sometimes".

My first VA job had three paper machines. With two machines, the one side method was totally accurate and with the third machine, it was mostly accurate. My employers were very happy with my performance - most- days!

I have learned alot since then. Like most of the people on this forum, I take it personally when my machinery fails without me knowing about it ( which hasn't happened in a long time- I'm probably due). I take a lot of different kinds of data now and I always run controlled tests on new machinery to find out what works best. I have been very succesful with Enveloped Acceleration and studying the waveforms.

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

David,

Which mill are you at?

I am at the one closest to downtown. We used to get a lot visitors here but the new owners don't like company.

Aubrey

Hi Aubrey,
I spend most of my time at the pulp mill in Brunswick. I work for SKF Reliability Systems.

Hi David,

I work at a paper mill in LA California and I am real curious as to what settings you all use. I have two velocity points at 30k and 6k with high resolution. I use enveloping 2,3 and 4 and am fairly successful (3 and 4 works best). I take axial readings at bottom of bearing on both drive and tendings ends. I have tried various setings in the past for one or two years durations, until settling for previous stated settings. Unfortunately ou plant is shutting down. Any one need a vibration analyst?

We take axial readings on the top of the tending side and bottom of the drive side. 80 orders fmax and peakvue. I just got our database to get the peakvue and velocity readings at the same time so I don't lose too much time.

Aubrey

David, eating at the Fourth of May on St. Simon's used to be a hobby of mine.

Aubrey,
Do your dryers have plain bearings or roller bearings,(or Carb bearings)? I take readings on 2 paper machines in Washington. Most have been converted to Carb bearings. I agree with Ralph. We take ours axially at 10 or 2 on the Tendings side and try the same on the drive side. Most of the drive side bearings are cabled out to a box as we have dryer gears and can't reach the bearings.

Sorry to take so long to get to you Jim. We have regular bearings. Our drive side is not wired out yet. As usual that side is hotter than 40 meters of heck. We are working to get them wired. We just put new guards on those steam joints so I will pretty much be locked into getting my readings at 7 oclock on those.

Thanks for all the answer guys,

Aubrey

We try to collect dryer bearing data in the axial direction as near to the bottom as we can get on both sides of the machine. The main objective is to collect data from the same place each time. I shy away from bolt heads, even though they offer a more direct path to the bearing, because the grade marks make for a rocking surface. A flat magnet may move around while collecting data. You could replace one of the bolts with one that has been machined flat. Good luck with that, I can't even get the mechanics to torque them. As for as seeing DS defects on the TS, it happens and with defect frequencies close, it sometimes becomes a SWAG as to which bearing to change. I called a TS bearing once and we found nothing wrong with it. After starting back up, I collected data again to see the same spectra. Though the amplitudes were higher on the TS, the DS bearing was bad. Changed it next outage.

CARB bearings have done a lot to settle down many of our dryers. With 60+ year old machines, the sole plates look like rolling hills and the frames often work loose. Getting rid of the rockers and bolting the TS housings down reduced vibration in some locations by 90%.

Have a good one,
Gary B