Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #1 
I monitor a kiln drive gearbox which I have essentially no information on.  Have taken 4 sets of data now (quarterly) and prior to last week the gearbox has always been very quiet.  But last week it was audibly noisy, and listening to the accelerometer, the noise was quasi-random.  I am thinking excessive clearance in a bearing, or bearing-to-housing clearance, and a gear is "bouncing around". 

The H.S. shaft speeds have been 110, 245, 276, and now 418 rpm.  The output speed was 9.83 rpm @ 418 rpm input, though these numbers may vary slightly since it is driven via a VFD motor and then a belt-driven jackshaft.

I am fairly certain the H.S. pinion has 23 teeth as this frequency is very prominent; there are 166 cpm sidebands about the 23X component.  I believe it is triple reduction though I can't see much of  the 1st intermediate as it sits directly below the H.S. shaft and there are guards covering it up (this is an MSHA site).  Can't get to the bearing covers on the 1st intermediate either.

John, can you calcualate ballpark ratios/tooth counts or point me towards rule-of-thumb ratios?




__________________
"The trend is your friend"
John from PA

Sr. Member
Registered:
Posts: 963
Reply with quote  #2 
Quote:
Originally Posted by RustyCas
I am fairly certain the H.S. pinion has 23 teeth as this frequency is very prominent; there are 166 cpm sidebands about the 23X component.  I believe it is triple reduction though I can't see much of  the 1st intermediate as it sits directly below the H.S. shaft and there are guards covering it up (this is an MSHA site).  Can't get to the bearing covers on the 1st intermediate either.

John, can you calcualate ballpark ratios/tooth counts or point me towards rule-of-thumb ratios?


I'm afraid I can't be of much help in this case due in large part to the possible triple reduction.  My software only can handle a double reduction.  I think I did one of those for you once and it took about 2½ days (seriously!).

Probably best to go back to the OEM or possibly check an onsite parts warehouse for some spares.
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #3 
John, do I assume correctly that as you move from input to output that the pinion teeth decrease in number, while the gear teeth increase?  I'm coming up with 23, 19/57, 15/63, and 77.  Is that a reasonable arrangement?  I'm just looking for gearmesh/speeds to look for in the data.  This gearbox is so old, we couldn't find a nameplate, they can't find a manual, and there is no spare.


Input Speed:416.991  
    
 PinionGearSpeed
High Speed23  
1st Intermediate1957168.2595
2nd Intermediate156350.74494
Output 779.885377
Measured Output Speed:9.83  



__________________
"The trend is your friend"
John from PA

Sr. Member
Registered:
Posts: 963
Reply with quote  #4 
Quote:
Originally Posted by RustyCas
John, do I assume correctly that as you move from input to output that the pinion teeth decrease in number, while the gear teeth increase?


I can't confirm the generalization that the tooth counts would vary as you state, although I'm sure some drives exist where that might appear to be the case. Generally one looks for total ratio but strives to get a good balance in durability rating in any given pair of meshing gears.  It would not be desirable for instance to have a pinion with a durability rating of 500 HP meshing with a gear with a 200 HP rating.  The gear could wear out well before the pinion. 

Quote:
Originally Posted by RustyCas
I'm coming up with 23, 19/57, 15/63, and 77.  Is that a reasonable arrangement?  I'm just looking for gearmesh/speeds to look for in the data.


My issue with this set of tooth counts is a 15 tooth pinion.  I'm assuming this is either helical or spur gearing.  In such a case, for a pressure angle of 20 degrees, the minimum number of teeth is 17.  Lower than that you need to have an undercut tooth form which can result in strength considerations.

Let's back up a bit.  You had mentioned you were reasonably sure it was a triple reduction.  That can be confirmed by comparing the input and output directions of rotation.  Have you done that?  The total ratio for your RPM's is about 42.5:1 which could readily be obtained in a double reduction.


Ralph Stewart

Avatar / Picture

Sr. Member
Registered:
Posts: 235
Reply with quote  #5 
Do you have any data showing more frequencies similar to the (23T)(418cpm) frequency and 166 cpm sidebands?

Sometimes it is fairly easy to use data to "reverse calculate or back up" and figure extremely close to the tooth counts and shaft speeds. Close enough to pinpoint the problem.

Just my opinion and I could be totally wrong. [smile]

__________________
Ralph Stewart
AlertAnalytical.com
John from PA

Sr. Member
Registered:
Posts: 963
Reply with quote  #6 
Rusty, any more input relative to the design of that gearbox?  My question, repeated, are you sure it is a triple?
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #7 
John, thanks for your always valuable insight and information. No, I’m not sure it’s a triple. And I’m sorry but I don’t have any more information. This site is 4 hrs away and I only get out there once a quarter. I have to have an escort there, so collecting data is always more rushed than I am comfortable with. They seem to either not have much information on their equipment, or are maybe too busy to find it and get it to me. It is a very old plant, bought by new owners a few years ago and being dragged kicking-and-screaming into the 21st century. Vibration analysis is sort of a novel idea for them, but they are slowly becoming believers. The plant manager (who brought me in) is very forward thinking and is a great supporter.

I will definitely get more info next trip, perhaps staying over an extra day to do some research and see what kind of info they have available.

__________________
"The trend is your friend"
Danny Harvey

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,403
Reply with quote  #8 
Rusty,

If you can get one, a picture might help us narrow down the manufacturer, too.

John told me that on Philadelphia Gear drives (which were very common back in the day) the serial number is punched into the bolt flange in the horizontal split of the casing. We found it but had to do it on a down day because it was right next to the input shaft and covered with dirt , grease and rust.

Others have their own look.

Measuring the shaft diameters and center distances both horizontal and vertical can also help narrow it down.   

You should be able to get at least the high speed gearmesh frequency from your data. That will give you the starting point.  From there, look for sidebands around things that look like gmf in the ranges you suspect. You might try switching your spectra from velocity to accel or even using a log scale to make the sidebands show up.  

What version are you running? If you can send me something in 5.42, I'd love to take a crack at it.
Vibe-Rater

Avatar / Picture

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,398
Reply with quote  #9 
I'd like a crack at it too Rusty. I run 5.61 but that has no problem opening an earlier version. In my past I have "reverse engineered" quite a few gearboxes - up to 3 stages, 4 shafts. As you will know. Speedvue input and output to check proper ratio. then high speed shaft gearmesh and work from there.  I normally make a guess for hidden shafts at a ratio of 3 or 4 : the previous shaft.  Meaning slower speed. After that it is a numbers game and chasing frequencies throughout the box. rgds
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.