I was also going to suggest putting your analyser in live mode in the vertical reading and tap some wooden or plastic wedges in under the frame next to each rubber pad. We had a similar issue with a lamson blower when the rubber pads wore out and got brittle. Until we could get to replacing them on the next outage we got by with using the wedges. It at least got us by. until more permanent corrections could be made.

Ralph, your question got me to thinking so I went out and took some more phase data which I have attached. It looks to me like the motor is doing all the vibration and the outboard end is doing much more. Even when you put the accels next to each other, one on the foot and one on the base, there is a big difference in amplitude but not phase. I tried wedging the base with a bar in different places the other day with very little change. Maybe I should replace the rubber pads with steel so everything will be straight and the same height
and then realign. I am trying to picture the way this machine is moving and I still suspect alignment is a big factor. I did a running soft foot check and made sure every bolt was tight.

P-16_Phase_data_8-23-07.pdf (11 Kb, 14 downloads)

This probably isn't going to help you very much, but one way to make alignment easier is to dispense with the requirement to. We had a lot of problems with balance, alignment and premature bearing failure on units very similar to yours, so we put the fan runner direct onto the motor. A few hours work with an oxy-spanner and a welder. A heavier baseplate can be used for the motor.


They balance very nicely afterwards. Two less bearings to worry about. If you are going to rebuild the base, this might be an option.

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

before_and_after.doc (139 Kb, 36 downloads)

Ian - that is interesting! I can't say I have ever heard of anyone doing that before. At our plant we are timid about making major changes like that. Out of curiosity, did you involve the fan or motor OEM in reviewing that configuration, or just tried it on your own?

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

quote:

did you involve the fan or motor OEM in reviewing that configuration

Pete, this is just my opinion/experience, but most of the problems I see are due to poor design/construction by the equipment OEM's. They are usually the last people I would consult with. But that's just me, and I'm known to be a contrarian.

quote:

Pete, this is just my opinion/experience, but most of the problems I see are due to poor design/construction by the equipment OEM's.

I must be a contrarian, too! Some of these fan designs are less than ideal. Aluminium fan impellers, pop-riveted togther and mounted on flimsy bases, high speed spherical roller bearings with almost no radial load, more axial load than radial, and we wonder why we have problems.

Pete: No, I didn't feel the need to consult with OEM's. I did do some checks to make sure that the fan runner weight wasn't going to be an issue for the motor bearings, but the motor rotor probably weighs 10 times more than the fan. I re-used the same (or made a dimesionally identical) fan runner, with a weld-on taper lock hub, so the actual fan performance is unchanged. I didn't actually see these as major changes, but then I don't work in the nuclear industry, and I suspect I don't have anything like the statutory/legal requirements that you face.

I did say that we are timid ,not that everyone else are cowboys There could be many weeks or months of hand-wringing at our plant before we ever did something like that. I kind of expected that you were going to tell me you hadn't consulted the OEM's, which only heightens the contrast.It's interesting to see how different the atmosphere is elsewhere.

Another consideration for the modification would be whether the new critical speed happens to coincide with running speed.

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

Roll-em, roll-em, roll-em, YEEHHAAAAA

Our site is 70+ years old; half of the OEM's have long since gone bust anyway. I'm in the very fortunate position of being able to make that sort of change as I see fit (subject to the area supervisor being happy to spend the money)

quote:

Another consideration for the modification would be whether the new critical speed happens to coincide with running speed.

I used 20mm plate (3/4") for the new motor base, fully welded to the existing frame and fan casing. Using the TLAR method, and being fairly certain that the original 5mm (<1/4inch) base was near to the critical, I was confident that there wasn't going to be too many issues. Also, I knew I was going to be able to balance it much more effectively. So I didn't worry over much about that.

Ian, just noticed your pics. I like it it. Sounds like my kind of place. Pete, if you are in a nuclear plant, please keep wringing your hands... I really don't want to look south and see anything glowing!

Going back to the beginning of this string...Was this a single plane balance job or a dual plane balance job? I am sure resonance, looseness and misalignment all have a part in this, but consider this: If the future try balancing the fan in two planes.

I had a similar unit go south on me. All indication in phase reading were single plane. I balanced it using single plane and the motor reading went up by 10X. I thought it was resonance, a loose base - come to find out I had an unbalance motor.

I pulled the weights off the fan and "ATTEMPTED" a dual plane balance job on both the fan and the motor to below 0.1 IPS.

In 17 years I never had to do this until then, and never had to do it again, either.

Quick check: uncouple the motor from the fan and check vibe levels/phase.

DanV,
Politically Correctness Alert!!!!!

You made the common mistake we Yankees all make sooner or later. Never refer to a machine/component/situation as 'going South' when you mean 'going bad'.
I made it 30 years ago explaining FFT plots to the plant manager of Scott Paper in Mobile, Alabama (can't go no further South!).
I was explaining the 'haystack' at 2000 Hz as a clear indication that the ball bearing was going South...I stopped immediately, looked at the plant manager and he smiled and softly replied, 'We like to say the bearing is going North!!!' Then he let me continue!!

Thanks DanV for your reply.
I have always balanced these fans in single plane mode but have done it before with multiple measurement planes. The impeller is 36" diameter and only about 3 1/2" in width at the top. In your post you said you attempted a dual plane balancing job on both the fan and motor. Are you referring to dual measurement planes or dual balancing planes? If you were measuring on both the the fan and the motor, where did you put your sensors, and if the motor was unbalanced how did you fix it? I will definitely run the motor uncoupled to see how it does but any motor we put in this position is always higher in vibration than the fan.

Randle,

I balanced the motor and fan at the same time. This would be a dual plane balance job with two measurement points (motor ODE HORZ and Fan ODE HORZ). I was trying to balance everthing in between the two most outer bearings. This particular fan was not an overhung blower, but a center hung unit. Try balncing the blower as a "unit".

By the way, in the past I have seen units having 3.5" in width needing dual plane balancing, on some overhung units. It depends on the phases you are getting. I would use two plane, with four measurements.

Thanks, DanV. I am still a little confused, did you reduce the vibration on fan and motor by putting correction weights on the fan only, or did you also put correction weights on the motor?

Randle,

I placed correction weights on both the motor and the fan. I think (and I strees that "I THINK") I was getting a "cross effect". Maybe someone can shed some light on what cross effect is - maybe a topic for a new string. I heard that term before but never actually seen a machine produce it.

To make a long story short, don't outrule imbalance issues in the motor or coupling until you attempt a balance job on the motor/fan as a unit.

Phase analysis on the motor can help you determine this. But I would try to reduce the 1x rpm peak at the motor by adding some weight to the motor (even as a trial).

Regards,

Dan

Here is a link to a tech paper - check out CROSS EFFECT:

http://www.compsys.com/drknow/aplpapr.nsf/19fcc759cbee3...05eab97?OpenDocument

Randle, many years ago I balanced a thin (about 4" wide) aluminum fan rotor in an IRD B-50 soft bearing balancing machine. I was amazed at how much of a couple it had in it, and how, unrestrained in the soft bearings, it produced a heck of a wobble. I used clamp on weights to do a 2-plane balance, but when I started adding the final weights to the wheel only 1/2" in from the center of mass of the trial weights, it threw it out of balance. I would never have believed it could make that kink of difference.

Since then I have seen several overhung rotors which could not be balanced in the field because of axial runnout in the wheel (usually due to poor fabrication, or the wheel being 'warped' as they tried to get it off the shaft for some reason). So I don't spend much time on these any more. "You need a new wheel" is not what they want to hear, but it's true just the same.