Hi guys,

We use a Dyna-bal soft balancer in our shop. Actually, we have three of them. We balance using mils displacement. This is measured in the horizontal where the machine is allowed to float. As an example, if I balanced a rotor to .2 mils at 400 rpm, what would be expected at 1200 rpm? Along the same line, when a customer specifies a balance spec, it is normally at the maximum service speed. When balancing at a lower speed, how should it be approached? A lot of balancing questions lately, so I thought I would add to the mix.

Sean

Shawn,

The movement of the rotor that you see, or .2 mils, is the displacement of your rotor mounted on soft supports, and that motion is caused by the centrifugal force of the residual unbalance left in your rotor.

Although it is said that the centrifugal force increases as the speed squared, so is the rest of the rotor body that acts as a restraining/restoring force also, else your rotor would fly off to the moon.

Now, that is in the case of soft bearing machines. A rotor balanced at one speed would behave the same if the speed was increased.
Simply because the rotor body which is considered in lay terms as the parasatic mass or restraining force would also increase as the unbalance will. But, no significant displacement difference in terms of mils would be noticed.

Considering the same rotor installed in a hard bearing machine, the approach is different. The hard bearing m/c pickups measure the force exerted on them by the unbalance in the rotor. Therefore, for an equal speed as one of a soft bearing m/c, the force will produce a certain reading on your meters and will augment if the speed of the balancing m/c is increased, a larger reading will be shown on the meters.
Resulting in the following statement : If you can't see it, you can't correct for it !

To conclude and answer your question : Higher speed on a soft balancing m/c will not increase the displacement you are reading, in theory.


I think that this answer may raise some eyebrows amongst other members.

Regards,

MarkoLeo

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

Sean, I assume this machine is variable speed? Why not just ease it up in increments of 50 rpm or so, record the data, and plot the results. Do this for several different rotors and you will have your answer.

"Believe nothing you hear, half that you see, but all that you do."

"No matter what they tell you, they're not telling you everything."

Sean,

Rusty's suggestion of increasing the speed is worthy of a try.

But the other statements : about ignoring the knowledge and experience of others that take the time to offer you an honest opinion and refraining to tell you all the thruth, are all out of context on this forum.

I strongly believe in honestly trying to help everyone that has a question or a problem, else this board is useless...

Regards Sean

P.S. Sorry about mispelling your name before.

MarkoLeo

Thanks for the replies. I will give the variable speed a try. So close I couldn't see the forest for the trees. Anybody else care to offer two cents.

Sean,
1. Take a rotor that you have balanced down to a good level.
2. Place a known weight at a known radius on one end of the rotor while locking down the other pedestal.
3. The readout will give you an increase in vibration.
4. Repeat for the other side of the rotor.
5. That vibration will give you the sensitivity of your machine for that rotor.
In other words, a given amount of unbalance at a given radius is the true definition of a balance specification or tolerance. You will then be able to bring the rest of the unbalance down until you have met either the customer's balance spec or one you have adopted from the ISO balance specifications.

Let me throw this out. I believe MarkoLeo is correct. In theory, once I have established the difference between the rotational center and the center of mass, it should not change. I think this is what the soft balancer is trying to measure. I think air flow and system stiffness account for the variations after that. Any takers?

Sean

Sean,
As the other threads have discussed to the point of unconsciousness, there is no way to know what the final vibration will be in the completed, assembled motor, running in situ. This goes for hard bearing and soft bearing machines.
You can only balance to a known specification, which is a given amount of residual unbalance at a given radius.
Knowing that the vibration of your soft bearing pedestal is 0.2 mils at 400 rpm is not an unbalance tolerance specification.
I understand what you were asking about, however the steps I indicated are necessary to understand what the residual unbalance tolerance is.

Sean - you can balance a part using mils as an indicator of your progress, but you cannot use mils as a final indication of the unbalance on the part. For instance, what heavy spot would 0.4 mils relate to for a 10 lb part vs. a 1,000 lb part? You must convert from mils to unblance units such as oz-in or gr-in in order to know the balance quality. And, Markoleo is correct in his statement that a linear soft suspension will exhibit the same displacement at all speeds as long as the rotor does not change. Forces increase exponentially with speed, but a soft suspension does not constrain the part, therefore it is allowed to displace at will due to a given heavy spot.

Sean

Like it has been mentioned; the only way to correctly balance (in a balance machine) and know the actual results are to balance to an ISO tolerance i.e. 4w/n, G1. G2.5 and so on. Does your equipment give information on how many oz. or grams to add or remove at a given location after calibration? If so you can calculate the residual unbalance tolerance when you reach .2 mills. You will probably find out that you may have balanced too much or in some cases not even close to what it needs to be. IRD balancing web site provides endless balancing information. Check it out and it will most likely answer most of your questions.

Most likely the only acceptable shop balance tolerance will be in terms of residual imbalance. 4W/N is not ISO – at least yet. I’ve been lobbying for this.

What does 0.2 mils mean for a soft bearing machine? As stated it measures the mass center of the piece, more or less (long rotor vs. short). The mass center indicates the imbalance.

So, if one has a 10 lb piece (use metric as you like – no problem – The 4W/N is in customary US units.). 0.2 mils pp would correspond to 10 X 0.0002 /2 lb-in (/2 because of pp measurement assumed) = 0.001 lb-in (or 0.016 oz-in) of imbalance.

ISO G balance grades have units in mm/sec. Convert the displacement into mm/sec and you have some type of version of the ISO balance grade (This wouldn’t satisfy most customers.). One would have to use the service speed to find this version of G.

quote:

the other statements : about ignoring the knowledge and experience of others that take the time to offer you an honest opinion and refraining to tell you all the thruth

Markoleo, I think you misunderstood what I meant. The two statements are "quotes" of things I heard many, many years ago. I'm not advocating that anyone ignore good advice, but the best way to "know" something is to "do" it, rather than just believing what others say. Anything that can be easily verified, shoud be.

As for, "they've not telling you everything," that originally pertained to the idea that when you are investigating a problem, or troubleshooting something, even well meaning folks will leave out important facts. Of course, some people (with something to hide) will intentionally not tell you things.

Here of this board, even the best respondents leave things out, or (like me) sometimes just don't say things the best way.

Again, anything that can be experientially verified, should be. (IMO)

Rusty,

I am in agreement with your statements in the context of your explanation.

Keep up the good work.

Regards

MarkoLeo

In my ignorance or stubbornness I think I can balance to a specification in mils. The mils reading is directly related to the unbalance divided by the rotor weight. If your unbalance spec is G6.3 @ 1800 rpm on a 1000 lb rotor, the residual unbalance in oz-in is 21.05. The center of gravity displacement is 21.05 oz-in divided by 16000 oz or 0.0013" or 1.3 mils. I can run the numbers for any spec and rotor weight you have and it corresponds directly to the chart in an old Schenck Trebel book. It lists the G balance specs on a chart with the X-axis being maximum service speed, and a Y-axis of "Acceptable Residual Unbalance per Unit of Rotor Mass or Center of Gravity Displacement".

Back to my original question. My customer has a balance spec of 10 gram inches per plane on a 350 lb rotor. The rotor has a tapered bore to mount it on the end of a compressor shaft. No specification on keyway fill, no specification on allowable shaft runout, etc... To balance, I need to mount the rotor on a tapered shaft and balance to 0.00012 inches. Or 0.12 mils. Any runout on the customer shaft and I'm out of spec. If the keys don't match, I'm out of spec. They balance at 393 rpm in a Schenck CAB720. But this is a G0.4 spec. I would think that would be an in situ balnce requirement.

There is my most of my story.

Thanks,
Sean

Now that you have a "balance per plane" requirement, how do you relate vibration to mass eccentricity? A rigid rotor (body) does not simply spin about the mass center; it also has an inertia 'eccentricity' to be balanced.

The balance tolerence sounds close to what API wants, and people can achieve this generally.

API is 4W/N in oz-in, equivalent to a balance grade of 0.665.

quote:

Originally posted by William_C._Foiles:
What does 0.2 mils mean for a soft bearing machine? As stated it measures the mass center of the piece, more or less (long rotor vs. short). The mass center indicates the imbalance.

Bill,
I quess, a soft bearing machine is a close approximation of an ideal case when a rotor is rotating free in space or suspended horizontally with two ropes with zero stiffness in lateral direction (in horizontal). In this case if center of mass is orbiting at 0.2 mil radius the resudual unbalance is: U = RotorWeight * raduis. If this is this correct, then how soft has to be the bearing in order to produce acceptable error?

David

David

There is only one center of mass for a rigid rotor, and you have two bearings not collocated with either bearing in general. The rotor has an imbalance of the inertial axis, in that the line of rotation for an unbalanced rotor does not coincide with a principle axis of ineria - this is where plane seperation comes in to help or hurt, in balancing this moment component.

The rotor could have a couple imbalance about the center of mass, yet have the center of mass on the axis of rotation (at least in theory, acheiving this in practice is something else). In such a case, one still has vibration at the bearings (or supports), but the center of mass has no eccentricity. There is a moment imbalance.

Spin a rigid rotor with just a couple imbalance about the mass center with soft supports, and the vibration depends upon how far from the mass center the measurement is taken.

[QUOTE]Originally posted by William_C._Foiles:
Now that you have a "balance per plane" requirement, how do you relate vibration to mass eccentricity? A rigid rotor (body) does not simply spin about the mass center; it also has an inertia 'eccentricity' to be balanced.QUOTE]

I'm not sure I understand. The rotor will spin about the bearing rotational centerline when in service. The shift between the rotational centerline and the mass center of gravity line causes the vibration. I agree those two lines could meet in between the bearings and give you a couple only unbalance.

Sean

Imbalance type forces in a rigid body spinning about a line is a centerline are a function of the imbalance and whether the line of rotation coincides with a principal axis of inertia. To balance a rigid rotor one corrects the mass center and aligns the inertia axis (like with a couple shot). Two properties of the rigid body affect imbalance, the mass center relatiionship to the axis of rotation and the principal axis of inertia relation to the axis of rotation. Either or both can cause imbalance forces.