Learning Geometric Measurement

While at a recent maintenance conference where I was conducting a workshop on fundamental maintenance, I was asked about the future of machinery installation. My immediate response was that geometric measurement is the future.

Geometric measurement is the measure of straightness, flatness, squareness, parallelism, etc. The response from the participants was that this was nothing new.

“It’s true,” I agreed, “but the fact is that the instruments we use to measure these parameters today are more cost effective for users to buy, as well as being very user friendly, so you don’t have to have an expert on staff.”

However, the main reason why geometric measurement is the future is what we have learned in the past. Let me explain.

Let’s face it, back in the dark ages, when we used just a straight edge to do coupling alignment during the installation of a machine unit, it was very much a hit-and-miss process. We didn’t actually learn anything at this time because we were ignorant of what was happening between the machine shafts. We simply got them as close as we could with the straight edge and then clamped the machine down.

Then, after the industrial revolution, some of us started using dial indicators and we were able to get a quantifiable measurement, which allowed us to work to a tolerance.

For many, this method was still trial and error, because we used to ‘bump’ the motor to see if we were going in the right direction or not — unless you had some training in graphing or used a math formula. Still, the biggest challenge in this era was moving the machine and keeping it in place while it was locked down. One of the things we learned with this method of shaft alignment was that, in many cases, the machines would move while they were being tightened down.

The lasr beam

I remember setting a machine in a misaligned position and knowing that when I tightened it, it would move into alignment. Basically I was compensating for the machine’s movement. However I was still unaware as to why the machine was moving.

Then the light literally fell upon the righteous in the form of a laser beam and we started to use laser systems to do shaft alignment work.

These systems dramatically shortened the amount of time it took to perform an alignment, however we were still pulling, pushing holding and clamping the machine into position until we had a favourable result on the display unit; then we locked it down. The process quite literally involved using come-a-long to pull piping over, or tightening jack bolts to the point of bending or breaking.

It is a fact that many machine units fail prematurely due to the strain that they are left under during the installation process.

If you think I’m wrong, consider this: Theoretically, you should be able to take a pump and motor machine unit out of service and loosen all the hold-down and jacking bolts. Then re-torque them and put the machine back into service. If there is no stress, the machine will not have moved and you can confidently push the start button, putting the unit back online.

But would you risk doing this procedure?

Remember, when installing a machine unit, the goal is to create a stress-free environment in which the machine can operate. This will guarantee the optimum life for the machine.

With the advent of laser systems we learned so much about the machinery installation process. We learned that problem issues such as pipe strain, twisted bases, soft foot and coupling strain, etc., have a huge impact on the installation process, resulting in machinery movement when tightening the hold-down bolts, as well as shaft deflection and more problems. These issues have to be corrected before we can attempt to align the shaft. We know this because when we remeasure with the laser system, the measurement results will not repeat.

I’m not saying that we didn’t know about these problem issues before, but because of the simplicity of modern laser systems, it is very simple to recheck your work and if the results are not repeatable, then you haven’t done an adequate alignment job. The reason is usually one or more of those problem issues.

Some companies have made tremendous improvements with many of these problems, such as pipe and coupling strain and soft foot. However, twisted bases and thermal growth are generally still ignored; dynamic movement is still very much an unknown problem.

Measuring soft foot

Most laser systems come with a soft foot measuring program, but using them requires a word of caution. The majority work by measuring the shaft deflection when tightening and loosening the machine’s hold-down bolts. Soft foot is a condition that occurs when the mounting feet on the machine do not lie in the same plane as the base to which they are bolted. This is because a foot is bent or twisted. Because of this, the measurement taken on the shaft may not be as accurate as you may wish, so I suggest that you just use the laser program as a guide and confirm the reading using a simple feeler gauge.

You can spend a lot of time finding and correcting soft foot and it can be frustrating, because you can create soft foot just as easily as you can correct it. You do this by adding too much shim when you are correcting it. A good tip to remember is to always be a little under rather than over when adding shim.

Each of these issues — soft foot and so on — has to be individually identified and corrected. The process of doing this is too detailed to cover in one article but here we can consider the following example, based on a hypothetical question on the subject.

Typical scenario

You are installing a very large motor onto a base; the motor weighs one ton and the distance between the front and back feet is 48 in. If the base frame was twisted by 0.010 in., would you expect to find this twist using the traditional soft foot method — by tightening and loosening the hold-down bolts? Would the foot spring up? Remember, this is a twisted frame, not a short or bent foot. Also think about the structure of a motor.

If you think the answer is no, does that mean you would assume that there is not a problem, or would you try to measure the base in another way before installing the motor?

The point that I’m trying to make is that without the knowledge and the appropriate measuring instruments, we are pretty much unaware of the condition that the machine’s base frame is in. It’s much the same way that we were unaware of the problems with the machinery installation process back when we were just using straight edges.

I believe that the foot will not spring up when measuring a small amount of twist on heavy motors with lightweight frames. Look at Fig. 1. There is a 0.012 in. gap under only one of the motor feet. As you can see, the machine’s base, which was fabricated in-house, is very robust. The base has been grouted in place and it is actually two solid 4-in. x 6-in. rails that have been milled flat.

If you were to torque down this motor, the midpoint where the bearing is would be pulled down 0.006 in., thus creating a 0.006 in. offset between the motor bearings, resulting in internal misalignment. Bear in mind that the standard shaft alignment tolerance for a 1,725-rpm motor is 0.002 in. for offset.

This scenario opens the door for more questions. For example, is it the new motor frame that is twisted? If so, simply packing shims under the foot will not cure the problem. Is it the new base that is twisted? Is it only out by 0.012 in.? How would you know? Obviously you would have to measure using a laser or Total Station, etc. (to my surprise, the motor was the problem in Fig. 1).

Back to the future

Now let’s get back to the future. I keep hearing about the high cost of energy and how we are looking for alternative solutions to this problem. My guess is that more and more companies will augment their energy supply in some way, such as wind power, similar to E
uropean industries. Although I seem to digress with this topic, please read on as this theme is relevant to our discussion.

If you were in the wind-power business, you would know that the installation and maintenance of these machines must be taken very seriously. I’m not just talking about the turbine. I mean the whole machine, starting with the tower base, which has to be flat. This has probably got something to do with the fact that it’s a very big machine and users don’t want it coming down unexpectedly. These people document the installation process so that they can have a written record, it’s that important to them. They also have to produce documentation of the measurement results. Each section of the tower will be measured.

Now let me ask you this: Is this machinery any more important than the critical equipment in your plant? Do you have recorded installation records?

This past summer I did some machinery installation work as an advisor for a major automotive manufacturer. The work involved the installation of three pump-and-motor sets. I was shown the company’s machinery installation specification for pumps and motors, which for shaft alignment was 0.002 in. of offset and 0.0003 in. per inch for angular. It also used a 0.001 in. flatness tolerance for the base. I found this very interesting because very few companies have written documentation for shaft alignment tolerances and even fewer have a tolerance for the flatness of the base.

It looks like the future might have arrived.

John Lambert is the principal of Benchmark Maintenance Services Inc., Pickering, Ont. He can be reached at 905-509-6522, by e-mail at info@withinspec.com, or visit www.withinspec.com.