Last month, I discussed why a pump should be aligned nine times and declared that four of those alignment adjustments should be made before the pump is started up. The prevailing, yet erroneous, assumption is that the pump is pre-aligned to a high level of precision in the factory and the initial factory alignment will hold firmly in place during the transit and installation processes.
Driver-to-pump misalignment will cause many expensive issues, and the subsequent cost of operating the pump will be higher. Pay to align the pump upfront, or you will end up paying more later. Field data has consistently shown that without a proper alignment, you may expect reliable operations of one year or less without issues. If you take time to properly align the units, the runtime can be extended to eight to 10 years.
Thermal Rise Due to Expansion
Just because the manufacturer said that some piece of equipment will have a certain thermal rise due to expansion does not mean it will actually grow that much. My experience is it could be substantially more or less. That is why the hot alignment checks are necessary. Note: this comment also applies to cryogenic applications (shrinkage).
How much a pump or driver will rise depends on many factors including how it is mounted and the material involved. Two common mounting methods are foot mounted and centerline. As an example, typical ANSI pump (foot mounted) will grow about 0.001 inches per 100 degrees of fluid temperature above the ambient temperature. So, a pump with a fluid temperature of around 180 to 200 F with an ambient of 70 F will grow less than 0.002 inches and probably closer to 0.0013 inches.
The material of construction will change the total thermal growth a small amount depending on the C value for the material, where C value is the coefficient of linear expansion. Cast iron, carbon steel and austenitic stainless steels in the 300 range all have similar C values. Stainless will grow more than cast iron with carbon steel in the midrange between them.
There are several technical methods and associated instruments for accurately measuring the alignment at the operating temperature. I cannot go into detail here, but the point is to make you aware they exist.
Having worked for two large companies that manufacture turbines, I know that steam turbines, being foot mounted, grow up with both ends fairly parallel to each other. Gas turbines do not grow at the same rate per end and will present an offset. You will see angular differences at the coupling face in lieu of parallel.
Machine temperature is not the only thing that affects change to the alignment during machine operation. Note that foundation design, operating pressure, shaft torque and machine location will also be factors. A major and commonly overlooked factor is the pipe system design. Understand that if the unit is involved in a drive train of several pieces of equipment, that will be a major factor, and I suggest you consult specialists in this area.
Caution note: pipe strain can change substantially depending on if it is full of liquid or not. An alignment with the associated piping empty will likely change when the piping is full.
The acceptance criteria for alignment specifications is directly proportional to the machine speed along with the coupling gap distance. The higher the speed, the smaller the acceptable tolerances. Be careful with variable speed applications and make sure your criterion is for the highest speed that the unit will experience. On laser alignment equipment, one of the setup criteria in the base program is speed. Be careful that you don’t set the base specifications at 1,200 revolutions per minute (rpm) for a 3,600 rpm unit.
Having spent many years in the U.S. Navy submarine service, I know and understand how much the U.S. Navy’s practice of using tapered dowel pins to preserve equipment alignments has carried over to the civilian world. Please note two things:
- First, question if you really need to use a dowel pin (probably not).
- Second, when you take the unit apart, never reuse the same hole and dowel pin. It is a common field practice to drill and taper ream a larger hole in the same spot, which I suppose is an acceptable practice, but I still question the taper pin practice outside of warship applications in the first place.
Checking for Pipe Strain Effects
To check for pipe strain effects on the equipment alignment, an easy method is to place one or two dial indicators to read off of the coupling surface. If you have a laser unit, that also works fine. Offset the two dial indicators by 90 degrees (one at the midnight position and the other at either the 9 or 3 o’clock position) and zero them on the coupling. Release the bolts on one of the pump flanges and look for potential movement on the dial indicator. More than 0.002 inches movement is not acceptable. Repeat the test for the remaining pump flange.
You should be able to duplicate the final alignment readings to test that the procedure you used was valid.
Just because you have a vertical pump and motor with a machined motor support stand does not mean the alignment process can be skipped. I have witnessed several issues with mis-machined motor supports through the years. Never assume and always verify.
On vertical pumps with rigid couplings, I have found that something as simple as the bolt tightening sequence can alter the runout of the coupling alignment. I recommend a dial indicator on the coupling during the bolt tightening sequence to prove your work.
On horizontal pumps with either C- or D-face motor adaptors, the inherent alignment is not perfect because of manufacturing process tolerances. Depending on how the tolerances stack, you could be 0.007 to 0.015 inches off.
Last of all, and my favorite: “A pump is the most expensive pipe support you could ever purchase.” Do not use the pump as a pipe support.
Don’t be surprised when your pump has problems due to misalignment.
The more critical the pump, and the faster it operates, the more important it is to align the driver. A standard vibration test will tell the true story. It is not the pump’s fault if you do not do the alignment.
19 Tips & Common Alignment Mistakes
1. Safety first. Make sure the equipment is locked out and tagged out before you start. You are ultimately responsible for your own safety.
2. Do not waste time and money conducting an alignment on a unit that is already a piece of junk. If the foundation, base and supports are junk, then your alignment will be junk.
3. Before you start the alignment process, check the driver for soft foot first. Check one foot at a time while the others remain bolted. Know that there can be parallel soft foot and angular soft foot. Both types will drive you crazy if you do not know you have the issue. Angular soft foot is also sometimes referred to as “skew.” There can also be the phenomenon of “base soft foot,” where nothing is wrong with the driver foot, but the base has the issue.
4. Except in rare cases, always conduct the alignment with the coupling spacer removed or at the very least with the bolts loose or mostly removed. Realize and understand that you are really doing a shaft alignment, not a coupling alignment.
5. I recommend to match mark the coupling halves so they align and reassemble together. I also recommend that the keyways/key seats for the two shafts be placed at 180 degrees from each other. If the keys, keyways and key seats are all identical, it probably does not matter. They are not identical most of the time, and so this practice will help with balance issues.
6. Rotate both shafts together and in the same direction that they rotate in the operation.
7. Check for the proper distance between shafts prior to starting the alignment and watch for axial play that will corrupt the readings and, consequentially, the alignment.
8. Do not beat on the driver feet with sledge hammers.
9. Normally, you would never move the pump during the alignment process and always move the driver. But if the pump is a new installation and has not been piped in yet, it is acceptable to move the pump. A common mistake is to pipe the pump in before the first alignment is complete.
10. When using dial indicators, be aware of spanner bar sag, and know how to check and compensate for sag. Do not start the alignment without checking for sag.
11. Depending on what technique or equipment type you choose to conduct the alignment, be sure to check both shafts for runout. Should you encounter more than 0.002 inches runout on either shaft, you will likely have other problems. So, correct the runout issue first.
12. Dial indicator shaft buttons must be perpendicular to the surface they are touching.
13. Make sure the dial indicator button is actually touching the surface to be measured and that the travel length is not compromised. This is a common trick played on new personnel.
14. Do not expect a four-decimal place accuracy from a three-decimal place dial indicator.
15. Measure your shims. Just because it is stamped 0.010 inches does not mean that is the true size. In a game where 0.002 inches is the maximum tolerance allowed, understand that 0.001 inches can make a big difference. Keep shims clean and free of burrs and imperfections.
16. Never use more than four shims total (aka stack) per equipment foot and try to keep it to three or less. The more shims, the more spring and margin for error. Do not be afraid of machining custom shims to accomplish this goal.
17. Always use the same number and size of shims in both front feet and the same number and size in the rear feet. The front stack does not need to match the back and vice versa. Do not use a different number of shims (stack size) from side to side in an effort to roll the unit.
18. When using washers with the driver, hold down bolts, use a strong washer material (grade 8) and watch the size (diameter). Many people incorrectly think they can make up the difference in a hole size simply by using a bigger washer. If the washer cups warps at all, then it is wrong, and you will have issues with holding the alignment.
19. If you have to open up (enlarge) the size of hole on a driver foot due to being bolt bound, do not open up the diameter more than the radius of the original hole size. It is also acceptable to machine the hold down bolt to achieve more space, but the procedure must be followed correctly. I am sorry I do not have room for that procedure in this article. A good practice is to keep close to the minor diameter, maintain the symmetry and provide a generous radius as stress relief.