November 29, 2011
Troubleshooting A Compressor Alignment
By Patrick Lawrence
We’re going to try something different this week. We see some interesting stuff out there. We want you folks to troubleshoot this one.
The OJT portion of a recent alignment training class involved checking the shaft alignment on an engine-driven reciprocating gas compressor. One of the great things about this part of the class is working on real equipment in the field and experiencing first-hand the challenges they can present. After selecting the tolerances and inputting the dimensions, the alignment condition was assessed and it looked great. All angles and offsets were well within tolerance–no adjustments needed to be made. End of story? Not quite.
Since we had the opportunity, we went through the steps on how to correct an engine on chocks. The XA Ultimate was already in live mode and the sensors were approximately horizontal (there was some guarding in the way). As we turned the shafts back to 12 o’clock, our beautiful alignment kept getting worse and worse. The engine’s vertical offset was becoming more and more negative. When we stopped at the vertical position our vertical offset had changed from ~-4 mils to ~-11 mils. Concerned looks were shot amongst the mechanics.
What do you think was going on and why? How would you verify your suspicions? Any test you’d like to try? Did I give you enough info? Mull it over, leave your comments or questions below and I’ll give you the rest of the story in a few days.
A few days later…
Thanks to those who replied offline. I’ll keep encouraging you to post your comments below so we can all participate.
So here’s what was going on.
As some pointed out, the problem is rolling the engine/compressor back–opposite of how we collected the data the first time around and opposite the normal rotation while operating. Consistency is the key to getting repeatable results. If you operate the machines you are aligning in a consistent manner, the results will be consistent and repeatable.
Remember, we were watching live data from the engine’s perspective. As we rotated the engine back to 12 o’clock and saw the numbers changing, we pondered how the engine could be dropping OR how the compressor could be rising. Sleeve bearings throughout and a six-cylinder compressor means there’s a lot of possible movement when rotating the shafts. Could the compressor shaft be riding up the side of the bearings? After kicking it around a bit, I had them try something–we continued rolling the shafts to ~11 o’clock and then back to 12 o’clock. The alignment fell back into place. We had the exact same vertical results as the first time. This confirmed our suspicions of the compressor shaft movement and reaffirmed their faith in just how accurate and repeatable the XA system is.