February 13, 2020

Balancing How To #6 – Balancing Tolerances

By Stan Riddle


By now, if you’ve kept up with the VibrAlign balancing “how-to” blogs, you should see that balancing is mostly just a math problem – counteracting forces.  But we haven’t discussed when the balance job is through.

When I started balancing fans many years ago, I stopped balancing when:

  1. The fan owner said, “that’s much better”, because the machine wasn’t bothering them anymore
  2. When I got the fan below 2 mils (PK-PK)
  3. When I couldn’t get it any better

None of these are very scientific, so let’s look at some specifications.

ISO 1940/1 covers balance grade qualities for many different types of machines – from slow-speed reciprocating engines, all the way down to gyroscopes.  In the cases of most industrial machines, the specification is G 6.3.  The 6.3 denotes the amount of residual vibration at running speed, in mm/sec.

Some high speeds applications, or electric motors operating “stand alone” might specify a G 2.5, or 2.5 mm/sec.

  • G 6.3 = 6.3 mm/sec, or 0.248 in/sec (PK), or 0.175 in/sec (rms).
  • G 2.5 = 2.5 mm/sec, or 0.098 in/sec (PK), or 0.069 in/sec (rms).

These are quite achievable, in most cases, IF unbalance is the only vibration at 1 times the rpm, and the machine is not operating in resonance.  But there are other sources of vibration that manifest themselves at 1 x rpm, such as misalignment, a bent shaft, looseness, and so on.

So, how do you know “when to stop”?

Balancing is both a mathematical, and a logical process.  If you add a correction weight to the right spot, you should expect the vibration to drop, and the phase angle to stay relatively steady.  Even a trim run should be predictable.  But if the vibration amplitude or phase does not perform as expected, look for:

  1. Any looseness in the system, either in the machine, or in your sensor or tachometer. If none is found,
  2. Check for resonance.  Is the machine operating at a natural frequency?  You can test for resonance in multiple ways, from a bump test, to a coast down test, to just changing the input on a VFD drive.
  3. Are there other components that could cause vibration at 1 x rpm?  Balancing can only effect balance quality, not misalignment, or other causes.

Lastly, most manufacturers of machinery specify balance quality for their machine.  You should balance to the manufacturer’s recommendations.