BRENNEN

Industries / Industrial equipment and services / Industrial services / Vibration analysis

Balancing and alignment

Balancing adds or removes weight on a rotor so its mass is spread evenly around its axis, and alignment puts two coupled shafts on one line.

4

Milestones achieved, from 1919 to 2001.

1

To watch for: announced, not yet confirmed outside.

2

Open deliverables. Nobody has shipped them.

The road

AchievedWatch forOpen

Achieved

  1. 1919

    Running past the critical speed

    A critical speed is a running speed that matches a natural frequency of the shaft, where it bends and shakes most. In 1869 W. J. M. Rankine concluded that beyond a certain "whirling" speed a shaft bends without limit; Gustaf de Laval then ran a steam turbine at about seven times its critical speed (1883 by one account, the late 1880s by another), Föppl showed in 1895 that the heavy side of a disc swings to the inside above the critical speed, and in 1919 Jeffcott's paper in the Philosophical Magazine showed stable running above it, with the heavy spot's angle shifting steadily as the shaft passes through. That angle shift is why a balancer measures phase, the timing of the vibration against a mark on the shaft: a 2016 balancing tutorial calls assuming no shift "a poor assumption" in most cases.

    Lessonbecause Rankine was far better known, his wrong prediction was widely accepted and, by one account, discouraged high-speed rotors for almost 50 years.

    archive.nptel.ac.inarchive.nptel.ac.insoftinway.comdoi.orgrotorlab.tamu.edu

    Credit

    Jeffcott; Föppl; Gustaf de Laval

  2. 1972

    Balancing weights by least squares

    In the influence coefficient method a balancer measures the vibration's size and phase, adds a trial weight, measures again, and uses the change to calculate the correction. Goodman published a least-squares method for computing balance corrections in 1964, and J. W. Lund and J. Tonnesen published multi-plane balancing of a flexible rotor in 1972. That July, Mechanical Technology Inc. reported to NASA that, with programs written by Lund, a 126-pound, 41-inch rotor was brought safely through three critical speeds after two or three balancing runs in most cases; a 1975 report extended the method to four bending critical speeds at up to 18,000 rpm. A 2016 rotor balancing tutorial by Ray Kelm, Walter Kelm and Dustin Pavelek of Kelm Engineering teaches the same method for balancing machines where they stand.

    ntrs.nasa.govntrs.nasa.govdoi.orgdoi.orgrotorlab.tamu.edu

    Credit

    Goodman; J. W. Lund and J. Tonnesen; J. M. Tessarzik and R. H. Badgley of Mechanical Technology Inc.; NASA Lewis Research Center

  3. 1976

    Grinding wheels balanced while turning

    A grinding wheel wears unevenly, so its balance changes during work and spoils the surface it grinds. Gebr. Hofmann of Pfungstadt, Germany, applied in November 1973 for a fluid balancing device with several chambers on the rotor, granted in the US in April 1976; a later Hofmann patent describes measuring the unbalance continuously and injecting fluid into the chambers while the wheel turns, and warns that a wheel speeding up as it wears passes through resonance, where the phase shifts by up to 180 degrees. Marposs now sells Marposs/Dittel balancers that "compensate any imbalance in real time during the cycle" on grinding machines.

    patents.google.compatents.google.commarposs.com

    Credit

    Gebr. Hofmann GmbH & Co. KG

  4. 2001

    Misalignment costs bearings, not power

    A 2001 University of Tennessee thesis notes that industry's spending on precision alignment rests on two assumptions: misalignment lowers motor efficiency, and it makes machines fail sooner. Stephen Jesse, advised by J. W. Hines at the University of Tennessee, tested 50 hp and 60 hp induction motors with four kinds of flexible coupling at about 15 alignment conditions each, at the university and at the Oak Ridge Center for Electrical Machinery System Testing. He found no significant link between misalignment and efficiency within the coupling makers' limits, but in some cases a loss of up to 50% of expected bearing life from an offset of 5 thousandths of an inch (0.13 mm); a 2012 US Department of Energy tip sheet also says misalignment "has no measurable effect on motor efficiency".

    Lessonthe case for careful alignment rests on bearing, seal and coupling life, not on the electricity bill.

    trace.tennessee.eduenergy.gov

    Credit

    Stephen Jesse and J. W. Hines, University of Tennessee

Watch for

  1. Balancing with no cables

    Acoem says its Wireless Balancer is "the world's first and only 2-plane wireless balancer", taking vibration and phase readings in step without cables near the rotor, with phase "precision up to 0.5° accuracy at 3000 RPM". It won Plant Engineering's Gold 2026 Product of the Year, a readers' vote, not a test.

    Achieved whenan outside party balances the same rotors with it and with a cabled balancer and publishes the phase readings, the number of runs and the vibration left.

    acoem.usacoem.us

    Claimed by

    Acoem, June 2026

Open

  1. 01

    Robot alignment, tested openly

    A rig in which small motors turn the jackbolts, the pushing screws that slide a motor across its baseplate, under command of live readings from a laser alignment kit, until the motor's shaft lines up with the pump shaft, and a published tally of how well it does against a technician. The motor stays unpowered and locked out, and the shafts are turned by hand for each reading.

    The brief
    Credit · unclaimed
  2. 02

    Balance fans from shared numbers

    A public table of influence coefficients, the measured response of vibration to a known weight, from many units of one small fan model, and a test of whether the averaged coefficient balances a new unit in one correction, without the trial-weight run that field balancing normally needs.

    The brief
    Credit · unclaimed

Open 01 · the brief

Robot alignment, tested openly

Credit · unclaimed

The first to publish is credited on this line, and the deliverable moves from Open to Achieved.

What to build

A rig in which small motors turn the jackbolts, the pushing screws that slide a motor across its baseplate, under command of live readings from a laser alignment kit, until the motor's shaft lines up with the pump shaft, and a published tally of how well it does against a technician. The motor stays unpowered and locked out, and the shafts are turned by hand for each reading.

Why it is still open

Machines are still moved by hand almost everywhere. A US patent granted in 2000 for a portable jackbolt system lists the usual tools as portable hydraulic jacks, wedges, pry bars, cranes and sledge hammers, and calls their use "very time consuming, dangerous to personnel and machinery". Automated moves have been patented and offered by one small maker: Danny Harris's US patents of 2012 and 2019 describe actuators driven by the computed correction, and Harris Technology LLC's RobotAlign, listed in PECM magazine in April 2025, includes automated jacking software, with the jacking hardware sold separately. A web search in November 2026 found no published results for it or any other automated system.

patents.google.compatents.google.compecm.co.uk

Built with what exists

An alignment training rig, such as a SpectraQuest alignment and balancing trainer; a commercial laser shaft alignment kit; stepper motors and drivers on the horizontal jackbolts; a Raspberry Pi and Python. Vertical moves need shims under the feet, so a first version handles side-to-side moves only.

  1. First step

    Fit stepper motors to the two horizontal jackbolts at one end of a training rig and close the loop on side-to-side offset alone.

  2. Done looks like

    Twenty filmed attempts from random side-to-side starting offsets of about 0.5 mm (20 thousandths of an inch), each with the final offset read by the laser kit and scored against the limits of 2.0 thousandths "excellent" and 3.0 "acceptable" that DOE's tip sheet gives for 1,800 rpm short flexible couplings, and the time per attempt compared with a technician's.

    energy.gov

Open 02 · the brief

Balance fans from shared numbers

Credit · unclaimed

The first to publish is credited on this line, and the deliverable moves from Open to Achieved.

What to build

A public table of influence coefficients, the measured response of vibration to a known weight, from many units of one small fan model, and a test of whether the averaged coefficient balances a new unit in one correction, without the trial-weight run that field balancing normally needs.

Why it is still open

Shared coefficients are used inside makers' own programs but have not been tested in public. A Boeing patent filed in 2017 on engine balancing says a single set of influence coefficients "cannot be relied on" across engines of one type, because of tolerances, wear and measurement error, while noting that averaging across engines is known practice. Field balancing as taught in a 2016 tutorial still starts with a trial weight big enough to shift the vibration by at least 10%. A web search in November 2026 found no published test of shared coefficients across identical industrial fans.

patents.google.comrotorlab.tamu.edu

Built with what exists

Ten or more identical small ventilation fans inside mesh guards; an accelerometer and an optical once-per-turn tachometer; a USB data acquisition module or a Raspberry Pi with an analog-to-digital converter; Python; weights fitted only with the fan stopped and unplugged.

  1. First step

    Measure the influence coefficient of three identical fans at one speed and publish how far apart they are.

  2. Done looks like

    For at least ten fans, the share brought below a stated vibration limit by one correction from the shared coefficient, against the number of runs the trial-weight method needs on the same fans, with the spread of the coefficients in size and angle published.