Howard W Penrose, Ph.D.
Vice President, Electrical Reliability Group
T-Solutions, Inc.
Developing Your Motor Repair Specification Part 8
Section 4 involves testing of the electric motor and associated components. One key issue in this section is the lack of pass/fail criteria for any tests other than insulation to ground tests.
In section 4.2.2, “Insulation Resistance Test,” the recommended test limits follow the IEEE 43-2000. Unfortunately, this means that you can receive a rewound motor back with an insulation resistance as low as 5 MegOhms for motors under 600 Volts and 100 MegOhms for all other larger motors and DC Armatures. Is this acceptable to you? I sure hope not! Most modern insulation systems present final insulation to ground values in the GigOhms and TerraOhms. If you are under the GigOhm range, there will most likely be some kind of issue in the insulation system that bears investigating, in newly rewound machines.
Section 4.2.3 involves the Polarization Index. To understand the issues of insulation to ground testing and PI, a copy of my paper on IEEE 43-2000 and Polarization Index testing will be available 12/23/2004 in the morning to download from http://www.motordiagnostics.com/downloads/insulation.pdf.
In sections 4.2.6 (Turn-to-Turn Test) and 4.2.7 (Surge Comparison Testing), absolutely no pass/fail standards are applied.
In all, the primary winding tests focus on insulation to ground which, as I have previously published, represent less than 0.8% of faults in application. In the repair side, based upon newly installed insulation systems, it is virtually impossible to have an insulation to ground failure in a new winding system, whereas scratched wire, missed turns, reversed coils and other coil and connection defects are far more common.
It would be recommended that pass/fail criteria for these tests are included in your repair specification to Your Standard! If the repair shop understands your requirements, then you will be more likely to receive a repaired motor that is far more reliable.
Now, the area that really bothers me, personally, about the test part of the standard: 4.5 No-Load Tests (people keep reminding me to say what I feel):
4.5.1, Speed: It states that for AC motors, no-load running tests should be made at rated voltage and rated frequency. The speed should be measured and compared with nameplate speed.(!?!) Let me put it this way: If you have a 1785 RPM (60 Hz) AC induction motor being tested at no load, what RPM would you expect to see from this statement? 1785? If you do see 1785 at no load on this example AC induction motor, then you have a serious problem! In an unloaded AC induction motor, the RPM should be closer to synchronous (1800, so possibly 1796 in this case) due to the lack of load. If you are running very close to the nameplate speed, then you have a torque problem that may indicate rotor bar problems, etc.
4.5.2, Current: Compare to nameplate? At no load? Again, there are no guidelines presented by the specification.
4.5.5, Bearing Temperature: Tested until stabilized.
In both of the above cases, no limits are specified. These are areas that need to be addressed.
4.5.6 Vibration Tests: Limits are presented, here.
4.6 Performance Tests: Leave the load test question wide-open.
Conclusion from the Testing Section:
While there are a number of tables presented that seem impressive, there are few to no pass/fail test limits or recommendations outlined in the specification. This is a very serious issue, as, if you were to use the specification as it stands, it leaves the quality of the final product (repair) open to the repair shop’s interpretation. It also leaves out a large number of newer, highly accurate, technologies (Electrical Signature Analysis, for instance) and only lightly touches upon other new technologies (MCA, which is downplayed to ‘phase balance,’ for instance).
However, the reason that these questions exist, within the specification, is that the limits will be somewhat dependant upon the specific motors being tested. In this case, it is depending upon the motor repair shops’ understanding and knowledge of electric motors. Unfortunately, in far too many cases, that dependancy is unfounded.
That statement leads to another discussion for another day.
Howard W Penrose, Ph.D.
Vice President, Electrical Reliability Group
T-Solutions, Inc.
Developing Your Motor Repair Specification Part 7
In Section 3, the EASA AR100-2001 discusses Rewinding. This section has the potential for significant impact on the reliability of the machine but has few limits cited.
3.1.2 Core Laminations: Provides no limits on pass/fail of the stator core condition. For instance, it does not cite that the maximum rise for any ‘hot spot’ should be no more than 10C above the average core temp. Also, it does not spell out the requirement for a core loss test to be performed and that the recommended limit is 7 Watts per lb with no increase in W/lb losses when tested before and after removing wire. You may wish to include such limits and specific tests in your specification.
3.2 Rewind Specification: The one sentence that makes up this part of the rewind spec is extremely significant. It also leaves one key item unaccounted: The definition of the term ‘same electrical characteristics as the original.’ A common practice in the electric motor repair industry is to convert ‘basket wound,’ or concentric, windings to lap windings, and, also, to change the windings slightly to allow for reducing wire size or the number of turns, as some newer motors have very tight core slots. If performed properly, the motor can be rewound to similar characteristics of the original (in some cases, better). If not performed properly, you will have reduced reliability and efficiency (increased operating costs). It has been cited that 81% of motor repair shops change winding design through the rewind process with 73% doing so for convenience and ease of rewinding and only 4% for improved reliability. It is highly recommended that you expand your requirements in this part of the spec.
3.3 Stripping of Windings: For an area that has been very controversial since the 1970’s, there are few solid recommendations in the specification that are covered by other documents. There are several basic winding removal methods with the most used being burnout ovens. The temperature limitation on these should be 650F in a temperature controlled oven, as higher temperatures both have an impact on older core steels and have an impact on soft foot (stress relief of housing steel). Mechanical stripping methods should be limited to low-temperature (400F or below) hydraulic systems, as pneumatic systems have a history of damaging core materials. The preferred winding removal process and temperature limitations should be included in the specification.
3.14 Impregnation of Windings: This section is geared to the most common (least expensive for the repair shop) varnishing process: Dip and bake. It leaves out trickle impregnation (higher quality insulation system, but expensive equipment), pour-through systems and vacuum pressure impregnation.
In general, this section is effective, but requires some limits and requirements that are not included in the specification as it stands. It is recommended that you include the limitations that you wish as the rewind portion of the EASA AR100-2001 allows for low quality repair.
Tomorrow we will cover Section 4, Testing.
The next section of the Motor Repair Specification Lecture Blog will continue next week on Monday, December 20, 2004.
We will be concentrating on rewind, testing, impacts of rewind on efficiency and reliability and the repair versus replace decision process.
In the following week, we will tie this part of the lecture series together and will make it available as an Adobe Acrobat file in January, 2005.
Howard W Penrose, Ph.D.
Vice President, Electrical Reliability Programs
T-Solutions, Inc.
howard@motordiagnostics.com
Developing Your Motor Repair Specification 6
In working with the EASA AR100-2001, the second section provides an overview of mechanical testing and evaluation.
The standard is pretty comprehensive including tolerance charts, etc. However, it leaves out one key item: Best practices for repairing mechanical components.
Bearing Housing Fits:
There are several ways to repair bearing housings which include:
1. Re-Manufacturing – This involves completely refabricating or replacing the complete endbell. This is normally not the most cost effective method for mechanical repair.
2. Sleeving – Involves removing metal, pressing in a metal sleeve then machining concentric and to the correct tolerance. This is typically the best method for repairing the housing, with the best reliability.
3. Tolerance Ring – Involves removing metal from the bearing housing and placing a spacer into the fit. Requires less accuracy in machining and costs less than sleeving. However, it involves very low reliability in the repair.
4. Metalizing – The housing is machined and threaded. A special metal-adhesive is sprayed into the housing and the housing is then machined to the correct tolerance. This is better than a tolerance ring. However, contaminants on the prepared surface will cause the material to lift and can damage the bearing(s). This is best not used in high speed applications or belted applications.
5. Welding – Is usually not performed on endshields due to the potential for stress fractures and other damage to the housing, itself.
6. Fillers – There are commercial fillers and adhesives that are sometimes used to fill worn bearing housings. These have a tendency to put undue stress on the bearing when the shaft expands, due to heat, axially. The result is a very-reduced bearing life and the possibility of an eccentric rotor or rotor bow.
In most instances, I usually cite sleeving as the preferred method for repairing endshields. If the repair center prefers to use any other method, then they have to obtain permission prior to initiating repairs. Also, all micrometers and other measuring equipment requires quarterly calibration.
Shaft Fits:
Shaft fits can be returned in one of several ways:
1. Re-manufacturing – The best way involves the machining of a new shaft. This can be the most expensive method, but involves the greatest reliability.
2. Sleeving – Depending on the application and the type of shaft damage, a sleeve can be pressed on and machined.
3. Welding – involves welding on metal and machining to size.
4. Metalizing – same as the bearing housing.
The primary concern with repairing the shaft is the potential for stress fractures and metal fatigue. Either of these can generate severe problems with the shaft during operation, such as the shaft breaking or bowing.
The best methods are re-manufacturing or sleeving.
Howard W Penrose, Ph.D.
howard@motordiagnostics.com
Vice President, Electrical Reliability Group
T-Solutions, Inc.
Pardon the short break in the motor repair BLOG. Normally I have the opportunity to sit for an hour or so while traveling for most shows and conferences. Not so at the IMC Conference in Bonita Springs, FL.
With over 850 attendees and a really neat concept of Learning Labs for several vendors (including ALL-TEST Pro), things were very busy. Kudos to Terry and Kelly O’Hanlon and their staff on an unbelievable show and conference.
In addition, we initiated the working group for the start of the Electrical Motor Diagnostics group. This group will be involved in the development of standards and certification for EMD.
So, I will continue the motor repair BLOG lecture starting tomorrow.
Sincerely,
Howard W Penrose, Ph.D.