We published a motor tip in the July 17 issue that created a great deal of feedback. All of the feedback offered a 180 degree response to the original Motor Tip and attempt to correct what they state are mis-statements by the original tip author.

We accept responsibility for publishing the original tip from a new source who happened to be a supplier to this industry and we apologize for not vetting the information more carefully.

We appreciate all of those who emailed and we are sending Maintenance Tips hats and Stainless Steel. Diamond Plate Coffee Mugs as a small thank you. We hope the replies below go some distance to provide new information.

Respectful discussion is always encouraged throughout the Reliabilityweb.com network and we hope you can learn from our missteps. Additional comments are welcome.

Thank you for your patience and understanding.

Terry O

Original Motor Tip

A reader recently asked the question how could preventive maintenance be promoted to management on the basis of energy savings. One fact related to electric motor operations is that rewound motors operate less efficiently than new motors.

By definition, a “burned out” motor has suffered from a mechanical failure. The winding insulation failed to contain the electric current - regardless of the cause (thermal breakdown, chemicals, vibration, voltage spikes, or moisture.) Motor rewind shops will agree that seldom - if ever - is it feasible to rewind a motor more than once. This is due to the cumulative damage to the windings during failure.

Rewind shops will also admit that not all motors can be rewound, for the same reason. What they may be reluctant to admit is that rewound motors will operate less efficiently than new motors. Depending on who you talk to, estimates range from 50% to 90% efficiency for rewound motors versus new. Hence, it can be demonstrated that repaired motors will increase energy consumption.

This is avoidable by:
#1: Only replacing failed motors with new motors.
Or
#2: Monitoring motor condition and taking worn motors offline BEFORE they fail.

Worn motors can be “reconditioned”, that is reinsulated before any damage to the windings occurs. A reconditioned motor returns to like new efficiency and capacity. In addition, the cost of reconditioning is usually less than half that of rewinding, and nearly 100% of worn motors can be reconditioned.

Of course, this is contingent on “catching” motors before they fail. What is required is lucky timing for a manual test or the use of automatic testing systems - like MEG-Alert. So preventive maintenance CAN lead to lower operating costs, AND lower maintenance expenses, overall.

Proactive maintenance will ALWAYS cost less than reactive, emergency repairs!

Reader tip by Randy Reek MEG-Alert Hot Springs Village Arkansas

Motor Tip Feedback

A recent article from the Reliabilityweb Maintenance Tips I received on July 17th, has gotten my blood boiling!

The author of the article made many statements in regards to motor maintenance, repair, rewinding and replacement that must be challenged, due to their blatant inaccuracy and falsehood!

In the article he states, quote (One fact related to electric motor operations is that rewound motors operate less efficiently than new motors.)Unquote. That statement is absolutely false! I direct your attention to the following publication from the Electrical Apparatus Service Association, Inc., EASA,
http://www.easa.com/indus/index.html
titled, “The Effect of Repair/Rewinding on Motor Efficiency”. I quote, (The 22 new motors studied were divided into four groups to accommodate the different test variables. The test results summarized below show no significant change in the efficiency of motors rewound using good practice repair procedures (within the range of accuracy of the IEEE 112B test method), and in several cases efficiency actually increased.) Unquote.

Later in the article the author states, quote (What they may be reluctant to admit is that rewound motors will operate less efficiently than new motors. Depending on who you talk to, estimates range from 50% to 90% efficiency for rewound motors versus new. Hence, it can be demonstrated that repaired motors will increase energy consumption.) Unquote. Once again, he is in error. I quote (The average efficiency change for each group also falls within the range of accuracy for the test method, (plus or minus 0.2%), showing that motors repaired/rewound following good practices maintained their original efficiency, and that in several instances efficiency actually improved.) unquote.
http://www.easa.com/indus/index.html

The author might want to do his homework before he submits another sales pitch disguised as an article. I would suggest the following, “A Guide to AC Motor Repair and Replacement”. He can find it here, http://www.easa.com/indus/index.html

Tip response provided by Stephen Powell
Service Manager
Turnupseed Electric Service, Inc.
Tulare CA

More Motor Tip Feedback

Original Tip: “Motor rewind shops will agree that seldom - if ever - is it feasible to rewind a motor more than once. This is due to the cumulative damage to the windings during failure.”

Feedback: The windings are replaced so no cumulative damage

Original Motor Tip: “One fact related to electric motor operations is that rewound motors operate less efficiently than new motors….”

Feedback: Efficiency should be almost the same as windings are replaced with the same as before - you may have a little more heat loss through the iron because the windings on smaller motors (1000 hp and less)are burnt out in ovens @ a high temperature.

Feedback provided by Lucy Malone
Catalyst Paper
Campbell River,B.C.

Even more Motor Tip Feedback

Original Tip: One fact related to electric motor operations is that rewound motors operate less efficiently than new motors.

Feedback: Only sometimes, if not done correctly, blanket statements are seldom all encompassing.

Original Tip: Worn motors can be "reconditioned", that is reinsulated before any damage to the windings occurs

Feedback: Motors can be baked out, cleaned, revarnished, new lead wire installed.

Reinsulated implies much more.

Neil Gillespie
Electrical Engineer

Thanks to Stephen Powell for jumping on this right away. Thank you for your links to EASA as a plus.

When the breakdown of the insulation between the laminated plates of the stator (due to excessive heat-too many restarts in a short period) is left undetected, a rebuilt motor with new windings is not going to run properly or efficiently. If the stator or rotor has been machined on its circumference to correct an out-of-round or small contact damage, there will be less efficiency because the air gap is now larger even if it has new windings. Technically, this correct and true.

In practice, all of the motor repair shops that I have dealt with are very much aware of these conditions. All of them use VPI dip treatment which insures that the windings have no movement at all in the slots. All of them balance the rotor after rework. All of them replace the bearings on the rotor or armature, insuring a proper fit also in the end bells or brackets. In most cases, not just as a rule, the efficiency does not decrease less than 1% and in alot of cases actually improves. That alone is a testimony to the knowledge, skill, and best practices of our motor repair shops. All of them have called first with a recommendation after their findings to inform the customer and get the final go ahead. In what is mutually agreed upon, the customer bears some responsibility for deviations from the recommended repairs or partial fixes if the motor continues to have problems or has lost efficiency.

To minimize disappointments and maintain credibility, it is a good practice to to have a written report and/or the recommendations signed off by the customereither by fax or email messaging before proceeding. Word of mouth is quicker but ink is the sticker. (enough corn).

Louis G. Pagliaro
Full Spectrum Diagnostics, PLLC
Cornelius, NC

Hello Terry, it has been a while since we last talked. I saw the retraction and the fire fight going on and thought you might need a hand. As you know, I don't have a dog in this particular fight. I don't sell anything to anyone who has a stake in this disagreement. I do want to make sure that people out there are getting an accurate representation of the problem so they make good decisions, and that Reliability professionals provide services to manufacturing that are of the highest standard. We are pros, and pros base decisions on hard data, not on out dated rules of thumb.

Here is my input on the motor issue. Most of the responses I have seen contain some truth and some speculation. Since not all motors have the same degree of damage coming in and not all shops have the same standards for repair and testing it is very easy to get a wide spectrum of results, all of which are "true facts" to a given individual based on a particular situation. You cannot dispute someone’s experience, but the conclusions they draw can be subject to bias on account of the particular situation that prompted them.

Let's first examine what can cause loss of efficiency of a repaired motor and examine how the repair process can affect those.

1) Hysteresis losses in the laminations (often called core loss). This is perhaps the most significant source of potential losses. What can cause increased hysteresis losses?

Generally (but not exclusively) core loss is increased when the interlamination insulation (the "varnish" on the individual lamination surfaces) deteriorates. This deficiency cannot be repaired except by building a new stator. What can cause it to deteriorate during the repair process or during the motor life?

Excessive temperatures.

Motors that have experienced fire or other extreme overtemp damage before repair or that have had the winding varnish "burned out" using uncontrolled methods, or that have simply had far too many "burn outs" in their life will exhibit high core losses due to degradation of this insulation. Most reputable shops have closely controlled "burn out” furnaces so this issue is not as common as it used to be. This expected degradation was the reason behind the old rule of thumb to rewind a motor no more than 3 times. Years back most shops did not have (and many still do not have) the ability to measure core loss. Or, they had it but the client didn't ask for it and; a rebuild is a sale, but a scrapped motor is a lost sales opportunity - right? That is why it is a best practice to test for core losses before committing to rebuild and doing a final check before re-assembly if the motor was rewound.

Another source of core loss is when there has been an rotor - stator rub, usually the result of a bearing failure gone terminal. There is smearing of metal at the end of the laminations that effectively negates the inter-lamination insulation with resulting core loss. If the damaged area is bored out or ground out (don't laugh, some shops will do this) this creates an irregularity in the field, and depending on the method used and the care taken, some metal will still bridge from lamination to lamination, or the inter lamination insulation at the rub point will still be damaged by local heating associated with the rub in the first place. The net result is still core loss which means efficiency loss and higher operating temperatures.

2) Rotor-Stator gap and eccentricity. Excessive clearance or eccentricity both result in decreased efficiency. Removing metal to "repair" smearing after a rub if done uniformly by grinding or boring (the infamous "clean up cut") increases clearance. Bearing failures often require boring and sleeving to get a proper bearing fit. Poor registration before machining, or distortion of the end bell by improper clamping to bore it can result in eccentric and excessive clearance. Loss of efficiency along with a host of other problems results.

Likewise a bowed rotor, either from a bent shaft, heating from a rub, or unbalanced heating caused by a broken rotor bar can cause the same eccentricity problems described above.

Often a motor will check out OK under no load, but when subjected to load will heat up, a cracked rotor bar will open up causing uneven heating, vibration, efficiency loss, and sometimes a rotor-stator rub. T compound it these problems can't be reproduced in the shop on an unloaded motor. Your rebuild specs do contain the requirement to test motors under load on a dyno, don't they?

Loose or cracked bars in and of themselves are as I have pointed out a source of loss. And as I mentioned, some of them don't make themselves known until the motor has been under load for a while. Whether the rotor bows or not the motor needs to be at operating temperature before a good assessment of the condition can be made.

3) Circulating currents. Some motor designs require an insulated bearing on one end to reduce circulating currents - particularly in high voltage motors. These in turn cause heating and heating is efficiency loss.

Improper replacement of the bearing or the wrong bearing type can result in loss of insulating effect. Dealing with insulated bearings in the correct way and testing for it is part of your repair specifications isn't it?

4) Winding Configuration. If the windings that are replaced do not exactly match the original coil configuration (diameter, # of turns, geometry, packing density) then efficiency may suffer. It is easy for this to happen even if no changes in configuration are desired if the shop does not have the right capability. Sometimes it is the unintended consequences of requesting upgraded slot insulation, or upgraded coil wire (such as upgrading from standard to inverter duty wiring). Again finding the true condition requires standards and testing.

I am not a motor expert and I urge you to seek out people who are when developing your specifications and test standards. I am sure that there are other issues that I have missed. These lessons all reflect real pain I or my clients have felt over the years.

The simple truth in this argument is that what you get back from a rebuild shop depends heavily upon what you send them, what you ask for (standards), and whether or not they have the inherent process capability to meet the standards.

I hope that this puts a little water on the motor controversy fire instead of gasoline.

- Sam
Samuel S McNair, Jr PE CMRP
Life Cycle Engineering
Phone: (843) 744-7110
Cellular: (843) 670-3784
smcnair@lce.com
www.lce.com

I read the motor repair tip today and also the responses on the forums that were copied to me.

I actually had a number of issues with the tip:


1. A motor can be repaired to original efficiency as shown in the studies mentioned in the two-part article I did in Uptime Magazine. It requires the repair shop to follow strict quality control processes. When such processes were not followed, as identified in third party studies, the decrease in efficiency ranged from 0.5 to 3% per rewind. As such, the US DOE published an estimation of 0.5% reduction in efficiency per rewind in Standard Efficient Motors – not in premium or energy efficient motors, as was found in both the CEA and EASA studies.

2. I do have an issue with the EASA study, that was performed without third party oversight, after the results from the CEA study did not meet full expectations.

3. Insulation to ground testing will not detect the 80% of winding faults that are initiated as winding shorts. Insulation to ground does not see or detect these faults. You cannot ‘re-insulate’ a turn fault.

Hope this is helpful.

Howard W Penrose, Ph.D., CMRP
President, SUCCESS by DESIGN Reliability Services
Vive la révolution de l'entretien!

Author: “Physical Asset Management for the Executive (Caution: Do Not Read This On An Airplane)” and “Electrical Motor Diagnostics, 2nd Edition”

Cell: 860 575-3087
Fax: 860 577-8537
Web: http://www.motordoc.com

quote:

Let's first examine what can cause loss of efficiency of a repaired motor and examine how the repair process can affect those.

1) Hysteresis losses in the laminations (often called core loss). This is perhaps the most significant source of potential losses. What can cause increased hysteresis losses?

Generally (but not exclusively) core loss is increased when the interlamination insulation (the "varnish" on the individual lamination surfaces) deteriorates. This deficiency cannot be repaired except by building a new stator. What can cause it to deteriorate during the repair process or during the motor life?

www.lce.com

There are two sources of losses in the iron of the electric machine. One is the hysteresis loss and another is the loss due to the eddy currents in the iron. The hysteresis loss is there even if there is no insulation between the individual laminations. The “elementary magnets” in the iron have to follow changes of the AC magnetization. The alignment of the “elementary magnets” with the constantly changing AC magnetic field requires energy, it means it creates losses. The amount of energy increases roughly proportionally with the frequency, and is proportional to the square area of the hysteresis curve of the particular steel material.
The other source of losses are the eddy currents. Since the magnetic material is conductive, there is always a conductive path around the magnetic flux. As soon as you have a conductive path around the AC magnetic flux the current starts to circulate around the magnetic flux and creates losses. Luckily there is solution. The thinner the sheets, the lower the losses. So in order to keep the losses low, you have to keep the insulation between lamination sheets intact. Those losses increase roughly with the square of frequency.
Hence the hysteresis losses are not often called “core loss”. The “core loss” is the result of the “hysteresis loss” PLUS the “eddy current loss “.
jank

quote:

By definition, a “burned out” motor has suffered from a mechanical failure. The winding insulation failed to contain the electric current - regardless of the cause (thermal breakdown, chemicals, vibration, voltage spikes, or moisture.) Motor rewind shops will agree that seldom - if ever - is it feasible to rewind a motor more than once. This is due to the cumulative damage to the windings during failure.

Rewind shops will also admit that not all motors can be rewound, for the same reason. What they may be reluctant to admit is that rewound motors will operate less efficiently than new motors. Depending on who you talk to, estimates range from 50% to 90% efficiency for rewound motors versus new. Hence, it can be demonstrated that repaired motors will increase energy consumption.

I would really like to know who was talked to about only rewinding one time and the reduction in efficiency by '50% to 90%'. I have yet to find a motor repair shop that will make a statement that a motor can only be rewound once and I have never found a study that shows a rewound motor operates at half of its original efficiency. It would be a radiator!

There were numerous studies performed in Canada during the 1990's which the US Department of Energy adopted. I discuss each of these studies in the two-part article in Uptime Magazine earlier this year. The most significant third-party study was performed by the Canadian Electrical Association in 1994. While initially supported by EASA, the study did not present the results they wanted, and the 'private' (non-reviewed) study performed in the UK was initiated. The primary difference between these studies and the core loss studies performed by EASA and General Electric in the past had to do with the core material.

The general losses in a machine are:

• Core Losses - which account for 15-25% of the overall losses: a combination of eddy-current losses which are the result of stray currents from the fields within the motor. In effect, induction heating of the stator; and, Hysterisis losses, which are the losses relating to the re-alignment of the magnetic fields in the core steel. The first is determined by the thickness of the steel and the condition of the inter-laminar insulation (thicknesses tend to fall between 0.019 to 0.039 inches for energy and premium efficient motors). This can be effected by burnout oven temperatures in older standard efficient core materials (as low as 400F) and excessively high temperatures in energy and premium efficient cores (around or above 1,000F, depending on the interlaminar insulation). Hysterisis losses can only occur with a change to the core material itself.
  
  
• Friction and Windage Losses - account for 5 to 15% of the overall losses: changes to the type of bearings (ie: shielded to sealed bearings can reduce efficiency an average of 3% or more - ie: 94% to 91% efficient), the surface areas of the rotor, and the fan (changes to the number of blades, etc.).
  
  
• Stator losses - 25 to 40% of the overall losses: These are the I^2R (current-squared times resistance or 'eye-squared r') losses that are directly related to the DC resistance of the windings and the amount of current. If the repair shop reduces the cross-section of the wire, it will increase these losses.
  
  
• Rotor losses - 15 to 25% of the overall losses: also the I^2R losses of the rotor. In a standard induction motor you should not expect to see any changes.
  
  
• Stray load losses (IEEE Std 112 - calculated)accounts for 10 to 20% of the overall losses: all other losses not accounted for including leakage.
  
  

In a 95% efficient motor, these total losses would amount to 5% of the energy introduced to the motor. The amount of effort to reduce the efficiency to less than 50% would require the repair shop to significantly reduce the wire size, totally destroy the inter-laminar insulation of the stator core, modify the friction and windage, install sealed bearings, and alter the rotor windings. On startup, the inrush would be extreme and the operating current would be too far above the nameplate at no load.

Fascinating.

All of this to sell an in-line megger, which would only detect less than 15% of winding faults in any case - according to the 1983 EPRI study and post 2000 EASA study on motor repair.

On the other hand, users of the MegAlert that I have talked to seem to like it well enough.

Howard

Welcome back to the forum Howard.

All overhauls should be AGAN (As Good AS New). If not, pls say so. If any upgrades or changes made, pls tell so.

Thanks Dave. I have just not had time - things have gotten very busy.

Josh - we would hope so. What we have found is that a good motor repair specification, such as the ANSI/EASA AR-100 (free from EASA.com) and the new IEEE 1068, once completed and released, are a good idea. The new 1068 provides a grading criteria for evaluating motor repair shops. If you keep the facilities in check and commission motors before sending them to a repair shop (>15% of motors are 'no problem found') and check when they are returned, with an agreement that failed checks will be returned, you will find you maintain efficiency and have good, reliable results.

When you do not monitor the repairs, or evaluate and work closely with your repair vendor, you will most likely end up with the average 0.5 points per rewind reduction, or worse, and potential warranty issues.

Sincerely,
Howard

What is the average 0.5 points per rewind reduction? Is it efficiency reduction?

Josh

Yes. Basically, if a motor was 94.2% efficient, it would be 93.7% after the first one, 93.2 after the second and so on. The work done at Ontario and BC Hydro actually stated an average of 1% per rewind. The US DOE and EASA agreed to 0.5% during the development of a number of US DOE publications.

Howard

Well, in that case, is there a limit to the number of rewinds to ensure AGAN and/or go for replacement unit? Of course, users have to economize on the number of rewinds then.

You can rewind as many times as you wish if you have good history of the motor. Since it is only the iron that can deteriorate by too many rewinds (affecting the efficiency), it is useful to have a graph of losses as shown in the attached file. If the motor comes for rewind sometimes in the future we can retest it. If the watts go up, some decisions have to be made.
The other important thing is putting the same amount of copper into the slots and correct configuration of the winding. It is almost always possible to add little more copper. However it is not the rule.
jank

keyera200hpWO21406Watts.pdf (33 Kb, 17 downloads)

Josh

It depends on what your goals are.

There are three ways that a repair shop can effect the overall efficiency of the motor, in common cases. One is to reduce the wire size (cross section), second is to convert the winding from say concentric to lap without doing a proper conversion, and the third is core loss. The first two will have an immediate, significant effect on I2R, and other, losses.

We have been preventing this issue in a number of ways - one is to provide a standard or specification. At some of our clients they have produced their own spec, while at others we had first adopted the ANSI/EASA AR-100, but are now already adopting the draft IEEE 1068 (not the 1976 version), including the scorecard for selecting repair facilities. Even then, incoming commissioning tests are still finding modifications to repairs.

It is important to remember, as well, that reduced efficiency also means reduced reliability.

1. Have a specification in place. If you don't have your own, adopt one of the above. EASA includes the AR100 at no cost along with a motor repair shop checklist on EASA.com.
2. Commission repairs. DO NOT take the motor repair shop's word for the condition of the motor. This both ensures that they are more vigilant and adds a layer of protection.
3. If you are primarily concerned about efficiency, track it using a program such as MotorMaster Plus or the efficiency tests built into the online Baker, PdMA, Areva, or ATPro.


If you are just looking for a rule of thumb, the industry tends to use 'three' as the number of rewinds before replacement. That is even the recommendation in the Canadian Electrical Association report and several US DOE documents. However, in both cases, they are making the recommendations based upon getting people to go from standard efficient to energy efficient.

The challenge in the energy efficient and premium efficient motors is that there is little room for the copper and the slot paper tends to be a little thicker. EE and PE rewinds should be primarily evaluated to ensure that the wire size or number of turns is not reduced. We have caught numerous instances of each recently.

Howard

Thanks for the replies guys. Very enlightening!

My goal from the reliability point of view is that any overhauls including for electric mtors should be AGAN. This is one of basic assumptions before making reliability analysis.

Now, I am a bit puzzled by the statement that reduced efficiency means reduced reliability. Pls explain further. I understand that we should change from standard efficient to premium efficient for energy saving. Are premium efficent motors more reliable than standard efficient motors because the former is more energy efficient?

Josh

Actually, I was referring to the loss of efficiency. Any of the steps that would reduce the efficiency of the motor would reduce its reliability. Or, should I state - long term availability, in that you would have either hot spots in the core (core damage), and/or increased I2R losses, and/or loading issues.

However, yes. The energy and premium efficiency motors are built using better materials in order to obtain the higher efficiency rating. It is less the result of the higher efficiency, but the materials used, that make them more reliable.

Howard