Comparing a Repair Versus New for an Electric Motor. Which is Better?

Comparing a repair versus new for an electric motor. Which is better?

The picture at the top of this post is a repaired ABB electric motor just about to leave our UK workshop. It’s not a big motor and there’ll be many people saying things like ‘it’s not worth repairing’,  ‘you’ll be better off buying new’, ‘A repair isn’t as good as a new motor’ and “you’ll lose efficiency if you repair it’.

Well I’d like to share this post and tell you why I think this repair is as-good-as, if not better than a new motor. I’d like to challenge the thought concept of ‘new is better’, by explaining the processes and engineering standards that we and other quality progressive repair workshops follow. That’s right, this is not a plug for my company but the whole of the repair industry who follow best practice guidelines.

Firstly there’s the argument and I find it mainly comes from the manufacturer, that a repair or rewind will have lost some of the original efficiency. Well that’s easily dealt with by following EASA/AEMT good practice guide; and to my knowledge and experience I’ve never heard a manufacturer challenge the report, but I’ll come onto the efficiency of the repair/rewind later.

Take a look at the repaired motor again, looks good doesn’t it? well I’d now like to let you know how good I think it is, inside and out. I’ll be discussing the major processes and stages of a repair and then compare those to a new motor bought off the shelf.

The motor is a 37Kw, 2 pole, 200M frame. When we received the motor we found the windings were blown, there was a blow between phases on the connection end. So taking a look at the winding of the motor first, this is where manufacturers will tell you that a rewind is detrimental to the motor’s efficiency. Well that’s simply not true providing critical steps are followed. We already knew the windings had blown outside of the stator slots, even so, we still carry out a core loss test on the stator with a Lexseco core loss tester. This gives us a reading of the watts loss per Kg of iron, and an indication of the condition of the stator. The watts loss per Kg pre-burnout was 4.270. The burn off process is carried out in a temperature controlled pyrolysis oven set to a maximum of 370 degrees Celsius. A post core loss test is then carried out and we saw a reading of 4.263 watts per kg. The slight difference is due to the tester repeatability but most importantly there has been no alteration of the stator core from the burnout process, so, the process hasn’t damaged the core. Since the early 1980’s we have been using a full class ‘H’ insulation system with full phase barriers as standard. After copy winding, we also VPI varnish the new windings as standard. Class ‘H’ is an insulation system higher than class ‘F’ which is usually a manufacturers standard specification.  I’ve seen manufacturers taking the winding wire to the terminals too and certain manufacturers of a budget range, make connections within a coil, supposedly when bobbin had run out. Things a decent repair company would not do. So I’m saying our rewind specification is BETTER THAN NEW.

Now let’s look at a few mechanical aspects of the repair. This machine has a roller bearing at the drive-end and a ball bearing at the non-drive-end. Correct measuring of the housings and journals are critical to the reliability of the machine, how do you measure these fits? Well first of all you need to know the fit and tolerance for the bearing, this means having data tables and a method of accurately measuring. As anyone will know, the fits are down to microns and you need a measuring system accurate enough to measure this. We’ve experimented with numerous methods over the years and the best and most repeatable method we’ve found which gives a printout of results is a coordinate measuring machine (CMM).

CMM Measurement of a bearing housing

Using this method, we found the drive end bearing housing for a NU312C3 roller bearing with a J7 fit (tolerance of 129.986mm – 130.026mm) measured oversize at 130.049mm and oval. It was oversize by 23 microns. Left uncorrected this would have caused premature failure of the motor by allowing the outer race to spin in the housing. The housing was bored and sleeved then re-measured. Incidentally the rest of the housings, journals, spigots and shaft extension, keyway, run-out and foot-flatness were within specification. As a note, we’ve stripped brand new motors from a variety of manufacturers and found bearing fits not to specification. So I’m saying our bearing fit measurements are BETTER THAN NEW.

Now on to balancing the rotor, most manufacturers as standard will balance their rotors to G6.3 (ISO 1940:1 2003) or G2.5 as an added cost option. We balance all our rotors to G1.0. Incidentally the rotor and cooling fan in this motor, weighing 58Kg had residual imbalance after balancing of 0.62g and 0.90g per plane. So I’m saying our balance grade is BETTER THAN NEW.

Ok so now the majority of the component parts of the repair have now being completed so it’s just a matter of fitting some bearings and rebuilding? Well no, not really. Bearing fitting is a skill and needs proper training and correct equipment to do this properly. I’ve heard many times in this industry people saying ‘it must have been a bad bearing!’ following a premature motor failureWell I don’t agree with that, I’ve never known of a genuine quality bearing being clever enough to be bad or destroy itself. Incorrect or poor fitting will damage a bearing for sure. I’m wondering how many people have blamed a bearing for their poor technique?

Trained engineer fitting a bearing after using a temperature controlled degaussing bearing heater in a clean environment

Now I’ll move on to the motor being assembled correctly and clean lubricant being correctly introduced to the bearings and lubrication channels, it’s time to carry out a test run of the motor. We ensure a one hour run test is carried out on our repairs and measure the bearing temperature rise and plateau temperatures. We also carry out a vibration test which would be expected, but what exactly is being measured and what do the results actually mean? We have set vibrations limits within frequency ranges that our repairs must be within before a motor can be classed as good-to-go, these are as follows.

Freq Range Hz Limit mm/s rms
15 – 40 0.72
40 – 60 1.35
60 – 175 0.72
175 – 425 0.54
425 – 1000 0.54
1000 – 2000 0.54
Acceleration Limit g’s peak
18 – 2000 0.5


We’ve sampled a number of new motors over the years and found that very few would even be under these limits. So I’m saying our repairs are BETTER THAN NEW.

There are a few useful features too that we add to our repaired motors that we believe enhance the repair further. Just taking a look at the repaired motor again, firstly a couple of coats of paint are sprayed. Can you see the dust caps on the lubrication ports? we feel this is a low cost yet vital addition, you don’t want dust being forced down a lubrication pipe, the first or any time you come to add lubricant. We label when the lubrication was charged and what lubricant was used too. A simple sticker denoting a roller bearing is fitted to the motor, useful so this motor isn’t used where a ball/ball arrangement is needed. The shaft extension is coated with an anti corrosion coating and an impact protection sleeve is added for greater protection. The feet too are coated with an anti corrosion coating and pads will be fitted to prevent damage. To finish, a repair label is added to the motor with our job reference number and to maintain our innovation in the repair industry, there’s a QR code on the label which links directly to the repair documentation for this motor.

I am saying that this repair is BETTER THAN NEW and there are many quality repair companies around the world who can say the same thing. So maybe it’s worth finding out the lengths your repair company goes to in repairing your equipment; maybe you will think twice about making the decision to buy new.

Incidentally this motor is an Ex tD A22 T125 Degrees C machine.

Fletcher Moorland is an IECEx Certified Service Facility working to IEC 60079:19, holding certificate no IECEx SIR0009 for d, e, n & c protection concepts.

To find out more about Fletcher Moorland visit or why not visit our workshops to see the repair process for yourself. call 01782 411021.