Motor-Drive Discussion--Lengthy!
Ok, Russ, here goes! First, some analysis and then some conclusions.
Let's start with analysing the load on the motor remembering that the motor lead amps are the vector sum of magnetizing amps and torque-producing amps. Nameplate data gives us 447 total amps full load and magnetizing amps on this motor will be about 125 amps. Solving for full load torque amps gives us 432 torque producing amps. This means that at no shaft load (0 ft-lbs) there is 0 torque producing amps and at rated load (1200 ft-lbs) there will be 432 torque producing amps.
Russ tells us that at 56 Hz, the total amps is 370. Again, solving the vector equation remembering that magnetizing amps are still 125, we find that there are 352 torque producing amps. 352/432 is .81 which tells us that the motor is producing 81% of its rated torque. From this we can conclude that the motor is not overloaded at this speed and load.
Russ also tells us that at 26 Hz, the total amps is 125. Since this is essentially the same as the magnetizing amps, we can conclude that the motor is developing almost no torque which would be normal with a fan or centrifugal pump load. (It must be that the magnetizing amps on this motor are slightly less which would leave a few amps left to produce a small amount of torque). But certainly we can conclude that the motor is very lightly loaded at this speed.
From the above analysis, we can safely say that overload is not the problem in this application.
Next, let's look at temperature data. The nameplate states Class F insulation and also mentions thermistors set for 155 degrees C. Checking the Electrical Engineering Handbook, we find that Class F insulation is rated for 155 degrees C so this matches. Now, 155 degrees C is mighty hot! That's 311 degrees F! That's right, the windings can withstand temperatures up to 311 degrees F. A common rule of thumb among motor people is to figure that this will permit up to about 180 degrees F on the outside skin of the motor. That is still plenty hot and the touch test is no good at this temp. 150 degrees F is too hot to hold your hand on. Now, it may be that these motors are judged to be too hot by the "hand" method or by the thermistor method. (Russ didn't say how he concluded the motors were too hot). I suspect that the motors are very hot in an Alabama plant environment but not too hot. Certainly, they are not too hot because of any overload condition or lack of cooling due to slow shaft speed.
Next, let's look at the phase-ground failures. Winding failures due to reflected waves from drive-generated pulses almost always cause phase-to-phase failures and very rarely phase-to-ground. When a new motor fails phase-to-ground it almost always is a winding defect and I suspect both failures were of that type. Any good motor shop can analyse a slot insulation failure to determine whether it was temperature induced or something else. Weg actually should be able to tell you this if you press them. From the above I conclude that the phase-ground winding failures were not caused by the drive system and that motor lead reactors or reflected wave filters are probably not required.
Finally let's look at the bearing failure. In my view, this is the big issue and needs to be addressed immediately. It is a well developed fact that motors in NEMA frames 400 and larger and rated 250 HP and larger can and often do suffer from currents being induced into the rotor shaft. (Tom Jenkins mentioned this earlier). This current has no place to go in a standard construction motor and ends up building up enough voltage in the bearings until the arc punctures the lubricating film and blows thru, in the process pitting the race and the ball or roller. It doesn't take much of this before the bearing goes straight to catastrophic failure. Russ' motor has done that and I am sure that, in his 586 frame motor, shaft currents are the cause. A good motor shop and certainly Weg can identify this type of bearing failure by examining the failed bearing.
Now, the conclusions. First, to solve the bearing problems, the motors must be modified with either insulated bearings or a shaft-to-frame grounding brush system. The preferred way on a new motor in this HP range is insulated bearings but that would require pulling the motors and modifying the end bells. For operating motors, it is often easier to install a slip ring on the motor shaft and a carbon brush assembly that grounds the slip ring and shaft to the frame. This can usually be field installed. You will find that most large reputable motor shops will be able to do this--the mom-and-pop variety will not. You don't really get a choice on this one, Russ. You will eat bearings and maybe even crash a rotor until this is dealt with. And I wouldn't necessarily call this Weg's fault since the motor is a general purpose design. ("Inverter duty" gets even fuzzier in over-NEMA motors)
Second, I would doublecheck that the motors are really running too hot. If your thermistor system is properly calibrated, depend on it rather than any kind of touchy-feely approach. (Russ, if the thermistors actually are indicating overtemp, let me know. We may have to rethink this). It is very well possible that bearing deterioration is, even now, causing additional heat. If you have the ability to do vibration analysis, you may want to take a look at the motor bearings. They may be leaving you as we speak!
Third, the drive may give you the option of several different carrier frequencies. Sometimes, this will be disquised with names like "low noise" or other code words. In your industrial application, it is best to select the lowest carrier frequency the drive offers. It is easier on motor insulation and further reduces any chance of reflected wave trouble. Check with Robicon (good luck!) if you don't see it in their user manual.
Finally, if this in some way doesn't make sense to you, Russ, let me know. I've tried to operate on the data presented but, if something important has been overlooked, please say so.
And, for everyone else. How'd I do? Still think I'm trying to hide something? (Ok, Ok, I've learned my lesson. Maybe I should read the rules!)