Category: Blog

Home  ›  Blog  ›  Page 2

AQ: DC link fault in 3 phase frequency inverter

Question:
Our one frequency inverter which drives 0.37 KW 400 V dosing pump motor intermittently (once in a month or once in two months) shows DC link fault and the speed is reduced to zero. This motor used to do changeover weakly. Pump NO: 1 never has such problem, pump NO: 2 only have this problem. We checked the motor found OK, checked the control circuit found ok, replaced with same new inverter still the same problem comes. We thought of incoming power supply problem so we swapped power supply cable from motor 1-2 but still the DC link fault comes in pump NO: 2. Then some of our experts said it is because the inductor is connected in the circuit, once remove the inductor this fault will not come again. But after removal of the inductor also same problem comes. From the previous history of work orders we found that this motor is a rewound motor, before rewinding there was no fault history at all. This motor is running always perfectly without any faults in manual control. Fault comes only in automatic control.
Could you please tell me what is the real problem?
Is it because of rewinding of the motor; winding geometry might have changed that affects the frequency inverter?
If this is the problem then why this fault is not coming whenever it is in service? (It waits for 1month or two months some time the fault comes in a weak also)
Is that the inverter will cause any problem because the inductor is in disconnected condition?
What is exactly the DC link fault and what are the reasons it can come in the inverter?
Why the DC link fault comes in when it is in automatic operation only?

Answer:
Have you compared the good unit to the bad unit?
Could there be any mechanical issues loading the motor?
Check that the current level on the bad motor is the same as the good motor.
It sounds as if the rewind data is not correct and the motor is taking high current. If the rewind data is correct the core loss may be high.

The procedures you have gone through would indicate that the motor is the issue. My advice would be to go to the OEM and purchase a new motor or if it is a standard motor your regular supplier should be able to supply them. It could even be beneficial to purchase two new motors and keep the existing good one as a spare.

AQ: Induction motor noise level

The noise level created by the motor at any speed is in a fixed environment, take two motors same HP, Speed, Enclosure, and the applied voltage could be a factor of the noise, the installed conditions of a 1000 motors could vary from alignment to load, to piping connected to load, to actual load.

What are you or the customer looking for? One 5 HP motor versus another 5HP motor, one in a 50,000 sq foot plant, the other in a 500 square foot plant, while the motor under ideal isolated test conditions might be X, the noise generated from the motor in different conditions could be blamed on the motor.

Would be interested in the question broken down to specific reasons/needs. I know very few who shop based on noise at particular speeds, most either accept the sound levels, which range from pitch to volume to whatever, as an irritant.

Often shielding of the motor can contain any noise that might be a factor in other areas around the motor.

I am not a manufacturer of motors, except for the modification of specialty applications. For example we changed out several hundred motors for the National Weather Service contained in a dipole antenna body. Existing motor was a single phase permanent split capacitor synchronous motor, 110 volt, 1800 RPM DESIRED, due to a feedback tachometer mounted on the motor to verify the speed as these receivers accepted upper air feedback of weather conditions, from weather balloons launched two to three times a day. Location and tracking of the balloons were critical, if the tach feedback was off by one rpm [from 1800] the tracking electronics could not deal with the inconsistency.

I attempted to purchase motors for this application, but because the motor was mounted vertically in a solid cone, no ventilation, plus they were single phase, with induction synchronous rotors, voltage was a consideration, and the units were mounted from Hawaii to Guam to Florida, across the US and Territories.

I took the existing single phase PSC SYNCH MOTOR, which few ever had the torque, or would stay at 1800 rpm, or fail do to the heat.

While they only needed around 300 plus active motors, they needed half as many as spares, considering the past history of failures and the lack of ability to deliver accurate timely weather data over an exact path.

It was not a case of excessive NOISE, it was a case of perceived sound, it sounded different, so for those involved with any Governmental Agency knows that form, fit, function is their mantra and excuse to not accept anything.

We had several complaints of noise, turns out the noise was in no way a danger or at levels of any concern, just different.

While testing 4. 6. 8. 2 pole motors for “noise” in a controlled environment, is only data from those conditions, out in the wild west, those conditions are going to change, mounting, structure, all explained above will affect the motor’s “noise” levels, or perceived “noise” levels.

In the fact that no load, [NEMA] testing is not going to be exacting as other possible more exacting, different parameter type testing, if noise is a concern, is under full load, which again is a variable.

How many vanilla NEMA motors ever operate at “full load”?

Many run below the full load capacity, let alone service factor capacity, some operate slightly overloaded, few ever see the exact applied voltages, with changing of applied voltages during seasonal or daily changes in many variables effecting voltage supply.

AQ: Motor babbitt bearings

Babbitt Bearings, properly cared for and maintained will last much longer than anti-friction bearings. Properly set up and aligned there is no contact between the babbitt and the journal other that a scuff mark at the bottom that is caused by minimal contact at starting. As long as the motor is properly aligned to the load, the oil is kept clean and continually fed to the bearing the bearing will last a very long time. There are two simple checks that should be done to check if all is well on Babbit Bearings. The first one is oil sampling. Cheap insurance which will tell you what contaminants are in the oil. The second one is an annual check on the air gap at the bottom of the motor. If the air gap is found to be getting smaller your bearings are wearing.

Babbitt bearings are normally found in larger motors and almost always in direct coupled applications.
In high speed motors the replacement of anti-friction bearings is essential every 1-2 years depending on the severity of the application. It is not uncommon to find a 2-pole Babbit Bearing motor, 20 years old or more with the original bearings. These motors can be overhauled and the windings cleaned up on a regular basis but the original bearings are re-installed.
Babbitt bearings are also affected by shaft currents and we often find a NDE babbitt bearing insulated from the housing.
Babbitt Bearings are also much quieter than anti-friction bearings. Another area where sleeve bearings are very common is in the fan motors in home furnaces. If ball bearings were installed in these motor you would get and annoying clicking sound coming through your ductwork. The bearings in these motors are not made from babbitt, there are made form an oil-bronze material.

Babbitt Bearings are much more expensive than anti-friction bearings and you can’t buy them off the shelf like a 6316. If they are worn, (normally because of mis-aligniment or lack of proper lubrication), they need to be re-babbitted. For a high speed motor with a 3″ journal the re-babitting can cost in the region of $1,200.00 to $2,000.00 per bearing.

When a new motor is being purchased it will cost much more with Babbitt Bearings than it will wit anti-friction bearings and users should be aware that, in the event of a bearing failure, it will not be a 2-3 day turn around on the motor.

Babbit bearings have an initial high cost but properly looked after they are more economical over a long period of time.

AQ: Aluminum or copper wire in motors?

Some motor manufacturers go from CU to Al because they try to reduce costs.
Then it just works the other way. Al wire needs to have a larger diameter than Cu if you want the same motor performance.
Then you may face problems with the slot opening and the slot fill, there may not even be enough space at all for the Al wire. You may need to change your lamination.

Furthermore, the blade gap of your inserting tools may not be suitable anymore so you will need a new set of toolings.
Also the end turns will have more volume which may cause problems at the end turn forming and even assembling process.
Besides, due to the properties of Al wire, your rejects will increase during the manufacturing process.
These are just a few examples.

Before changing to Al wire, manufacturers should consider the pros and cons carefully.
Some may invest more than they will safe with the cheaper Al wire.

Copper – at least at the purities and alloys used for electrical conductors – is fairly scarce, which tends to make the price pretty volatile. Aluminum, on the other hand, is fairly abundant in the alloys used for conductors … and hence pretty stable in price (not to mention cheaper than copper).

Neither raw material (copper or aluminum) is used in its pure form for electrical conductors. Both have some other materials added, primarily for mechanical strength. The key factor in determining how much of each to use is the conductivity: 98 percent for the typical copper alloy (ref UNC C11000), 61 percent for the 1970s aluminum alloy (ref EC 1350), or 56 percent for the modern aluminum alloys used in busbar material (ref alloy 6101).

Material properties:
Tensile strength (same cross section) lb/in2: Cu = 50000 Al = 32000
Tensile strength (same conductivity) lb/in2: Cu = 50000 Al = 50000
Weight (same conductivity) lb : Cu = 100 Al = 54
Cross section (same conductivity) % : Cu = 100 Al = 156
Coefficient expansion per deg C x 10^-6 : Cu = 16.6 Al = 23.0

The choice between Al and Cu usually boils down to either cost or weight.

Care must be taken because Al is not as strong (more problems with the forces generated by fault conditions) AND because it has a higher susceptibility to dimensional change under high temperature conditions (such as those occurring during electrical faults).

Another consideration for an aluminum-winding machine are the connection points for real-world transmission: care in terminations is a must. Galvanic action between dissimiar materials is a known difficulty that can be further aggravated by airborne (chemical) contaminants.

AQ: DC Motor Armature Testing

For a DC Motor Armature, There is a simple method of determining the condition of the Armature.
Drop Test Method: Give a DC Voltage across the commutator Segments for one pole pitch area from a Power supply or Battery. Connect Positive end of the DC power supply at one end and the Negative end at the opposite end.
For example if the total number of commutator segments are say, 40 in the armature to be tested and the total number of poles is 4, then one pole pitch area will be 10 segments.

Now measure with a Milli volt meter say 0 to 10 millivolts range, the Voltage Drop at the center point, that is between 5th and 6th segment. again rotate the Armature Clockwise or Anti clock wise and measure the next set of segments.
Like this complete measurements for all the 40 segments pairs. simultaneously recording the readings.

If there is any defect in the winding, that is shorted or open, it will show in the readings.
If the reading of Milli volt Meter is uniform for the all the 40 segments pairs, than the armature is good. If there is short between winding or the winding coil between one particular pair of segments, the reading will be less drop in millivolts. If there is any loose or open, the reading will be more than normal readings. Thus one can determine the condition of a DC armature for short or lose or open winding.

When testing a DC armature there is a series of tet should that should be done. The first is. Ground insulation test or more commonly known as a mugger test, usually done at 500VDC. If the ground reading is above 1 meg ohm the armature is good to go to the next test which is a bar to bar test. There are 2 pieces of equipment to conduct this test the best. One of these combined with the mugger test will tell you if the armature is satisfactory return to service. The first bar to bar test is conducted with a “DLRO” digital low resistance ohm meter. The meter will circulate about 8-10 amps thru adjacent successive bars and measure the milli ohm resistance of the circuit. If there is more than a 5% variation then the armature is shorted turn to turn. The next tester which is called a high frequency bar to bar tester. The tester has 4 tet points and as you move it around the armature a high frequency voltage is introduced across the pairs of successive windings and the meter will show a variation if there is a shorted turn. If it passes either of these 2 bar to bar test and the ground insulation test then it can be returned to service.

AQ: Improve induction motor efficiency

The efficiency of an induction motor is determined by intrinsic losses that can be reduced only by changes in motor design. Intrinsic losses are of two types: fixed losses – independent of motor load, and variable losses – dependent on load. Fixed losses consist of magnetic core losses and friction and windage losses. Variable losses consist of resistance losses in the stator and in the rotor and miscellaneous stray losses. So by reducing these losses we can improve efficiency of induction motor.

Changing the rotation direction will not improve efficiency.
Core loss and copper, those are the dominant losses. Improve them and you will get better efficiency. Changing the slot shape etc will help considerably, as will using copper in the rotor. BUT, you can’t do either one without affecting the performance of the motor, specifically the starting torque and current as well as the maximum torque and current. In addition, if the motor is designed to have aluminum cage, then changing the cage material to copper won’t help the efficiency much since the rotor slot and end rings are not optimally designed.

Improving slot fill will help your copper loss, by putting bigger wires in the stator slot, the wire resistance will reduce and the copper loss will go down. Reducing the end turn height of the windings will also help reduce copper losses.
Stray losses are the only one which can improve efficiency without affecting size of the induction motor. This can be reduced by reducing harmonies in the machine, which can be controlled by selecting slot combination, winding layout, size of air gap, saturation, concentricity of air gap etc.

If an induction motor has to run in both direction and uses a bi directional fan it is inefficient. uni directional fans are used in higher ratings to improve efficiency. further direction of rotation is determined by the driven equipment and cannot be changed at will. Minimising losses both core and copper and stray losses, better cooling ,improvement in cooling fan design a combination of all this suitably balanced will improve efficiency but there is always a limitation on max value imposed by certain conditions of application, materials, willingness of customers to pay.

AQ: What factors cause Current unbalance

1. Voltage unbalance in supply side (1% volts could easily be 10% current).
2. Physical differences between individual stator coil shapes and connections causing small (but noticeable) resistance changes.
3. Unsymmetrical magnetic circuit – not as big a deal in the smaller “ring” lamination designs, unless highly saturated.
4. Lightly loaded machines will exhibit far higher unbalance than those loaded closer to the full nameplate rating (mostly due to the magnetizing current requirements and associated core/stray loss).

For quick solution measure the current in the three phases, then change the three supply terminals by shift the three terminal to rotate the motor in the same direction, and measure again the current, if the high current move with a certain phase (example: phase L1 of supply read high current in the two case above) the problem is from supply, you can then measure the voltage at motor terminal to be sure that the control circuit and cable are good.

AQ: What’s the difference of variable frequency drive and soft starter

variable frequency drives are two different purpose products. VFD is for AC motor speed control, it’s not only change the output voltage but also change the frequency; Soft starter is a regulator actually for motor starting, just changing the output voltage. Variable frequency drive has all the features of soft starters, but the price is much more expensive than the soft starter and the structure is much more complex.

Variable frequency drive is converting power supply (single phase VFD and three-phase variable frequency drive.

Soft starter is a set of motor soft start/stop, light-load energy saving and various protection functions devices to control motors.

Soft starter uses three opposite parallel thyristors as regulator, plug it into the power source and motor stator.  When using soft starter to start the motor, the thyristor output voltage increases gradually, and the motor accelerates gradually until the thyristor is turned on completely. The motor operates at rated voltage to achieve a smooth start, reduce starting current and avoid start overcurrent trip. When the motor reaches rated RPM, the startup process is completed, the soft starter uses bypass contactor to replace thyristor to provide rated voltage to the motor, in order to reduce the thyristor heat loss, extend the soft starter service life and improve efficiency, also avoid harmonic pollution to the power grid.

AQ: VFD control loop circuit faults analysis

The affection on variable frequency drive life in the control loop circuit is the power part, the buffer capacitor in smoothing capacitor and IPM board. The ripple current pass the capacitor is a fixed value which won’t be affected by the main circuit, so its life is mainly determined by the temperature and power-on time. Since the capacitors are soldered to the circuit board, it difficult to determine the capacitor deterioration by measuring the electrostatic capacity. Generally, we calculate its life base on the ambient temperature and service time.

Power supply circuit provides power to the control circuit board, IPM drive circuit, operation display panel and cooling fan, the power is obtained from the main circuit DC voltage rectified by the switching power supply. Therefore, if one power short circuit, besides itself damaged, also affect other parts power supply, such as misoperation causes power source and the public ground short circuit, result in switching power supply circuit board damaged, the fans power supply short circuit etc. Generally it’s easy to find out by observing the power supply circuit board.

Logic control circuit board is the core of a variable frequency drive, it includes CPU, MPU, RAM, EEPROM etc large scale integrated circuits, the failure rate is very rare due to high reliability. But sometimes all control terminals closed simultaneously during startup which will cause the VFD drive appear EEPROM fault, in such case, just reset the EEPROM.

IPM circuit board contains drivers and buffer circuit, and over-voltage, phase loss protection circuits. Logic control panel PWM signal input to IPM module by voltage drive signal optical coupling, so, we should measure the IPM module optical coupling during module detection.

AQ: 3 phase induction motor designs

For 3 phase motor designs, there is hardly any slot combination that will yield a perfectly smooth torque-speed curve. Keeping the following rules in mind will (mostly) avoid the combinations that tend to amplify magnetic noise, harmonics, and parasitic torques.

Let the number of stator slots be S, and the number of rotor slots be R, and the number of poles be P. Undesirable combinations occur when any of the following are true:

1. S – R = 0
2. S – R = +1 OR -1
3. S – R = +2 OR -2
4. S – R = +P or -P
5. S – R = +(P + 1) or -(P +1)
6. S – R = +(P + 2) or -(P + 2)
7. S – R = -(P * 2)
8. S – R = -(P * 5)
9. S – R = +(P * 3) or -(P * 3) .. or multiples of +/-(3 * P).

We know the stator should have an even number of slots to make winding easier – although for certain pole counts, it too can be an odd integer value. And except for a few cases, the number of rotor slots can be either even OR odd.

Then it comes down to the accuracy of the compound die or indexing die for the slot stamping.