I am currently investigating the design of a three phase axial flux PM motor, but replacing conventional materials with high temperature superconductors. I'm interested to know the thoughts of group members regarding design rules/rules of thumb relating to the number of stator coils and rotor poles. Many in the amateur wind turbine community seem to use a 4:3 ratio (magnets:coils), but I can't seem to find anything 'official' on the topic.
An equal number of magnets: coils would cause problems with starting the motor and with cogging/torque pulsations.
The only textbook I've found dedicated to the design of axial flux PM motors is Jacek Gieras's book on 'Axial Flux Permanent Magnet Brushless Machines', but this seems only to mention examples of coils: poles ratios (e.g., 12 stator coils and 8 rotor poles, 9/8, etc.).
"Design of Brushless Permanent-Magnet Motors" by J.R. Hendershot Jr. and TJE Miller is an excellent design book and pages 3-50 thru 3-55 illustrate the 3 phase winding patterns you describe (8/6, 8/9, and 4/6). Whether axial air gap or radial air gap the principles are the same. I assume with an axial air gap machine you do not want phases overlapping each other, that is the common factor in the three patterns above. This keeps winding simple and compact and is commonly used on smaller 3 phase brushless motors.
These windings do not automatically guarantee a true BEMF sine wave form. If you want a sinusoidal waveform you will have to do some work on tailoring the magnetic design (gap between magnets, skewing, air gap profiling, etc.). Some servo motor manufacturers do just this to get a true BEMF sine wave to match their sine wave controllers for ripple free torque operation.
Another decision is does the coil center have a laminated steel pole or only and air center. Air gap windings should be axially thin and have no hysteresis component which is good for high speed operation. A slotted pole winding can handle more wire bulk but a laminated construction may be difficult to implement, you might look at an AC Powdered Metal for the Armature and teeth.
If you allow phase coils to overlap there are a great many other winding patterns possible (listed in the reference book), some are better for Trapezoid controller drive and some are better for sine wave controller drive (BEMF should match controller drive type). Just depends on you end goals.
AC Motors - Variable torque: AC motors have a speed torque characteristic that varies as the square of the speed. For example, an 1,800/900-rpm electrical motor that develops 10 hp at 1,800 rpm produces 2.5 hp at 900 rpm. Since ac motors face loads, such as centrifugal pumps, fans, and blowers, have a torque requirement that varies as the square or cube of the speed, this ac motor characteristic is usually adequate.
AC Motors - Constant torque: These ac motors can develop the same torque at each speed, thus power output varies directly with speed. For example, an ac motor rated at 10 hp at 1,800 rpm produces 5 hp at 900 rpm. These ac motors are used in applications with constant torque requirements such as mixers, conveyors, and compressors.
1- Synchronous motors generally offer more efficiency than induction ones, and hence in higher ratings (about 5000 hp and higher) they may be more cost effective considering Life Cycle Costs. The exact size of preference to switch to Synchronous shall be determined based on LCC analysis of specific application.
2- A Large reciprocating compressor is a highly variable load and a synchronous motor will keep its speed in this situation while the induction motor would respond with fluctuating speed.
3- Based on API 618 (with reference to IEC and NEMA), a synchronous motor used for reciprocating compressor may tolerate 66% variation in current, while an induction motor is allowed to have only 40% variation in current which in larger compressors may be exceeded (because of variable load).Also Higher efficiency induction motors with less slip, cause more current variations and are prohibited.
Synchronous motors are characterized by limited starting torque, the ability to actively control power factor and less current in-rush than the induction motor. The synchronous motor also requires active matching of torque demand with motor output. Synchronous motors started “across-the–line” also produce oscillatory torques at the twice slip frequency during acceleration (i.e., starting at 120 Hz and decreasing to 0 Hz at full speed). These torques generally require additional transient torsional analysis because of the potential for damage.
Synchronous motors are usually advantageous on slow speed applications (e.g., low speed reciprocating compressors operating from 200-400 RPM) and also on machines larger than about 10,000 to 15,000 HP. With both motor types, it is important to match the compressor torque versus speed requirements with motor torque versus speed capabilities as discussed in Sections 6.0 and 7.0. Both induction and synchronous motor types can be coupled with a VFD for variable speed operation.
If the motor is being driven by a variable frequency drive with sophisticated drive algorithms, i.e. controllers that can track the load torque variations, then both the efficiency and transient stability problems can be solved together.
The other significant thing is the starting problem. The transient load torque is also present at starting so the motor has to be able to accelerate through the load transients and be capable of starting when the compressor is sitting at the highest load.
How to add a separate AC line reactor / DC choke in case the variable frequency drive doesn't have it? Can we use a separate line reactor if it's not built in with the VFD drive? What all parameters I would have to look into, if I want to add the line reactor? Is there any sizing criteria? How would I have to install it?
It depends on how much THD you want to have and how much money you want to spend. If this is for electric motor protection there are additional methods of spike suppression and better reactors/filters.
Size for amps and voltage.
THD will vary will design and specifications. You want the reactor to filter or tune out the unwanted frequencies, mainly the AC drive carrier frequency. One often overlooked parameter is what rejection frequency the reactor is wound for. You want a reactor wound for the rejection frequency you have your VFD drive set at.
This will make you want to raise the carrier frequency to make the reactor smaller, less turns, and less expensive. Before you do this look at the de-rating tables and other factors involved with a high carrier frequency.
It's always best to first check with your VFD installation and operation documentation. It is likely that the motor drives manufacturer makes recommendations for reactor ratings. That said 3 to 5% reactance at the VFD drive's rated input current is always a good solution. If there is no internal bus choke or reactor in the VFD then use 5%. Don't sweat the voltage drop. The drop is in quadrature to the source voltage and so mostly subtracts at a 90 degree angle. Thus, the drop will be less than half the %reactance.
As far as I know all variable frequency drives with vector control can also be run with just V/F control.
A drive in vector control mode has several tuning parameters to increase or decrease motor performance. With factory default parameters a VFD in vector mode will have higher performance than a drive in V/F mode. Sort of like a "sport or racing" computer option in a modern automobile.
Depending on the application using vector control can use a lot more power. If you have a rapidly surging load the vector may be really struggling to keep the speed constant while a variable frequency drive in V/F mode never notices the speed change. If the application has a steady mid-range speed and load or has a slow rate of change a vector and V/F may be very close in amp draw.
If you have an application where you need the vector for starting or stopping quickly but you are using a lot of current at speed you can change vector parameters to reduce the current. In some applications it is cheaper to oversize a V/F drive to get starting or stopping torque if you don't need precise speed control.
I accept the fact that, in the practice, V/f is considered by many the better choice for fan loads, but I see few reasons why V/f approach could result in better efficiency.
One reason could be that, since it doesn't try to regulate anything, practically it can't oscillate due to weak stability, although oscillations may still occur (I've seen a heavily vibrating torque measurement on a fan driven by a V/f variable frequency drive).
Another could be that, while non-linear V/f curves (suitable to non-linear loads as fans) are quite common, the same is not done for the flux reference (magnitude) in vector control.
And, of course, the few parameters of a V/f control are far easier to tune than a vector scheme (which companies don't really share).
However, one interesting thing that can be done with vector control is, for slow dynamics applications, to automatically tune the flux reference to achieve a minimum loss control during the control operation. I don't think this would be possible with V/f.
From a manufacturing economics standpoint, there is often a trade off in the decision to add a DC bus choke or not based on its ability to reduce the DC bus ripple. This is because it can reduce the DC bus capacitance necessary to present a clean DC source to the transistors. For some AC drive manufacturers who have the internal capability to wind their own component chokes, this often represents a component cost benefit compared to buying capacitors from outside vendors and being more subject to market volatility. On the other hand if the AC drive manufacturer IS also a manufacturer of capacitors, it works exactly the other way around.
I believe this is why we often see small component class drives being made without DC chokes primarily by companies, mostly in Asia, for whom capacitors are a very low cost commodity. When EU and US manufacturer make larger variable frequency drives, it's usually less expensive for them to wind chokes, but that option is often perceived to be too physically large for component class drives so they farm out their designs and production to Asian manufacturers. Ironically then, users will add an external AC reactor anyway, but fail to observe that the overall footprint is now larger than it would have been with a DC choke.
I attribute this to the same false market perception that society uses in buying airline tickets. We now shop on the internet based on one criteria, price of the ticket. The airlines have finally figured that out, so they now appear to have lower ticket prices, but charge us extra for bags, snacks, leg room etc. and we actually are paying MORE than we used to. So to relate that back to the AC drives, the market demanded smaller and smaller packaging of VFD drives, which became a primary selection criteria, leading to the smallest physical package, the ones without DC chokes, being dominant in that low kW realm to the point where virtually everyone else gave up and joined the party.
That said, there is still validity to the added protection for the front end of the AC drive provided by the reactor compared to a DC choke. If there are multiple AC drives in an enclosure however, that benefit can still be realized with one larger reactor ahead of the entire inverter drive input power circuit.
The dU/dt at the output of the variable frequency drive combined with the motor cable length will result in very high voltage peaks at the motor terminals. This is a concern for the isolation in motors not designed to be driven by VFDs.
On the other hand the maximum motor cable length depends also on the switching frequency used due to the charging effect of the motor cable capacitance (this is a limitation on the variable frequency drive side, not on the motor isolation).
The dU/dt at motor terminals normally is very different from the dU/dt that you can calculate from IGBT and its driving characteristics (turn on time, gate resistor, etc) at variable frequency drive terminals. As the cable acts like a distributed LC impedance, the dU/dt calculation on VFD terminals will give you very high values that can be apparently dangerous, but in practice, will not happen at motor terminals.
For long cables, the combination of cable impedance, high frequency input impedance of motor and VFD switching frequency can lead to reflection of voltage pulses that gives origin to large voltage overshoots on motor terminals. The problem increases as increasing switching frequency because the time between voltage pulses will be smaller, so, a voltage pulse reaching the motor will add to the pulse being reflected. This “double pulsing” can results in extreme voltage overshoot and dU/dt that will result in motor insulation failures. For the variable frequency drives side the increasing switching frequency will be a problem (besides power losses) if you have a big capacitor filter at converter output, that can lead to high current pulses at inverter side.
The determination of the resulting dU/dt at motor terminals from the dU/dt at VFD drive terminals is very difficult if you try to use simulations. For this task you’ll need the high frequency parameters of cables (that also depends on installation details) and motor, that will not be available from standard datasheets and are very difficult to obtain from measurements. In practice almost all VFD manufacturers make extensive measurements and establish some criteria in order to orient applications. The approach is to determine if it is necessary or not to have an output filter for a known application (cable length).
For instance, a common specification is:
For cable lengths up to 100 meters (and motor suitable for variable frequency drive applications) it is not necessary a filter; for lengths from 100 to 200 meters, a series reactance can be used; for greater lengths it is necessary an LC filter at VFD terminals. The limit lengths can be different from different manufacturers and voltage levels (LV/MV). Iacdrive, for instance, can give complete orientation for application of its drives considering the needed cable length for the application.
Always the top brands will be the most popular PLC and over many years it is my opinion that this is because of their marketing strategy, history, reputation and worldwide acceptance more than any other reasons. This does not mean they are better or worse in any way, just means they are more accepted world wide and more people are experienced with their software. Thus there is some security for the owner in respect to programmer support or future resources etc (people come, people go) and a basis on which management may dictate what hardware is used. There is also the consideration on the capital outlay for programming software which can be very expensive.
Choice most often depends on your application and infrastructure. Example: if an entire factory or whatever was "x-brand" and communicating with each other through "y-protocol", it may be wise to keep to the same-same. Other brands PLC may talk same protocol but then you need to think about software and the experience of your programmer resources, spares etc.
The alternative may be a more task or machine specific PLC that can communicate the same protocol but at the cost of the programmer not knowing the device or software, or the costs of additional software and also there may be less skilled programmers in this hardware choice constricting the owners future options in using this alternative.
Experienced programmers fall into two basic categories. Just like Joe-Builder who has had 25yrs experience - now Joe, was that 25years experience doing different things or was that 1years experience 25 times? I have encountered this so often, fantastic CV but doesn't know anything because has been in same job, day in day out, year after year. Very good at THAT job mind you but no real (other) world experience. PLC programmers are often the same, know x-plc (or software language) inside out but nothing else.
Just my opinion but a good programmer is someone skilled in ladder logic, functions / function blocks, structured text, CRC etc and knows when to use it. Someone also familiar with the hardware and its associated costs. Someone who knows how the hardware device scans and can makes efficient use of its resources through the above mentioned skills. Someone also who is mind-full of who will maintain / modify and what can be modified and what should not... etc. Bit of a mouth full I know, but such a person can then make choices of hardware based on the end result required and not be constrained in his/her thinking based on what already exists or what they themselves know or what they or their management consider to be the current reality.
So, a long story to ask another question. Are you really asking which is the most popular brand PLC because a quick google search using the a brand name would tell you that in seconds based on the number of millions of pages available for THAT brand or are you asking which PLC should you choose?
As further comment...
Today I would go task specific by choice. If you want ultra speed, complex math or fast analogue and. or heavy processing etc... then you are looking at a soft logic PLC that will talk the same protocol as the other PLC's in the factory. If the task is simple logic and minimal analogue and does not require ultra fast scan times (i.e. 10ms+ is acceptable) then many top brands offer a range that will do this.
There are many things you can do in ladder logic that will satisfy a situation admirably. There are lots of things you can do in structured text that is impossible / impractical to do in ladder logic. All soft-logic PLC's I have experienced are totally useless at complex ladder logic. This is WHY I choose by what the task requires as opposed to choosing because of what constrains my current reality thinking or comfort zone.
The end result is a functional task, machine or project that is maintainable - not what a particular
We recently evaluated a 500 HP 4 pole motor on a pump application. The motor is started with a soft starter. Upon examination of the bearings we discovered fluting inside both the variable frequency drive and opposite drive end bearings.
If it were shaft currents, especially on a pump, the fluting would be typically on the non-drive end only, excess shaft current would be drained through the apparatus attached to the drive end shaft. We would more likely suspect a vibration issue with the assembly while inactive. What is base condition for the pump? Is it on a stable foundation or is it mobile? If mobile, and transported you need to "lock" the shaft to avoid axial or radial motion.
PAM winding is still a feasible alternative to VFD where simply two or three discrete speeds are necessary without the need for servo-like control, mostly for high power applications as was mentioned above. Only several extra leads and contactors but no nasty harmonics, reduction of insulation life and no additional variable frequency drive that takes space & is not cheap to buy or maintain, might become obsolete and most likely will not last as long as the motor.
Note that some shaft couplers are insulating; and therefore, won't drain shaft voltages.
However, all of the soft starters that I have used are line (mains) frequency phase angle modulating. Hence they act as three phase variacs (variable autotransformers). I have not run across any stray voltage problems with these units. However, some soft starters modulate only two of the three phases. I don't know what this will cause.
Regarding VFD's, three steps are needed to protect the motor: 1) High enough winding voltage withstand voltage (dielectric strength), 2) Adequate thermal capability to counter the extra (5% or so) winding heading due to the harmonics, and 3) protecting the bearings from developed stray voltage (grounding, bypassing or insulating).
A soft starter is in the circuit for so short a time, it is not likely that the fluting is coming from the drive. My logic is that fluting is a low current long time event. Bearing damage that could occur from the very short and very infrequent duration of starting would have to be a very high energy (for that short time), and would more likely be pitting.
In evaluating all possible sources:
There have been instances where the external current is coming from the plant piping. This would be eliminated by insulating the piping from the pump (if a flanged connection, use an insulative gasket [no metal fibers or rims], and plastic sleeves & washers for the bolt set).
Other motor related sources: the API motor specs say to insulate one end where the shaft voltage exceeds 500 mV. This can be done many ways, and usually done on the non drive end. (Have you measured the shaft voltage?)
I am not a big fan of shaft grounding brushes, and grounding the plant piping may not be enough. Brush contact is not reliable, and may not drain all the current (same for grounding the pipe).
Anecdotally: an electric utilitie had system grounding problems that elevated the potential of "ground" in a dairy. The path to lowest potential was through the cow to the milking machine to "ground". Milk production went down, it took a while for the farmer to get the utility to check their system. Finally they did, fixed the transmission system grounding, and the problem disappeared.
Is it AC line reactor more important than DC choke in a frequency inverter? If AC line reactor is missing in the inverter, what are possible impacts to the inverter? And how about DC choke?
Quality frequency inverters incorporate either an AC Reactor or DC Reactor (choke). Their inclusion in the basic design of the frequency inverter allows the design engineer to maximize the advantages of the choke. Their function is to reduce the current distortion caused by the input stage rectifiers by slowing the rate of change of current, and thus charging the internal capacitor at a slower rate over a longer time.
The Harmonic Distortion caused by a frequency inverter is related to its size & load, choke size, and the supply network parameters.
With no AC Reactor or DC Choke, the harmonic distortion will be greater.
Another consideration should be a properly sized source transformer that provides enough impedance. The sized source transformer used as an isolation transformer (although a bit more of an investment) should provide 3 to 5% impedance yet also provides Voltage Transient mitigation with ten to one reduction in impulse peaks, as well as noise reduction through the use of a Delta primary to Wye secondary with center tap ground. It provides additional protection for the frequency inverter front (Converter) end while proper ground of the Source to inverter, frequency inverter to Motor and Motor to Voltage Source assists in mitigating high frequency noise, especially when flat braid is used as the grounding straps. This protects your investment and assists in keeping the variable frequency inverter from generating noise into the supply that can compromise your nearby instrumentation, and PLC power supplies, etc. As well you can tap up the transformer giving you a higher input voltage mitigating the voltage drop issues resulting from the higher impedance.
The DC link assists in mitigating DC Bus Ripple and increasing the input impedance enabling a slower inrush for power on and sudden demand current requirements furthering the life of your capacitors, while a sized supply transformer protects the front end of the frequency inverter drive by providing voltage noise protection and adding input impedance for smoother current and adding a capability to change taps to prevent a voltage drop, while input reactors slow inrush current furthering the life of your input components and capacitors but add no protection from Voltage impulses or noise to the drive converter components, and add voltage drop increasing stress on those components. The important thing to remember is that "Proper" systemic design protects your frequency inverters and system components investment.