I would like to control the power of a three phase Linear Induction Motor (LIM) with phase angle control. At my hands I have a couple of SSRs (Solid State Relay) and a microprocessor. I get a trig to the CPU when the first phase crosses zero.
Now the only (?) thing I need is the algorithms and/or tables to calculate the needed firing angles to supply the whished power to the LIM. With a single phase this would be rather simple, but three phases drive me mad.
I have made an extensive search on the net without finding anything but finnished products. Seems hard to come by material about needed algorithms and example control programs.
Would be great if anyone could provide some useful info!
Hi, you cant use SSRs, you need 2 thyristors for each phase. Use an opto coupler, 1 for each thyristor pair, to give you feedback signals to the micro for sync. When a thyristor pair go off a voltage appears across them, this turns the opto on which signals the micro. The software times from this point for a delay period and refires the pair at the end of the delay. You need to fire 2 pairs at once to get it started but after that its self sustaining as long as you dont go too low. If you need to phase down to zero output then you have to fire the six thyristors in pairs but on different phases, you can get the sequence from any textbook on 3 phase dc drives.
If you want to control speed (not torque) on a linear induction motor, I think you're going to need a VFD (variable-frequency drive, not vacuum fluorescent display). Those devices output a synthesized sine wave at the required frequency and voltage to maintain torque and change speed. They are not a beginner project in any way. The last big one I heard of (big linear motor) had an "event" that resulted in the vaporizing massive bus bars and cubic meters of control cabinet due to a firmware "issue". On a smaller scale, one tends to blow a few sets of expensive power semis.
If you just make a 3-phase phase control you'll just reduce the torque. In some special applications (fans, for example) that is good enough as the motor won't stall over a fairly wide range and the slip increases, thus changing the RPM.
Best regards, Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
The solution with SSRs was recommended by the company that manufactures the motors and I was under the impression that they had tested it and knew what they where talking about, but since you seem to be well informed I'll have to call the guys and confirm.
As far as I knew a SSR is basically composed of a couple of SCRs (thyristors?)? I'm just the computer guy that are going to program the micro but I'm having kind of a crash course in high power electronics now :)
I guess that the "any textbook" that you are referring to is what I'm after. Could you perhaps name one or even better, point to some resource on the net?
Would you care to elaborate on that? Since the SSR solution was recommended by the guys that manufactures the LIM I'd be very interested to be a little enlighted since high power electronics is not my expertise!
Well, the LIMs will be used to transport stuff on a rail in small "cars", the lims will be mounted on the rail with the reaction plate in the car. Fun stuff. The phase control solution was recommended by the lim manufacturer mainly for cost reasons I suppose. I have been given the mission to make use of the SSR solution that have been presented to me. To my frustration it doesn't seem to be as simple as just getting the firing angles and implement the thrust control in the micro. There almost seem to be a little "secret" in building a good firing circuit. Is this really so complicated? Or have I been too damaged in my relatively simple digital world? :)
Hi, you could try "power electronics handbook" by F.F.Mazda isbn 0-408-03004-6. It's quite dated insofar as it doesn't have much on inverters, but it has everything and more with regard to thyristor circuits.
Consider rectifying the AC, then drive the motor phases with a 6 FET bridge. To make the job easy, take a look at the TI TMS320 line of DSP chips. TMS320LF2402, 06, 07, etc.
They have motor control libaries for C and Asm. Creating a 3 phase control with one of these chips is very easy.
You have to constuct phases out of nothing. Or even worse you have to constuct phases out of a 3phase supply. As some said, you can use thyristors. These are semiconductors that conduct upon a trigger until the current reaches zero. It is doable and it is done. When you make a mistake with the trigger, the whole can blow up due to overcurrents. So make the semiconductors replaceable as simple a possible. The algorithm may be looked up in a textbook, I heard it once as a one year lesson with
2 hours per week. Since the usual solidstate AC relays have zero crossing detectors to switch on, you're somewhat limited. Better have a controlled rectifier to get DC first, then some Mosfet that allow microsecond resolution. Then you have to PWM as many phases as you need. A major project that keeps you busy for several months.
Well, I am under the impression that the procedure on this project will only involve chopping up the voltage supplied from the network, frequence will not be modified. The SSRs picked should obviously be able to switch on not only at zero crossing since the resolution without phase angle control won't be good enough.
Do you have a name on a good textbook on the subject or even better, some place to find this algorithm on the net? I'm rather axious to get started so I'd hate to have to wait a week or two on a book.
I'm rather curious thou, how will it be possible to blow this thing up just by triggering at the wrong moment? Shouldn't the SSRs (and the thyristors in them) be able to handle triggering at anytime?
An induction motor consists of a stator with windings and a short-circuited rotor. The magnetic field in the stator is made to rotate at the rate of the supplied AC (or a submultiple thereof). In three-phase motors the rotation occurs naturally because of the phase sequence. In single-phase motors the rotation is created with a starter capacitor or a split and shorted magnet pole.
The magnetic field induces a voltage in the rotor winding which quickly converts it to a hefty current (the winding is a short - see the rails along the rotor). The current in the rotor creates a magnetic field which makes the rotor follow the rotating field. The rotating rotor field creates a back-EMF in the stator windings which opposes the supply voltage and thus limits the total current in the motor.
There is always a small lag between the field rotation and the rotor rotation (e.g. 1500 rpm field rotation creates the usual 1450 rpm European motor). This lag keeps up the rotor current and thus also the torque the motor is able to deliver. When the rotor drops well out of synchronisation with the supply frequency, the back-EMF does not work anymore as supposed, so the motor quite quickly converts to a short-circuit. This is the cause of the hefty start-up spike of an induction motor.
You want to create an out-of-sync condition by starving the supply feed so far that the motor speed drops. With fan load it may sometimes be done without winding smoke, as the torque demand of a fan is roughly proportional to the square of the rotational speed, so it drops quite fast with dropping speed.
IMHO, the only working method to control induction motors is a variable-frequency drive with voltage tracking. The back-EMF changes with rotational speed, so it has to be compensated for.
There are thyristor and triac circuits which can directly convert the frequency of three-phase AC to another frequency, but they are far from simple. Google for 'cycloconverters'.
I think everyone has got the wrong end of the stick here. This is a linear induction motor. Not a rotary one. My understanding is that you drive the winding segments in sequence along a linear track. The velocity of the item you move along the track is dependent on the rate at which you drive the segments and the distance between the segments. Unless you want to achieve rail-gun like velocities you can probably energise one segment at at time using SSRs.
It is a linear induction motor, one of these flat ones you know? Not a revolving one, the stator and rotor is spread out. Currents induced in the reaction plate (rotor) by the stator travelling field creat a secondary magnetic field. It's the reaction between these two fields which produces the linear thrust...
By what I learned of Spehro Pefhany, phase angle control will only reduce the torque, not the speed of the motor, this will in turn make the slip increase, thus making it possible to control the speed of the reaction plate.
Yes, right on target, this is exactly what I'm trying to do. I'm going to put together the micro that is going to be put at each LIM motor. This micro is going to talk to an overlaying system to get the acceleration needed on its LIM when the item passes over and then fire the SSRs accordingly.
So my problem is to understand how to get the right firing sequences of these SSRs in order to get the required thrust.
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
Since thristors conduct until the current becomes zero, and an inductor has the current as integral of the voltage, a thyristor is triggered on the falling slope of the sine. Assuming an inductive load, there is current flowing well into the region where the sine is negative. When you trigger at the peak voltage (90 degrees) then you get almost 180degrees of conduction. Per phase. Doing 4 quadrant control, you also need a negative conducting branch.
Using MOSFETs for a frequency converter is limited to the smaller converters. It depends on the size of you system, ahem the power it has. I guess 10kW is doable with MOSFETs, whereas 1MW probably is not doable anymore.
Be able to synthesize the frequency from zero up when a lot of power is at play. The surge may not be bearable.
As to the blowing up of thyristors, yes with a suboptimal setup, where the line impedance is too low, especially when the recommended isolation transformer is omitted for cost reasons, and the thyristors are coosen for the operating load, then you can blow them up. No, fuses won't help. Fuses have a I^2*t characteristic, and they are too slow. In the shortcircuit case of a 10 Amp fuse, you can easily get 500Amps for a period or so. Have an isolation transformer for this.
Actually, building a frequency converter is more suited to sci.electronic.design than comp.arch.embedded
The basic physics stays the same, regardless whether the motor is flat or rolled into a cylinder. The speed is still tied to the excitation frequency. In a linear motor the field wanders a pole set each cycle.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.