Foi este post abaixo, explicando como religar um motor de indução AC para conseguir uma potência de pelo menos 6x maior do que a original.
O post é deste forum aqui:
from: http://forums.aeva.asn.au/forums/forum_posts.asp?TID=1237&PN=9&title=changing-an-induction-motor-voltage
E o autor é este cara aqui:
coulomb
Ok, I'll have a go at this.
1. For ages, we've known that it is possible to push an induction motor past its nominal power point, simply by loading it more than its nominal power. The controller can increase the mechanical load by increasing the "slip". We call this "overcurrenting" the motor, because it requires more current than the windings can continuously handle. This is OK as long as there is sufficient cooling, and the motor isn't overcurrented for more than short bursts, e.g. for acceleration in an EV. This is also OK for short hill climbing, but not long hills or mountains; they can overload the motor for long periods of time and cause overheating. This is not the subject of this thread; I put it here for context.
2. The other way of getting more power out of a motor is by increasing the voltage. But most motors are 415 V already; without more than a 600 V pack and a very expensive controller we can't increase the voltage, and it's probably not safe to go beyond 440 VAC since the motor insulation isn't designed for it. But we can reconfigure the motor so that its nominalvoltage is less than 415 V, and put up to 415 VAC on it anyway. We can reconfigure the motor's nominal voltage in several ways:
2a. Rewinding. That basically means chop out the old wire, and start again. This is beyond most people's garages. A figure of AU$800 has been mentioned for this process.
2b. Reconnecting. If we are lucky enough to have the windings split into two or more pieces, and there are 9 or more connectors in the terminal box, then we can rearrange the connections to achieve lower nominal voltage. Some American motors are like this, so you might be able to configure for 220 VAC as well as 440 VAC, for example. One special case: some Australian motors can be found (or specified if you are buying new) with 415 V star windings. This is the "S" option for ABB motors. In that case, with the standard 6 wire terminal box, you can rejumper your motor to be 230 V delta. This is a fixed voltage reduction of sqrt(3) = 1.732.
2c. Rewiring. That's what this thread is about. Some, but as we have seen not all, motors have windings in series that can fairly easily (in the home garage) be wired in parallel. Depending on how many similar windings are in series, we might even have a choice as to what factor the motor becomes undervoltaged: half, third, perhaps even smaller fractions. The big advantage of this method over 2a is that it's cheap and easy, but if the motor isn't suitable for rewiring, then 2a is the only real option for changing the motor's nominal voltage. 2b is even better, since we don't even have to open the motor, but few motors available to Australian converters are likely to have this option.
3. So changing the nominal voltage of the motor is a prerequisite to overvoltaging. But it has some important effects. Firstly, the nominal power of the motor is increased. This is because when we increase the voltage beyond the nominal voltage, the speed will increase beyond the nominal speed, but the motor current remains largely unchanged. The bulk of the motor current depends on the load torque, so if we keep the load torque the same and double the voltage, we double the speed and hence double the power.
4. But we can't leave the motor at double voltage and nominal torque continuously, because the iron losses increase dramatically with speed. That's why I said "largely unchanged" in point 3 above. But we can reduce the continuous torque by a certain amount to compensate for this. Weber calculated that the derated increase in continuous power is about the 0.7th power of the voltage increase. 2^0.7 ~= 1.6, which is 80% of 2, so we can derate the continuous torque by 20%.
With double the speed and 80% of the torque, we have 160% of the original nominal power, and the motor can run at this power level continuously. Note: this is theory right now, yet to be tested.
5. Let's consider an example to make this clear. In a Barina class vehicle, it might take 16 kW to maintain 100 km/h on the level. Without rewiring, you'd need an 18.5 kW nominal motor to be able to drive at freeway speeds for as long as your battery lasts. You'd probably get away with a 15 kW motor unless you had a really big battery pack, as the thermal time constant for a motor is about an hour, so it might take an hour for the 15 kW motor to overheat, by which time you've travelled 100 km and drained your pack. Now let's consider an 11 kW nominal motor, 415 V delta (can't get the easy 1.732 voltage reduction). Let's say its breakdown torque is a little over 3x nominal, so this is a 33 kW peak power motor. Let's say this motor is suitable for voltage halving. Let's also say it's a 4-pole motor, so its nominal speed is a little under 1500 RPM; for simplicity I'll call this 1500 RPM. By rewiring, you end up with a 208V 11 kW 1500 RPM motor. So we can tell the controller that this is now a 160% * 11 kW = 17.5 kW continuous motor, with a nominal speed of 3000 RPM. We can still overcurrent this motor by 3x (note: this is theory again). So at peak we should be able to get 3x original nominal torque at 2x original nominal speed, for an overall 6x peak power, or 11 * 3 * 2 = 66 kW. That's 32% more than the peak power of a Barina class ICE.
But there's more: at 1500 RPM and below, we know that it can take 100% of its original continuous torque. The original motor's nominal torque dropped away at 1/f, so at 3000 RPM, the original motor had half nominal torque. After rewiring, it has 3/4 nominal torque at 3000 RPM.
In summary: by rewiring for half voltage, we've gone from an 11 kW continuous 33 kW peak motor, to a 17.5 kW continuous, 66 kW peak motor. You might notice that I've chosen the numbers so that a rather weak motor, not capable of continuous freeway driving, and only 2/3 of the original ICE's peak power becomes capable of continuous freeway driving, and more peak power than the original ICE. That's an impressive improvement. We can only get away with this because and assuming that EV power demand is very peaky, with the peaks of short duration.
| How about considering what a 5hp ac motor which would normally be too small to sufficiently move a light vehicle be applying the above parallel vs series process? |
Assuming by light you mean Barina class, you have the right idea, but the effect is not quite as dramatic as that.
I'm fairly certain that the best voltage reduction we can hope for in EV sized induction motors would be half. But let's suppose we found a 5 kW motor that we could quarter the voltage of. From Weber's equation, we can expect about a 4^0.7 ~= 2.6x improvement in continuous power, and a 4x * 3x = 12x (!) improvement in peak power. So that's 13 kW continuous, and 60 kW peak. The peak power is at 4x the original nominal speed, so it should be a 4-pole motor, so that peak power is a little under 6000 RPM. (Extra balancing will likely be required, and possibly also better bearings.) Whether this would be satisfactory or not depends on your needs. 13 kW will not allow continuous freeway speeds, but it might allow continuous 80 km/h driving, and certainly continuous stop/start traffic. So that would be great; but: can we quarter the nominal voltage?
Edits: use Weber's equation rather than chart; redid examples with this.
Continuing the big summary post above. Now we know that rewiring for lower voltage is good; what are the ways that this can be done?
6. Many EV-sized vehicles come with "concentric" windings. I'm pretty sure that this is an attempt to distribute the magnetic flux in a more sinusoidal way. Unfortunately for rewirers, this means that a lot of opportunity for rewiring is removed. In a two pole motor, these windings can span 18, 16, and 14 slots in a typical 36-slot motor. For 4-pole, the numbers are typically 12, 10, and 8 slots. The resistance of the windings varies according to how many slots are spanned, so paralleling these concentric windings would probably cause problems with circulating currents. This is not yet confirmed. If it was safe to do, and could readily be done, this would allow a 1/3 voltage rewire. However, when I attempted to do this, I found that it was very difficult to find the places where the wire changed from one pair of slots to the next (the "crossover wires"). This problem is made much worse by the way the windings are arranged; generally, one pair of windings is very much behind the other overlapping pairs, and this makes it virtually impossible to find all the crossover wires.
7. Despite that, most EV sized induction motors have two real poles. This is obvious for 2-pole motors, and at least some 4-pole motors use "consequent poles", meaning that the direction that the windings are connected forces these "virtual" or "consequent" poles to form. In the case of my 7.5 kW motor, these poles were actually connected in parallel. No voltage reducing rewiring is possible in cases like that. We need to find windings that are in series, and connect them in parallel instead. The paralleling in this case was obvious from the "Y" connections around the outside of the windings.
In the case of Electrocycle's 11kW motor, we didn't see any such Y points, so we assume that the poles are wired in series. In this case, it should be possible to do a fairly easy half voltage rewiring, and the varnished cambric sleeving should pretty much highlight the places were wires need to be cut and reconnected.
8. To prove that this works, we need to know the load we can put on a motor, both to see what load it can handle continuously without overheating, and what peak load it can handle. Two main ways of doing this are setting up a dynamometer (like ACmotor is doing), and driving an actual vehicle and logging as much data as possible. The latter has the problem that when you do get to the point where you've overloaded the motor, you might cause a traffic hazard, and/or have to pull over to the side of the road and let things cool off for a long time. The problem with the former is that things like thermal constants may vary greatly from 2kW class motors to 11 kW class motors.
Nenhum comentário:
Postar um comentário