Fieldlines.com: The Otherpower discussion board
Homebrewed Electricity => Wind => Topic started by: Adriaan Kragten on January 04, 2017, 06:23:01 AM

The title of this new public report KD 624 is: Ideas about a direct drive 46pole PMgenerator for the VIRYA6.5 windmill meant for driving the 2.2 kW asynchronous motor of a centrifugal pump. The report can be copied from my website: www.kdwindturbines.nl at the bottom of the menu KDreports.

far as I know, to get the most holding torque you need to short out these high pole count motors through a resistor equal to its internal impedance; otherwise they won't slow down the turbine.
my 30 pole motor had a 8.5 amp short circuit current, 97 volts at 440hz or about 1750 rpm, each phase was 1.1 ohm.
if i were to short out all three phases, 101 watts per phase would be dumped into the motor, about one half hp shaft power. this was at 1750 rpm. so the torque was about 1.5 foot pounds.
if i connected the motor through various resistors, i could stall a 1/2 hp induction motor at 800 rpm, or at least 3.3 foot pounds of torque, possibly as much as 5 foot pounds (the line current of the induction motor at stall was about 8 amps, full load amps was 6.6, so the torque at stall is unknown, but at least as much as 3.3 foot pounds)
I discovered this effect recently again, i found a 1/2 hp at 1250 rpm, 10 pole, 12 tooth concentrated pole motor (ferrite magnets) manufactured for the purpose of driving the fan in a relatively large heat pump. the motor had a built in inverter. If i shorted the motor phases out, the motor was difficult to turn of course, but once it was rotating above a few tens of rpm, the motor was effortless to turn.
As you propose making rather large high pole count motors, I suspect the torque cut off will be fairly sharp.
all this may not matter if you short the motor out after the rectifier.. i will have to try this again.

I have measured several 4pole PMgenerators for shortcircuit and you see that the torque is rising very fast at low rpm, then it has a maximum at about 400 rpm and then it slowly goes down at increasing rpm (see KD 78 figure 4). The whole Qn curve shifts to the right if the voltage is higher. For a 34pole or 46pole PMgenerator I expect about the same shape of the Qn curve but the point of maximum torque may lie at a much lower rpm as the frequency is much higher. So it might be that you first have to make shortcircuit over a certain resister to get a high torque level at moderate rotational speeds and when the rotor has slowed down enough, then you make real shortcircuit before the rectifier. If the generator is normally used in star, the star point has to be shortcircuited too to get the maximum torque level. For real shortcircuit, you will get the peak torque at the lowest rpm and this means that the rotor will turn only very slowly if shortcircuit is made.

Its also a function of the resistance of the copper windings. if the wires had no resistance, the motor would be effortless to turn, the magnetic flux would just be reflected by the coils into the air gap. so as the motor becomes larger, and proportionally the resistance of the copper decreases while the leakage inductance continues to increase, then the torque will drop off above a lower critical rpm.
This effect is mostly a result of geometry, the same effect causes high pole count induction motors (8 or 10 poles) to have poor power factor (on the order of 70%, when 2 pole and 4 pole machines can have 90% power factor)

Lets compare a 4pole PMgenerator with a 34pole PMgenerator and turn them with a rotational speed for which the frequency is 50 Hz for both. For a 4pole machine this 1500 rpm and for a 34pole machine this is 176.5 rpm. I have measured a 4pole PMgenerator made from a 2.5 kW asynchronous motor for 1500 rpm for different resistors as load and the performance of this generator is very good at that rotational speed (see KD 78 figure 28). The maximum electrical power is 3050 W for a load resistance of 30 Ohm. The efficiency is about 75 %. For a load resistance of 40 Ohm, the maximum power is 2950 W and the effciency is almost 80 %. So the maximum electrical power for a high efficiency is about a factor 1.2 higher than the nominal motor power of the original asynchronous motor.
If a 34pole PMgenerator is turned at the same frequency, I get the same variation of the magnetic flux in the stator so I don't see why this should result in a lower power factor than for a 4pole machine turning at the same frequency. If a 34pole generator turns at 1500 rpm you might get a much lower power factor but the VIRYA5 wind turbine is used at frequencies of about 50 Hz. The difference in between a PMgenerator and an asynchronous motor is that there are no big currents flowing in the armature and that there is no slip in between the rotating magnetic field of the armature and that of the stator. So what counts for an asynchronous motor may not necessarely count for a PMgenerator.
The only way to prove that a 34pole or a 46pole PMgenerator has an acceptable power level is to build one and measure it on an accurate test rig but I won't do this. In March I become 70 years old and I feel that I have built and tested enough. But sometimes I have ideas for which I think it is useful to make them public.

congratulations on your up coming birthday.
thanks for documenting all your experiments.

My point in bringing up induction motors is it is the leakage inductance that is affected by the geometry. For example it appears that linear induction motors just can't get higher than 60% pf. It is the leakage inductance that causes said low pf, the same leakage inductance that causes the torque to drop off. The higher the pole count, higher leakage flux.

Adriaan,
I don't know if you visit the Back Shed forum at all, but there have been a number of recent posts that made me think of you. Most of the discussion there centers around F& P smartdrive alternators, but there are some similar themes at times:
http://www.thebackshed.com/forum/forum_posts.asp?TID=9247&PN=1
http://www.thebackshed.com/forum/forum_posts.asp?TID=9239&PN=1
There are also a number of members here that contribute on Glenn's forum with the folks (largely) down under. Thought you might want to check it out.
Best, ~ks

Dear Kitestrings
I didn't know this forum. I have looked at it but although they also use a high pole number, it isn't the same kind of design. I think they are using the motor of a slow running washing machine and they rectify the current. I am using the housing of a big standard asynchronous motor and the three phases are directly coupled to the three phases of the motor a centrifugal pump .