I would like a second opinion on this alternator winding configuration

Thread Starter

l0vot

Joined Apr 29, 2013
107
I got a hold of a 5HP induction motor, I am rewinding it as a brushless alternator, I made a basic winding diagram that would result in a 9 pole alternator that would operate at full power @400RPM, and it may be able to be switched to 1200RPM or 3600RPM depending on how the windings are wired up (or 3 phase 3600,1800,1200,600RPM), but since i haven't done this before, I would like a second opinion on my design. I'm actually trying to drive the frequency up past what the motor was designed for , but that can of worms isn't relevant right now. (540HZ max, was shooting for 360HZ, but this core design does not permit that with single phase without leaving some of the slots unused)

Same coloured lines are pairs for single phase operation.

I will be using a binary capacitor bank to find the right cap size for this induction alternator, so don't worry about that. This is an experiment in order to learn, if the results are good enough, then it will become something more useful, and possibly a lot more valuable, but the primary objective is to learn what I can do with iron core tech, particularly since the information I'm looking for isn't that common since anyone dealing with anything besides mains frequency uses ferrite cores, and then pumps the frequency up to so many KHz/MHz.
 

Attachments

Thread Starter

l0vot

Joined Apr 29, 2013
107
i didn't get any replies, so i put test coils in the motor, the test coils were 3 winds of 12AWG solid core wire in each of the 18 "slots" in the picture I posted (54 winds total). it runs as a motor (hand start required, obviously), I will need to test it to see if the speed is right, and the volts/wind @60HZ so i have some idea of how it's supposed to be acting in generator mode.
 

shortbus

Joined Sep 30, 2009
10,045
i didn't get any replies,
May be because of the file you attached. I know for me I wasn't able to open it. Many don't use Explore and it won't open in Firefox which is what a lot or even most use here. Maybe change it to another form, but not a PDF, because many won't open them from people they don't know.
 

Thread Starter

l0vot

Joined Apr 29, 2013
107
May be because of the file you attached. I know for me I wasn't able to open it. Many don't use Explore and it won't open in Firefox which is what a lot or even most use here. Maybe change it to another form, but not a PDF, because many won't open them from people they don't know.
It's an SVG, most browsers, including Firefox support them. The problem appears to be FF will try to download SVGs instead of displaying them when they are in a link like that, but they display just fine on a web page, or if they are dragged and dropped into FF no clue why. I posted as SVG because it's the original format, and it's very compact while a PNG or JPG would have been much larger, even at a lower resolution.
 

Thread Starter

l0vot

Joined Apr 29, 2013
107
Mainly I wasn't sure if having the coils interlaced that much would be a problem, or what impact it would have on performance. This is my first time winding a motor/alternator after all.
 

shortbus

Joined Sep 30, 2009
10,045
Mainly I wasn't sure if having the coils interlaced that much would be a problem,
I'm no expert on this, but from any wind scheme I've ever seen they don't do it as discontinuous as you are showing. You seem to be jumping all over the place. Maybe show one winding so it's actually easy to trace. Usually the windings a overlapped on over the next, not randomly.

I have a stalled project that was to make a SRM (switched reluctance motor) from an existing 5HP tree phase motor stator. and came up with the same problem your having that there are not the correct pole numbers to work with.
 

Thread Starter

l0vot

Joined Apr 29, 2013
107
I'm no expert on this, but from any wind scheme I've ever seen they don't do it as discontinuous as you are showing. You seem to be jumping all over the place. Maybe show one winding so it's actually easy to trace. Usually the windings a overlapped on over the next, not randomly.

I have a stalled project that was to make a SRM (switched reluctance motor) from an existing 5HP tree phase motor stator. and came up with the same problem your having that there are not the correct pole numbers to work with.
the coils are in a very organized pattern, if you go clockwise, every other coil is a different pair that's offset by exactly 40 degrees, once you do that up to 9, then all of the pairs have been found. The idea is to have the magnetic field cut across a very large section of the rotor. It hurt my head to look at when I came across that pattern by throwing ideas at the wall to see what worked. It does look strange for sure.

Good luck with the reluctance motor, I have a 3phase one out of a washing machine, but no way to power/control it.
 

Thread Starter

l0vot

Joined Apr 29, 2013
107
Hi again maybe this will help you on your journey to get what you want. http://www.electrical4u.com/armature-winding-of-alternator/
only problem is that's for a brushed alternator armature, an induction motor and induction alternator are wound exactly the same way, except in the case of single phase induction alternators no start coils are necessary. My variac came in and it seems to saturate above 7VAC, and draws 20A@7VAC which is lower than I expected to get with 54 winds but w/e.
 

shortbus

Joined Sep 30, 2009
10,045
only problem is that's for a brushed alternator armature,
I don't think so. In an alternator the armature is not the same as in a generator. The alternator armature is the same as the stator in a AC motor, wound on iron surrounding the rotor.
 

Thread Starter

l0vot

Joined Apr 29, 2013
107
it looks like the design i was trying to follow was probably a 2 layer design (from the page you linked me to), and mine has WAY more than 2 layers. Due to the extremely poor performance I'm going to try rewinding it with a 2 layer design. The benefit of the 2 layer winding was the ability to completely utilize the whole core, but it looks like more than 2 layers is NOT a good idea. The current at no load is just too high, that motor may be big, but it shouldn't be pulling 140W at no load (weather it's spinning or not!) and have nearly no torque, even with test windings.
 

shortbus

Joined Sep 30, 2009
10,045
Like said earlier I'm no expert and haven't studied alternator design but that looks more like a workable one. If this is new ground, to you at least, you just have to experiment. Have you Googled for college thesis on what your doing? that where I usually start to look. Please keep showing results here of your experiments.
 

Thread Starter

l0vot

Joined Apr 29, 2013
107
I finished rewinding the stator core, and got the EXACT same results. Literally no difference. Hook it up to the variac, get to 7V, and it's drawing 20A again, but i tried logging the amps at 0RPM, and full speed, and I noticed the amperage did change a little bit, it dropped to 19A at full speed. Still no torque. At least the windings are far neater, and consume less wire.

I got to thinking, "what if it just needs more power?", so I hooked it directly to my AC arc welder, lo and behold I started getting actual torque out of this thing, the power draw was pretty high, but it wasn't saturating.

Industrial induction motors get around 95% efficiency at max load, less load their efficiency drops, more load and their efficiency drops, which is pretty weird, they are basically rotating transformers, and normal transformers made of the same materials can get 99% efficiency under certain loads, and consume little power when idling. I was wondering, where is that 5% going? it's not going to hysteresis or eddy current losses because they are made of the same exact stuff transformers are made of, they don't have this problem because the metal is specifically designed to have low core losses. Then it hit me: all induction motors have a pretty large air gap, they have to, the rotor and stator are not supposed to touch under any circumstances, so there has to be a gap. The motors work by having large amounts of flux cut through the rotor, but to get to the rotor, the flux has to fill the air gap first, and air is not a very good conductor of magnetic fields.

The problem seems to be the air gap coupled with the test windings, the windings don't have enough copper to fill the air gap efficiently, so they have to draw a ton of power just to magnetize everything, proper field windings would have a much higher inductance, and thus be able to fill the air gap a lot more efficiently.

The only thing to do now is hook the alternator up to the engine, and see if driving up the speed and frequency helps anything.
 
Top