Help with wind turbine magnet replacement

[llaind]

Hi, I'm new here and i signed up because i need help with something.
I've got a couple of istabreeze 2kw wind turbines that have failed and I'm replacing the magnets on the rotor, since the originals came off and crunched.

I had a go at it already and failed. I placed all the magnets with the alternating polarity around the rotor. but there was no resistance when turning and didn't produce any power. i decided to go for n55 magnets and the north and south is on the long face.

i can't figure out the orientation of the magnets and istabreeze won't reply to any questions about it.

help needed if anyone can.

the rotor has an uneven space and doesn't have an opposing placement 180deg.

here are the pics.

[DamonHD]

Welcome!

I can't directly help you, but I'm sure that someone will be along shortly who can...  B^>

Rgds

Damon

[llaind]

Welcome!

I can't directly help you, but I'm sure that someone will be along shortly who can...  B^>

Rgds

Damon

Thankyou, hopefully someone will know exactly what to do here.

[electrondady1]

   its handy to have a magnet marked north and south to help orientate the mags as your laying them out .
assuming the coils are full length, you need  to place a second magnet with the same pole up, end to end with that magnet
 then place two more in the adjacent slot with the opposite pole up , then so on and so on
maybe some glue is involved .
it does look like one of the ridges is wider than the others.
any chance of a photo of the coils?

[llaind]

   its handy to have a magnet marked north and south to help orientate the mags as your laying them out .
assuming the coils are full length, you need  to place a second magnet with the same pole up, end to end with that magnet
 then place two more in the adjacent slot with the opposite pole up , then so on and so on
maybe some glue is involved .
it does look like one of the ridges is wider than the others.
any chance of a photo of the coils?

Thanks for the input. However, already tried that. I glued the magnets so pairs stuck together end to end then placed them alternating north up and north down around the rotor. gluing them with appropriate adhesive. After doing this, i got no power output and there was no resistance when turning the motor like there used to be, it simply spun freely.  millivolt output of power. it's a 3 phase 48v ac motor and yes the coils are full length of the rotor.  the only difference is i'm using n55 magnets, where i suspect the originals were n42/45. Ill upload a photo of the coils.

[Adriaan Kragten]

In the third photo I have counted 12 grooves. If you have 12 grooves in the armature, the armature is normally a 12-pole armature and in this case you have six north and six south poles lying alternated against each other. The pitch angle in between the grooves isn't exactly 30° because you can see that there are eleven narrow bridges in between the grooves and one wide bridge. I don't think that this is a mistake but that it is done to minimize the fluctuation on the sticking torque. I assume that the stator has 36 grooves and that the winding is a 3-phase winding. If the stator has 36 slots, it will therefore have 36 poles. If the angle in between the armature grooves would be exactly 30°, you will get a preference position every 10° when an armature pole is just opposite a stator pole. So you will get 36 preference positions per revolution. But because of that one wide bridge, the fluctuation will be flattened a lot.

If you have glued the magnets such that you have six alternating north and south poles and if you get no voltage out of the winding, it might be that the generator must have not a 12-pole but a 6-pole or a 4-pole armature. For a 6-pole armature you have two rows of north poles besides each other. For a 4-pole armature you have three rows of north poles besides each other. So you have totally three options. Which option is valid, can be found by looking at the winding.

Assume that the armature has 12 poles. So the average armature pole angle is 30°. This means that the optimum angle in between the left and the right leg of a coil is 30° too. So if the grooves are numbered 1 - 36, you have a coil U1 in grooves 1 and 4, a coil V1 in grooves 5 and 8, a coil W1 in grooves 9 and 12, a coil U2 in grooves 13 and 16 and so on until the first layer is complete. The second layer is laid in the remaining 18 grooves.

Assume that the armature has 6 poles. So the average armature pole angle is 60°. This means that the optimum angle in between the left and the right leg of a coils is 60° too. But this isn't possible for a stator with 36 grooves. So one coil has an angle of 50° and around this coil there is a second coil with an angle of 70°. Another difference is that a 6-pole armature can't have a 2-layers winding but it has a 3-layers winding. The first layer contains coil U1 in slots 2 and 7, coil U2 in slots 1 and 8, coil U3 in slot 14 and 19, coil U4 in slot 13 and 20, coil U5 in slot 26 and 31 and coil U6 in slot 25 and 32. The second layer contains six coils V in slots 5, 6, 11, 12, 17, 18, 23, 24, 29, 30, 35 an 36. The second layer contains six coils W in slots 33, 34, 3, 4, 9, 10, 15, 16, 21, 22, 27 and 28.

Assume that the armature has 4 poles. So the average armature pole angle is 90°. This means that the optimum angle in between the left and the right leg of a coils is 90° too. But this isn't possible for a stator with 36 grooves. So one coil has an angle of 70°, around this coil there is a second coil with an angle of 90° and around this second coil there is a third coil with a coil angle of 110°.  A 4-pole armature can have a 2-layers winding. The first layer contains coil U1 in slots 3 and 10, coil U2 in slots 2 and 11, coil U3 in slot 1 and 12, coil V1 in slot 15 and 22, coil V2 in slot 14 and 23 and coil V3 in slot 13 and 24, coil W1 in slots 27 and 34, coil W2 in slots 26 and 35 and coil W3 in slots 25 and 36. The second layer contains three coils U in slots 19, 20, 21, 28, 29 and 30, three coils V in slots 31, 32, 33, 4, 5 and 6 and three coils W in slots 7, 8, 9, 16, 17 and 18.

So you have to study the winding carefully to find out if this is the winding of a 12-pole, a 6-pole of a 4-pole armature. If the winding would be the winding of a 6-pole or a 4-pole armature and if you give the armature 12 poles, the winding is completely wrong for a 12-pole armature and you will get no voltage out of it. I am afraid that this is the cause of your problem.

[Mary B]

   its handy to have a magnet marked north and south to help orientate the mags as your laying them out .
assuming the coils are full length, you need  to place a second magnet with the same pole up, end to end with that magnet
 then place two more in the adjacent slot with the opposite pole up , then so on and so on
maybe some glue is involved .
it does look like one of the ridges is wider than the others.
any chance of a photo of the coils?

Thanks for the input. However, already tried that. I glued the magnets so pairs stuck together end to end then placed them alternating north up and north down around the rotor. gluing them with appropriate adhesive. After doing this, i got no power output and there was no resistance when turning the motor like there used to be, it simply spun freely.  millivolt output of power. it's a 3 phase 48v ac motor and yes the coils are full length of the rotor.  the only difference is i'm using n55 magnets, where i suspect the originals were n42/45. Ill upload a photo of the coils.

Test your windings with an ohmmeter to see if you have an open from the magnets bashing around... and make sure to use the AC setting on the meter to check output from the windings! With no load you won't feel much resistance. You could short each set of windings to see if it gives you sticking torque. It will be very hard to turn into a short.

[Adriaan Kragten]

Another point is that the manufacturer has made a mistake such that the original magnets came loose. You should not make the same mistake. The following mistakes may have been made.
1) The grooves are made on a milling machine and the cutter is greased by oil. So this oil has to be removed carefully using a solvent which removes all oil. This can't be done in one move. So one must do it several times with a clean cloth.
2) One can have used epoxy glue. The disadvantage of epoxy glue is that the strength of the glue decreases strongly at increasing temperature. The strength is almost completely lost at 100 °C. No heat is generated in the armature but heat is generated in the stator due to iron and copper losses. If the stator isn't cooled well enough, it can become rather hot at hot days and the armature will become hot too because of inwards radiation. Anaerobe glue can be used up to much higher temperatures of about 200 °C and this glue type must be used if the armature temperature can become too high. But anaerobe glue hardens only in a narrow gap.
3) Some solvent may have remained and react to the used glue. I don't know which solvent has to be used for epoxy glue but only alcohol or acetone can be used for anaerobe glue.
4) If water can penetrate in the generator, the iron of the armature can rust and the rust may enter the glue gap and push the magnet outside the groove. Water can penetrate even in closed housings of asynchronous motors due to condensation at low temperatures if the seals start leaking after some time.
5) There must be a gap of at least 0.2 mm in between the outside of the magnets and the inside of the stator. It might have been that a certain grove was too narrow (or one magnet was too wide) making that a magnet could not touch the bottom of the groove. Another cause may have been that an iron particle my have prevented that a magnets could touch the bottom of the groove. So the gap for one magnet may have been very small. If the bearings start wearing, you get some bearing play and this play may have caused that that one magnet has touched the stator.
6) The magnets are pulled to the outside because of the centrifugal force which results in a certain pulling stress in the glue. However, if you would calculated the centrifugal force for the rather low maximum rotational speed which happens in a direct drive PM-generator of a wind turbine, you will find that this centrifugal force is very low and that the stress in the glue is very low too. So if the glue is in good condition, the centrifugal force will never make that the magnets come loose. In chapter 5 of my free public report KD 718 I give the calculation of the pulling stress in the glue for my new 16-pole PM-generator with magnets with a similar orientation.

[Mary B]

after 30 years pf running electronics outdoors in very harsh weather and exposed to extreme winds I can tell you this. NO sealed container is waterproof. Humidity will get in, condensation will happen, water will pool. I always make sure to have a weep hole where excess water can drain. Temps here range form -31f to 110f... winds to 95mph...

[llaind]

In the third photo I have counted 12 grooves. If you have 12 grooves in the armature, the armature is normally a 12-pole armature and in this case you have six north and six south poles lying alternated against each other. The pitch angle in between the grooves isn't exactly 30° because you can see that there are eleven narrow bridges in between the grooves and one wide bridge. I don't think that this is a mistake but that it is done to minimize the fluctuation on the sticking torque. I assume that the stator has 36 grooves and that the winding is a 3-phase winding. If the stator has 36 slots, it will therefore have 36 poles. If the angle in between the armature grooves would be exactly 30°, you will get a preference position every 10° when an armature pole is just opposite a stator pole. So you will get 36 preference positions per revolution. But because of that one wide bridge, the fluctuation will be flattened a lot.

If you have glued the magnets such that you have six alternating north and south poles and if you get no voltage out of the winding, it might be that the generator must have not a 12-pole but a 6-pole or a 4-pole armature. For a 6-pole armature you have two rows of north poles besides each other. For a 4-pole armature you have three rows of north poles besides each other. So you have totally three options. Which option is valid, can be found by looking at the winding.

Assume that the armature has 12 poles. So the average armature pole angle is 30°. This means that the optimum angle in between the left and the right leg of a coil is 30° too. So if the grooves are numbered 1 - 36, you have a coil U1 in grooves 1 and 4, a coil V1 in grooves 5 and 8, a coil W1 in grooves 9 and 12, a coil U2 in grooves 13 and 16 and so on until the first layer is complete. The second layer is laid in the remaining 18 grooves.

Assume that the armature has 6 poles. So the average armature pole angle is 60°. This means that the optimum angle in between the left and the right leg of a coils is 60° too. But this isn't possible for a stator with 36 grooves. So one coil has an angle of 50° and around this coil there is a second coil with an angle of 70°. Another difference is that a 6-pole armature can't have a 2-layers winding but it has a 3-layers winding. The first layer contains coil U1 in slots 2 and 7, coil U2 in slots 1 and 8, coil U3 in slot 14 and 19, coil U4 in slot 13 and 20, coil U5 in slot 26 and 31 and coil U6 in slot 25 and 32. The second layer contains six coils V in slots 5, 6, 11, 12, 17, 18, 23, 24, 29, 30, 35 an 36. The second layer contains six coils W in slots 33, 34, 3, 4, 9, 10, 15, 16, 21, 22, 27 and 28.

Assume that the armature has 4 poles. So the average armature pole angle is 90°. This means that the optimum angle in between the left and the right leg of a coils is 90° too. But this isn't possible for a stator with 36 grooves. So one coil has an angle of 70°, around this coil there is a second coil with an angle of 90° and around this second coil there is a third coil with a coil angle of 110°.  A 4-pole armature can have a 2-layers winding. The first layer contains coil U1 in slots 3 and 10, coil U2 in slots 2 and 11, coil U3 in slot 1 and 12, coil V1 in slot 15 and 22, coil V2 in slot 14 and 23 and coil V3 in slot 13 and 24, coil W1 in slots 27 and 34, coil W2 in slots 26 and 35 and coil W3 in slots 25 and 36. The second layer contains three coils U in slots 19, 20, 21, 28, 29 and 30, three coils V in slots 31, 32, 33, 4, 5 and 6 and three coils W in slots 7, 8, 9, 16, 17 and 18.

So you have to study the winding carefully to find out if this is the winding of a 12-pole, a 6-pole of a 4-pole armature. If the winding would be the winding of a 6-pole or a 4-pole armature and if you give the armature 12 poles, the winding is completely wrong for a 12-pole armature and you will get no voltage out of it. I am afraid that this is the cause of your problem.

Thank you for the very detailed reply, took a few reads to fully get it. but, i see now how there are many configurations of the magnets possible. My initial research led me to believe it was going to be a simple alternating sequence. I've taken a few photographs of the stator, both stators i have. one which i won't be using due to damage. They were both sold as a 2kw unit, but the wiring as you can see is different and overall one has more copper than the other. Both 36 grooves.

[llaind]

after 30 years pf running electronics outdoors in very harsh weather and exposed to extreme winds I can tell you this. NO sealed container is waterproof. Humidity will get in, condensation will happen, water will pool. I always make sure to have a weep hole where excess water can drain. Temps here range form -31f to 110f... winds to 95mph...

thanks for the input, i;ve checked all three phase wires and they all have similar resistance. the moisture does get in and in a short time. these units have had a few bearing replacements and the swivel electric contact too.

here is the new model, need to make a mounting plate and tail for it with some kind of spring auto-break.



the newer motors now have much thicker plate for the blades. as the first incarnation would bend in higher winds and cause the bolts one the blades to contact the bolts on the motor casing. the shaft also has a larger diameter and is tapered too now.

[Adriaan Kragten]

In the third photo I have counted 12 grooves. If you have 12 grooves in the armature, the armature is normally a 12-pole armature and in this case you have six north and six south poles lying alternated against each other. The pitch angle in between the grooves isn't exactly 30° because you can see that there are eleven narrow bridges in between the grooves and one wide bridge. I don't think that this is a mistake but that it is done to minimize the fluctuation on the sticking torque. I assume that the stator has 36 grooves and that the winding is a 3-phase winding. If the stator has 36 slots, it will therefore have 36 poles. If the angle in between the armature grooves would be exactly 30°, you will get a preference position every 10° when an armature pole is just opposite a stator pole. So you will get 36 preference positions per revolution. But because of that one wide bridge, the fluctuation will be flattened a lot.

If you have glued the magnets such that you have six alternating north and south poles and if you get no voltage out of the winding, it might be that the generator must have not a 12-pole but a 6-pole or a 4-pole armature. For a 6-pole armature you have two rows of north poles besides each other. For a 4-pole armature you have three rows of north poles besides each other. So you have totally three options. Which option is valid, can be found by looking at the winding.

Assume that the armature has 12 poles. So the average armature pole angle is 30°. This means that the optimum angle in between the left and the right leg of a coil is 30° too. So if the grooves are numbered 1 - 36, you have a coil U1 in grooves 1 and 4, a coil V1 in grooves 5 and 8, a coil W1 in grooves 9 and 12, a coil U2 in grooves 13 and 16 and so on until the first layer is complete. The second layer is laid in the remaining 18 grooves.

Assume that the armature has 6 poles. So the average armature pole angle is 60°. This means that the optimum angle in between the left and the right leg of a coils is 60° too. But this isn't possible for a stator with 36 grooves. So one coil has an angle of 50° and around this coil there is a second coil with an angle of 70°. Another difference is that a 6-pole armature can't have a 2-layers winding but it has a 3-layers winding. The first layer contains coil U1 in slots 2 and 7, coil U2 in slots 1 and 8, coil U3 in slot 14 and 19, coil U4 in slot 13 and 20, coil U5 in slot 26 and 31 and coil U6 in slot 25 and 32. The second layer contains six coils V in slots 5, 6, 11, 12, 17, 18, 23, 24, 29, 30, 35 an 36. The second layer contains six coils W in slots 33, 34, 3, 4, 9, 10, 15, 16, 21, 22, 27 and 28.

Assume that the armature has 4 poles. So the average armature pole angle is 90°. This means that the optimum angle in between the left and the right leg of a coils is 90° too. But this isn't possible for a stator with 36 grooves. So one coil has an angle of 70°, around this coil there is a second coil with an angle of 90° and around this second coil there is a third coil with a coil angle of 110°.  A 4-pole armature can have a 2-layers winding. The first layer contains coil U1 in slots 3 and 10, coil U2 in slots 2 and 11, coil U3 in slot 1 and 12, coil V1 in slot 15 and 22, coil V2 in slot 14 and 23 and coil V3 in slot 13 and 24, coil W1 in slots 27 and 34, coil W2 in slots 26 and 35 and coil W3 in slots 25 and 36. The second layer contains three coils U in slots 19, 20, 21, 28, 29 and 30, three coils V in slots 31, 32, 33, 4, 5 and 6 and three coils W in slots 7, 8, 9, 16, 17 and 18.

So you have to study the winding carefully to find out if this is the winding of a 12-pole, a 6-pole of a 4-pole armature. If the winding would be the winding of a 6-pole or a 4-pole armature and if you give the armature 12 poles, the winding is completely wrong for a 12-pole armature and you will get no voltage out of it. I am afraid that this is the cause of your problem.

Thank you for the very detailed reply, took a few reads to fully get it. but, i see now how there are many configurations of the magnets possible. My initial research led me to believe it was going to be a simple alternating sequence. I've taken a few photographs of the stator, both stators i have. one which i won't be using due to damage. They were both sold as a 2kw unit, but the wiring as you can see is different and overall one has more copper than the other. Both 36 grooves.

(Attachment Link) (Attachment Link)

In the top photo you can see that in between the left and the right leg of a coil there are two grooves. So the angle in between the left and the right leg of a coil is 30° if the stator has 36 grooves. This means that this is the winding of a 12-pole armature. However, the winding is very unusual as the left leg and a right leg of two adjacent coils are placed in the same groove. So one leg of a coil takes only half of the available space.

This isn't a usual 2-layers winding with coils of three different phases in one layer but a 3-layers winding with all coils of one phase in one layer. But one layer doesn't have six right hand wound coils but six right hand and six left hand wound coils. For the same number of wires in a groove, this gives the same voltage as for only six right hand wound coils but the winding is more chaotic as a normal 3-layers winding with only six coils. A slight advantage is that the coil heads are somewhat smaller. A disadvantage of using left hand and right hand wound coils is that a very orderly 2-layers winding with adjacent coils of phase U, V and W in one layer is now not possible. The coil pattern of only the third layer can be followed well if you look to the winding from the center of the stator but the other two layers must have the same construction.

As this is the winding of a 12-pole generator, the magnets in the armature must be positioned such that a row with north poles is followed by a row with south poles. But that is what you have done. So the fact that you measure no voltage if the armature is rotating is not caused by a wrong number of armature poles. Are you sure that the two magnets in one groove have both the same pole to the outside? When I make a generator, I first pile all needed magnets together and I place arrows in the same direction at all four small sides of the magnets. So you can always see if the magnets have been mounted correctly. I would mount first all magnets with the north poles to the outside and in the six odd grooves and wait until the glue is hardened. Next I would place the remaining magnets with the south poles to the outside in the remaining six even grooves.

The second photo is clearly from another winding as now I see only the legs of one coil in a groove. But if you follow the coils, you can see that this is also a 3-layers winding as for a 2-layers winding the right leg of coil U1 should be in the next groove as the left leg of coil V1.   

[SparWeb]

llaind,
Welcome to Fieldlines!

I am looking at your photos from earlier today (reply #9) and the wire looks discoloured and possibly damaged.  The individual wires are all separate, not bonded or tied together which makes me very worried.  This could be damage from the broken magnet, or the result of a winding process in the generator's past that did not include the (necessary) tying of the coil groups to prevent vibration.  Forces on the wire will bend them back and forth when they are not supported and the result is broken wire.  The heat damage also could have bad effects already.  If I were you I would want to test the stator for "hi-pot" or "high potential" insulation resistance and continuity.

The second stator looks much better than the first one - I hope you are keeping that one.

[Adriaan Kragten]

If you compare your 12-pole generator with my 16-pole generator as described in my recent public report KD 718, you can see that there are some similarities but also some differences. The similarities are that the magnets are mounted radial, so with the wide side to the outside, in grooves which are in parallel to the armature axis and that a slight shift of the armature pole angle with respect to the stator pole angle is used to flatten the peak on the sticking torque.

An important difference is that I use only magnets for half the poles and that the other poles are formed by the remaining material of the armature. This has two advantages.

The first advantage is that less magnets are needed but that now a magnetic loop if flowing only through one magnet. In chapter 3 of KD 718, I have calculated that the stator is saturated for the chosen magnet thickness and air gaps. So using only one magnet in a magnetic loop is no problem if the magnets are thick enough.

The second advantage is that the air gap is smallest for the poles which are not formed by the magnets and the magnets are therefore not touching the stator when the armature is mounted in the stator. The armature will always be pulled against one side of the stator during mounting and pushing the armature at the correct place requires a large force. The armature is only centered when the two bearing covers are pulled against the stator housing. If all poles are formed by magnets, certain magnets will scratch along the sharp stator poles and this may damage the nickel coating of those magnets. Uncovered neodymium is very sensible to corrosion and if some water has penetrated in the housing, corrosion will be accelerated. This corrosion then will spread to the glue gap and finally the rust will pull the magnet out of the groove. This may be the real cause of your problem. For the new armature, I advice to cover the outside of the magnets with epoxy paint and hope that this paint will prevent that the nickel coating of the magnets is damaged during mounting.

Another advantage of my armature construction is that mechanically the armature has sixteen poles but physically it has only four poles (see KD 718 figure 2). Therefore it can be used in combination with a standard 3-phase winding of a 4-pole asynchronous motor. But this is only an advantage if an existing motor winding is used. Another advantage of my armature construction is that it is fully rota symmetric and therefore no imbalance is created. For your 12-pole armature there is one large bridge in between the grooves and this means that the magnets are not exactly opposite to each other. This creates a certain imbalance but at the low rotational speeds for which the generator is used in a wind turbine this imbalance may give no problems. But if you would run this 12-pole armature at high rotational speeds, it will vibrate strongly.

[GreenTeam]

If it produced a millivolt of power, than I would say that all the windings are shorted out due to not being packed or varnished.
Being cooked before prolly doesnt help either.
It looks like a wave winding.

Also, if you do manage to get it producing again, ,make sure you drill holes on the bottom to let the water drain out.
As that is the source of all the rust. 

[llaind]

still scratching my head on this one, i decided to use the new motor and use the old end casing and it all fitted together nicely for a change. still need to figure out the magnets on the two spare motors i have, which miraculously, both have intact winds, reading correct resistance, bit of varnish after a cleanup should be good to go. hopefully with all the expertise here eventually it will get solved. thanks for all the input.