Author Topic: report KD 632 about 26-pole PM-generator available  (Read 1502 times)

0 Members and 1 Guest are viewing this topic.

Adriaan Kragten

  • Full Member
  • ***
  • Posts: 154
  • Country: nl
report KD 632 about 26-pole PM-generator available
« on: May 11, 2017, 12:19:10 PM »
The free public report KD 632 has been addes to my website: www.kdwindturbines.nl at the menu KD-reports. KD 632 describes a 26-pole PM-generator using the housing of a 4-pole, 3-phase, 0.75 kW asynchronous motor frame size 80 and 26 neodymium magnets size 40 * 7 * 3 mm.

JW

  • Development Manager
  • SuperHero Member
  • *****
  • Posts: 2335
  • Country: us
    • Flashsteam.com
Re: report KD 632 about 26-pole PM-generator available
« Reply #1 on: May 11, 2017, 07:09:55 PM »
How similar is this to what Zubby was doing

joestue

  • Hero Member
  • *****
  • Posts: 1411
Re: report KD 632 about 26-pole PM-generator available
« Reply #2 on: May 11, 2017, 08:01:36 PM »
How similar is this to what Zubby was doing

I don't think he did anything beyond the usual one pole per 3 slots.

I'm not sure you can get a higher torque density out of a concentrated pole motor (such as the 26 pole, 24 slot, or 30/34/38 pole, 36 slot) than you can out of an 8 pole, 24 slot machine, but the high pole count machines have far lower cogging torque. Though its true you can reduce the cogging torque by skewing the magnets, you get a lower voltage out, you don't have to do that with a concentrated pole motor.

high pole count machines work well with transformers, low pole count machines don't, simply due to the low frequency output.

this has nothing to do with the generator, but simply that a transformer's power density is proportional to frequency, a generator's power density (by weight) is proportional to both rpm and frequency. (provided of course, the generator was optimally designed for the pole count)


high pole count generators may get the same torque as a low pole count machine when only looking at the surface area and radius of the air gap, but a high pole count machine can do so with far less weight.


so when re-using motor stampings, i'm not sure if there is any advantage to a high pole count machine except for two things: lower cogging torque, and greater efficiency due to lower end turn losses.

the torque density is fundamentally limited by the air gap flux density. the air gap flux density can be increased with a high pole count design because you're no longer limited by the back iron, but the tooth width. if you don't care about iron losses then it probably costs less to get a higher air gap flux density with a concentrated pole motor than a traditional motor.

however, the concentrated pole motor needs thinner and smaller magnets.. these cost more than thicker magnets... so your milage may vary.

the difference seems to be as high as 3:1 for the magnet cost when dealing with 1 inch cubes (25mm) or 1/2 by 1/4 (12 by 6mm) magnets.
« Last Edit: May 11, 2017, 10:07:22 PM by joestue »

Adriaan Kragten

  • Full Member
  • ***
  • Posts: 154
  • Country: nl
Re: report KD 632 about 26-pole PM-generator available
« Reply #3 on: May 12, 2017, 12:52:47 AM »
I expect that the maximum torque level of the 26-pole radial flux PM-generator as given in report KD 632 will be about the same as that of the 8-pole axial flux generator as given in report KD 631. The total magnet costs of the 26-pole generator are about 18 and for the 8-pole generator are about 75. So the magnet costs of the 8-pole generator are much higher. The large air gap of the 8-pole generator results in a much lower magnetic field in the coils and therefore much more copper will be needed to get the same power at the same rpm. So the copper costs will also be much higher.

The advantage of using the housing of an asynchronous motor is that these are produced in large quantities and that therefore they are relatively cheap if they are ordered at the original manufacturer. Other advantages are that they are compact and the inside is well protected against the elements. It should be good if someone could make both generators and compare the technical and economical qualities. It depends also on the available machines and craftmanship which option is best for a certain person or company.
« Last Edit: May 12, 2017, 06:39:30 AM by Adriaan Kragten »

Adriaan Kragten

  • Full Member
  • ***
  • Posts: 154
  • Country: nl
Re: report KD 632 about 26-pole PM-generator available
« Reply #4 on: May 12, 2017, 07:37:12 AM »
Now I will explain more in detail the cause of the large difference in magnet costs in between the 26-pole radial flux and the 8-pole axial flux generator.
The 26-pole generator uses 26 magnets size 40 * 7 * 3 mm. So the volume of one magnet is 0.84 cm^3 and the total magnet volume is 21.84 cm^3.
The 8-pole generater uses 8 circular magnets with a diameter of 45 mm and a thickness of 15 mm. So the volume of one magnet is 23.86 cm^3 and the total magnet volume is 190.88 cm^3.
So the total magnet volume of the 8-pole generator is a factor 8.74 larger than that of the 26-pole generator.

The magnet costs depends on the magnet size, the quantity, the costs of transport, the quality of the magnetic material, if VAT is included or not and very much on the chosen supplier. Both magnets can be ordered at the same rather cheap Polish supplier Enes Magnets, so a different supplier is not the cause of the price difference. The costs of transport are relatively high if only a small number of magnets is ordered. Therefore I always order at least the quantity needed for manufacture of two generators. An extra advantage of buying a rather large quantity is that the price per magnet goes down at increasing quantity. For comparing both options, I forget the costs of transport. The quality of the small rectangular magnets is N38SH and of the big round magnets is N35. So the remanence Br will be a little less for the round magnets but this difference can be neglected.

For two generators I need 52 magnets size 40 * 7 * 3 mm or 16 magnets size round 45 * 15 mm. The magnet supplier gives on his website a list with prices (including VAT but excluding costs of transport) depending on the quantity.
For a quantity of at least 40 pieces, the price per piece is 0.70 for the magnet size 40 * 7 * 3 mm. So the price is 0.83 per cm^3.
For a quantity of at least 7 pieces, the price per piece is 9.88 for the magnet round 45 * 15 mm. So the price is 0.41 per cm^3.

So the price per cm^3 is about double for the small magnets. However, the total required magnet volume of the 8-pole generator is more than a factor 8 higher and this finally results in total magnet costs which are more than a factor four higher for the 8-pole generator than for the 26-pole generator.

The fact that the 8-pole generator has no iron in the stator has as advantage that it has almost no sticking torque and therefore starting will be easy and the maximum efficiency will be very high. But the high magnet and copper costs require a much higher investment in the generator and I doubt if the higher investment is paid back by the somewhat higher output (assume the same wind turbine rotor is used).

Adriaan Kragten

  • Full Member
  • ***
  • Posts: 154
  • Country: nl
Re: report KD 632 about 26-pole PM-generator available
« Reply #5 on: May 14, 2017, 02:38:38 PM »
A new chapter 7 has been added to public report KD 341. In this chapter 7, the 26-pole radial flux generator is compared to the 8-pole axial flux generator. The comparison is not only done for the magnet costs like given in my earlier posts but also for the estimated torque levels.

Adriaan Kragten

  • Full Member
  • ***
  • Posts: 154
  • Country: nl
Re: report KD 632 about 26-pole PM-generator available
« Reply #6 on: October 06, 2017, 11:23:03 AM »
Some new chapters have been added to report KD 632. In these chapters, a simple 2-bladed rotor with a diameter of 2.1 m and cambered steel blades is described. In the new chapter 9, a method is described to determine the number of turns per coil such that the rotor and the generator are matching well.

Adriaan Kragten

  • Full Member
  • ***
  • Posts: 154
  • Country: nl
Re: report KD 632 about 26-pole PM-generator available
« Reply #7 on: October 10, 2017, 01:09:01 AM »
Report KD 632 has been modified again. The rotor diameter has been increased from 2.1 m to 2.2 m and the strip which connects both blades has been made wider.

kitestrings

  • Hero Member
  • *****
  • Posts: 756
Re: report KD 632 about 26-pole PM-generator available
« Reply #8 on: October 10, 2017, 07:34:57 PM »
Hi Adriaan,

I just read through the 632 report.  Admittedly, it is a little out of the realm of my experience or expertise; still quite interesting to follow your methods.  I had a few questions:

What led you to select galvanized steel blades?

Why two blades?

In your detailed rotor description you describe a 2-bladed rotor/hub, but in calculating the Vstart you say that 3 m/s is acceptabl[y] low for a 3-bladed rotor with a tip speed of 4.75, so I was confused there.  I assume, perhaps incorrectly, that a 2-bladed rotor might have a bit lower starting torque; or require a bit higher wind to start might be better said.

Lastly, can you explain the Q-n curve?  Is that just the unloaded characteristic of the rotor?

Kindly, ~ks