Author Topic: K55 ceiling fan motor  (Read 39092 times)

0 Members and 1 Guest are viewing this topic.

ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: K55 ceiling fan motor
« Reply #33 on: September 25, 2010, 10:51:23 PM »
I don't think you need the extra magnet volume, especially if they won't easily fit.

A bit of skew certainly won't hurt.
36 neos : 36 slots is going to cog no matter what.
Could use 72 square neos, something like Spdlmt150 except 2 in a row instead of 4,
http://fieldlines.com/board/index.php/topic,128761.0.html

Plain epoxy is fine.
The "5 minute" stuff is NOT waterproof.  Just resistant.
The slow cure stuff is waterproof.
Make sure the neos are clean of hand sweat, etc (i use 70% rubbing alcohol).

JBWeld is probably not a great idea for the first attempt.  It is full of iron dust and it crawls all over the neos and makes a heck of a mess, which seems worse in small projects.

1/4x1/4 or 3/8x3/8 kind of leaves you guessing about the polarity.  They stick together in more than one way.
I would suggest also getting a couple neo disks to use for a magnetic reference, and mark each neo with a perm marker on the side with the same polarity.  It doesn't matter if you mark the N or S, but mark them all the same.
G-
www.ghurd.info<<<-----Information on my Controller

hiker

  • Hero Member
  • *****
  • Posts: 1661
  • BIG DOG
Re: K55 ceiling fan motor
« Reply #34 on: September 25, 2010, 11:06:30 PM »
try the jb-weld intended for sealing gas tanks..or water tanks..
it doesent have any iron in it...its stronger than epoxy and water proof..
what i do is to use super glue to stick the mags down--then bury the mags in the jb-weld..just the top face of the mag sticks out..
good luck..............
WILD in ALASKA

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #35 on: October 17, 2010, 11:29:04 PM »
What is the effect of using less than optimum width bar magnets on the new rotor. For instance, with 18 poles, 36 slots, and 36 stator teeth, the slots are .100" wide, the stator teeth are .185". One pole is therefore .185" + .100"+.185" wide or .470". The largest width 1" long x square cross section bar magnet that will fit the rotor will be 3/8" wide. What would the performance difference be between 1/4" wide and 3/8" wide magnets? Would the narrower 1/4" magnets lessen the cogging?

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #36 on: October 18, 2010, 10:20:11 AM »
On second thought, maybe the bar type magnets would not be the best configuration. What magnet shape, size, and configuration would work best on the new rotor considering power output and cogging? I have only a .471" wide space on the circumference for each pole using 1/4" thick magnets. Maybe a little less depending on magnet width. Thanks for any suggestions.

Jeff

ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: K55 ceiling fan motor
« Reply #37 on: October 18, 2010, 11:12:05 AM »
If you have the machines, might try two 3/8 square x 1/4 or 3/16" thick per pole, skewed like Spdlmt150 did.
0.050" or 0.075" offset?

Do not make tolerances so close it causes problems.
If a neo slides a few thousandths sideways, it can be a big problem.
The laminate may not be exactly round... or centered.
etc.

By cheap paper thin neo, I meant don't use 1mm or 1/16" thick.
I don't think it needs 1/4" thick neos.  There is a point when more flux really does not help.  If you can keep the average distance from the neo face to the laminate to maybe (?) half the neo's thickness, I don't think thicker neo will do much good for something like this.
If 1/4" thick fits, OK.  Not sure there would be a big loss going with 3/16".
Be sure whatever you decide is the perfect size... nobody will sell it.

If you make a new rotor out of solid steel, can get a decent increase in output too.  10% sticks in my mind for some reason.
It gets rid of the AL which acts like air.
The rotor can not be SS.
G-
www.ghurd.info<<<-----Information on my Controller

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #38 on: November 10, 2010, 06:53:18 PM »
Quick question. This is an 18 pole motor with 36 slots. I going to use the 1" x 1/4" x 1/4" bar magnets skewed at a, yet to be determind angle. When I place the magnets on the rotor, are they placed side by side alternately N and S, or are they arranged so that the same polarity magnets are always 180 degree opposite with each other? If the latter is so, it won't work with 18 poles. I am a bit confused on this point.

Jeff

Ungrounded Lightning Rod

  • SuperHero Member
  • ******
  • Posts: 2865
Re: K55 ceiling fan motor
« Reply #39 on: November 10, 2010, 07:03:58 PM »
Quick question. This is an 18 pole motor with 36 slots. I going to use the 1" x 1/4" x 1/4" bar magnets skewed at a, yet to be determind angle. When I place the magnets on the rotor, are they placed side by side alternately N and S, or are they arranged so that the same polarity magnets are always 180 degree opposite with each other? If the latter is so, it won't work with 18 poles. I am a bit confused on this point.

Jeff

The former.  2N poles where N is even have identical poles opposite each other and work just fine.  There's nothing magic about opposing positions.

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #40 on: November 10, 2010, 10:36:54 PM »
ULR,
So if the number of N poles, and S poles is odd, then there will be no problem if the opposing poles are different polarity?

Jeff

Ungrounded Lightning Rod

  • SuperHero Member
  • ******
  • Posts: 2865
Re: K55 ceiling fan motor
« Reply #41 on: November 11, 2010, 01:19:01 PM »
ULR,
So if the number of N poles, and S poles is odd, then there will be no problem if the opposing poles are different polarity?

Sure, if the slots work out for you.

Think of the rotor and stator as an infinite linear motor/generator with alternate N and S poles going on forever.  Then chop out an integer number of N/S pairs and bend it into a cylinder.

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #42 on: November 11, 2010, 02:36:51 PM »
ULR,
Got it! Thanks.

Jeff

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #43 on: November 11, 2010, 08:28:49 PM »
I tried a stator using 1" x 1/4" x 1/4" neo bar magnets. I built it according to the "offset method 2" described in the link below which looked ideal for bar magnets:

http://www.otherpower.com/images/scimages/3538/decogging_tutorial_V1.pdf

The 1' X 1/4" X 1/4" bar magnets cogged terribly with this method on the K-55 stator. I could not turn the shaft by hand. The author says the method hadn't been tried "but it  should work".  Ok, I've tried it...! 

The maximum practical magnet width to fit the circumference of the 18 pole stator is about 3/8" . Maximum depth is still about 1/4".  I can probably skew the bar magnets on the existing stator, but I am not so sure they would work much better than what I tried. 

I would appreciate suggestions on what magnet type and what anti-cogging configuration would work better on this shallow 1" deep stator lamination?

Jeff


ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: K55 ceiling fan motor
« Reply #44 on: November 11, 2010, 11:07:30 PM »
I believe you have too many poles for the offset method(s) to work.
Myself, I don't care for it.

I generally use something like on the top of page 8.
The magnets are curved.
The crawling JB Weld shows.
It does not cog much.


« Last Edit: November 11, 2010, 11:09:05 PM by ghurd »
www.ghurd.info<<<-----Information on my Controller

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #45 on: November 12, 2010, 10:29:04 PM »
Since the "Off-set method #2" didn't help much to minimize cogging, I decided to re-machine the rotor for a 14 degree skewed magnet mounting, and try it again. At 14 degrees, the 1" long bar magnets covered two stator teeth equally. The rotor went into the stator with a resounding "thump" like before (and comes out hard), but this time I was able to turn the rotor by hand. Some progress at least! It still cogs, but can be easily turned. It feels, and sounds much like a stepper motor now. I am not sure what the cogging torque is, but I think I can live with it since I don't know what else to try to further minimize cogging. The 4 winding coils are all isolated, and all four produced 92v no load at about 650 RPM. A little higher than I expected. Next will be to find some time to do some load testing.

Thanks all for the continued guidance.

Jeff

ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: K55 ceiling fan motor
« Reply #46 on: November 13, 2010, 09:46:09 AM »
That great!

Time to clutter up the board with a pic of it?
G-
www.ghurd.info<<<-----Information on my Controller

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #47 on: November 14, 2010, 11:08:13 PM »
Ghurd,
I was going to do some load testing, but figured that I had better epoxy the magnets to the rotor before doing so. I spun it up to 700 RPM and didn't loose any initially, but with load there are more forces at play. Here is what it looks like presently:



After testing, I intend to encapsulate the magnets in epoxy more so to protect them from any handling damage. I will likely be inserting the rotor into the stator quite a few times during the fabrication of the new stator framework and rear bearing support, and I don't want the magnet surfaces nicked or scraped. The rotor has a mind of its own when inserted into the stator. 

Jeff

ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: K55 ceiling fan motor
« Reply #48 on: November 14, 2010, 11:15:12 PM »
Looks very nice.

I use a 2 liter soda bottle, cut into a sheet, to protect the magnets when sliding them in.
Wrap the plastic around the rotor, pop it in, pull out the plastic.
G-
www.ghurd.info<<<-----Information on my Controller

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #49 on: November 14, 2010, 11:34:01 PM »
Ghurd,
Yes, I am aware of the technique, and it is a good idea. My plans might change, but that is the plan for now.

Jeff

hiker

  • Hero Member
  • *****
  • Posts: 1661
  • BIG DOG
Re: K55 ceiling fan motor
« Reply #50 on: November 15, 2010, 02:55:38 AM »
nice work...........
check my lastest post....if your going to bury your mags in epoxy..i would go for -epoxy stick- dries as hard as rock...
no more worries about your mags breaking off when you insert or remove your rotor...
its a whole lot better than using just regular epoxy--stronger and  water restance...
WILD in ALASKA

Ungrounded Lightning Rod

  • SuperHero Member
  • ******
  • Posts: 2865
Re: K55 ceiling fan motor
« Reply #51 on: November 15, 2010, 09:22:56 PM »
At 14 degrees, the 1" long bar magnets covered two stator teeth equally.

You want each skewed set of magnets to cover an integer number of stator poles AND SLOTS.  One pole and one slot, two poles and two slots, etc.  (With a bar magnet one end will cover the slots and poles the same way as the other end, but offset by a pole or two.)  That way you always have the same amount of stator pole material under the rotor magnet poles regardless of angle.

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #52 on: November 16, 2010, 11:20:40 PM »
I did some informal testing today on the K55 generator. Measurements and load were taken on only one of the 4 coils. All 4 coils seemed to perform similarly.
RPM       Volts AC (no load)
100         16.4   
200         31.0
300         48.5
400         63.8
500         81.0
600         96.4
700        115.1  Loads: 1. Added 15 watt light bulb load. Load was 0.13A at 111.5v
                                 2. Added 40 watt light bulb load. Load was 0.30A at 105.2v

For my purposes this output looks adequate. Usage will be at 500-600 rpm range. I will probably split the output. One pair of coils will be rectified to 14vdc, the other will feed AC lighting load at output voltage, through a rheostat. 

Thanks for all the help and suggestions.

Jeff



ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: K55 ceiling fan motor
« Reply #53 on: November 17, 2010, 09:42:35 AM »
Nice!

The layman is more impressed with more bulbs.
Might consider 10 #194 bulbs in series for 110VAC?
Looks like you could do 2 separate strings of them.  20 bulbs.  :o
(the sockets are crazy expensive)

Or could light up a whole bunch of 110VAC LED Christmas light strings!
G-
www.ghurd.info<<<-----Information on my Controller

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #54 on: November 17, 2010, 01:21:18 PM »
Ghurd,
I agree. The mentioned bulbs were just to see what the voltage droop would be at various loads. The droop was lower than I expected, so I think the generator is stout enough to handle well in excess of 100 watts. Unfortunately, I suspect the efficiency is quite low and my little 4 cylinder engine will be hard pressed to produce the needed torque for more than about 40-50 watts of power. I will likely be running it in the 81-96v speed range. The AC load would be an array of very small bulbs on one phase with a rheostat or dimmer switch to be able to vary the lighting wattage. The other phase (or maybe one coil) will be tranformed and rectified to handle the 6 watts of 12vdc loads.

Now to design and builld the generator frame and bearing support. Thanks again for your help.

Jeff

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #55 on: December 30, 2010, 06:39:30 PM »
It has been awhile, and this project is fiinally nearing completion. I have a few more remaining newbie questions. I have isolated each of the generator's 4 coils.  I am getting 115vac on each coil at the desired 700 RPM. These coils were originally wired two in series in each phase winding with two phases. My first question is, can I parallel two or maybe three of these coils for the AC output for lighting, and then use the remaining coil (s) for a rectified 14v DC output circuit.

My second question is should one leg of the AC output be frame grounded? And also, how should the AC to DC rectifier side be grounded? This is a demonstration model and safety is important.

As always, thanks for any assistance.

Jeff

ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: K55 ceiling fan motor
« Reply #56 on: December 30, 2010, 07:08:35 PM »
Maybe?
You can only parallel 2 coils. They have to be in phase, which is a pair that were in series. 
It may not be possible because of the winding and how the magnets were installed.
Feel how easy it spins when the coils are not connected.  If it gets any harder with the coils connected, you either have the wrong coils (probably feels like heavy drag), one is reversed (turns very stiff), or the magnets are making them a bit stiff they may be out of phase.

I think I would use them separately anyway.  Less issues with circulating currents.

Might be worth while to feed that 115VAC to a 12V transformer, then rectify to DC.
Direct rectification will result in a high DC voltage, and that's considered not good.

I do not think 'grounding' anything is a good idea in this situation.  If there is a good way to do it, I can't think of it right now.
If there is a good reason to do it, I can't think of it either.
Maybe someone else will chine in.
G-
www.ghurd.info<<<-----Information on my Controller

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #57 on: December 30, 2010, 10:24:49 PM »
Ghurd,
Thanks. I suspected only the phase coils can be paralleled. I prefer to have only one AC circuit and one DC circuit so I will try the two paralleled circuits first. I will try the individual coil circuits if the paralleled coils don't work out. Using the 4 individual circuits is a lot more wiring. I like the transformer idea, but I have to include a PWM regulator since the AC output voltage may vary as much as 15-20% depending on how the speed is set and how much load is applied. A unit this small has no governor.

I will skip the grounding for now, since I will rarely have an earth ground available.

Jeff

ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: K55 ceiling fan motor
« Reply #58 on: December 31, 2010, 12:53:38 AM »
Not sure what the DC power is intended for, but most "12V" devices can operate properly from about 10.5V to near 15.0V.

And a transformer will have a lot more variation due to loading too.  Would not be uncommon to have an unloaded 12V transformer making 18VAC, which will increase even more after rectification to DC.
G-
www.ghurd.info<<<-----Information on my Controller

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #59 on: January 28, 2011, 01:04:10 PM »
This project is nearing completion. AC output is working well at 115vac. A DC output is still needed to operate a 0.5 amp 12v computer case fan that will be used for cooling. I was planning on transforming the 115vac output down to about 18-30vac, rectifying to DC, then control the output to the fan with an adjustable LM-317 regulator. However, I understand that transformers are adversely affected by higher frequency. The generator produces approximately 105 hz. Can I just oversize a standard small transformer (if so, by how much?) and get by, or are there other issues? I was also considering using a regulated universal AC adapter for an all-in-one solution for the DC circuit.
Jeff

ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: K55 ceiling fan motor
« Reply #60 on: January 28, 2011, 01:40:48 PM »
The frequency is less than double standard 60Hz.  Should be fine.

Need (or should want to) go lower than 18-30VAC.
 VAC x 1.4 = VDC.   
30VAC goes to (30 x 1.4=) 42VDC, and LM317 blows up at 38VDC input.

A 10VAC transformer would be fine.  14VAC at the most.

LM317 is kind of complicated to get 12VDC.
They make a non adjustable version, 7812, sort of the same thing except the adjusting resistors are in it already.

There is a heat issue too.
Simplified,  317 or 7812 heat is Amps x Voltage reduction.
(30Vin - 12V out) x 0.5A = 18V x 0.5A = 18W
18W of heat in that little thing is going to make Magic Smoke.

Plus, why waste 18W in heat to get 6W usable output?

After the bridge on the DC side, use a big capacitor to keep things smoother.

The simplest way is get a 12VDC regulated output wall-wart.  All the good stuff is in it already.
115VAC in.  12VDC out.

If you are determined to make the thing youself, shoot me an email....
Armpit deep in all that stuff, except for the transformers.
G-
www.ghurd.info<<<-----Information on my Controller

rustkolector

  • Jr. Member
  • **
  • Posts: 69
Re: K55 ceiling fan motor
« Reply #61 on: September 19, 2011, 10:07:38 PM »
It has been quite awhile since updating this thread. This project is finally completed and running better than expected. It is not a wind turbine, but I was able to find the help I needed here to complete this project. This is a 1/10th  scale model of a slow speed 1912 vintage Bruce-MacBeth 100 hp stationary natural gas engine driving a similar vintage alternator modeled after a Westinghouse alternator design. The model alternator core is an Emerson K-55 ceiling fan stator. The alternator rotor uses 18 neo magnets for excitation and produces about 100 watts at 120vac and 750 RPM from each of the two windings. The 4 cylinder engine (1" x 1.12" bore x stroke) was built from bar stock and runs at 625 RPM on propane with an alternator voltage of 100v. The engine easily pulls 40-50 watts (and will pull much more), but runs nicely for demo with a 15-20 watt load. The clear plexiglass inspection door window in the engine is not permanent. It is for watching the splash oil system to select a good oil viscosity and RPM setting to maintain adequate internal lubrication.

I love to model working antique engines doing what they used to do. My thanks to all who helped me with this project. 

Jeff


http://www.youtube.com/watch?v=2-qdar5TEpQ

12AX7

  • Hero Member
  • *****
  • Posts: 814
Re: K55 ceiling fan motor
« Reply #62 on: September 19, 2011, 10:20:27 PM »
Very COOL!

TomW

  • Super Hero Member Plus
  • *******
  • Posts: 5130
  • Country: us
Re: K55 ceiling fan motor
« Reply #63 on: September 19, 2011, 10:33:19 PM »
Super Stuff!

Thanks for the share.

Tom

oztules

  • Hero Member
  • *****
  • Posts: 1477
  • Country: aq
  • Village idiot
Re: K55 ceiling fan motor
« Reply #64 on: September 20, 2011, 05:13:00 AM »
Took 3 times to attampts to reply.... but gee
Damn classy stuff. You need to be proud of that.


.............oztules
Flinders Island Australia

Bruce S

  • Administrator
  • Super Hero Member Plus
  • *****
  • Posts: 5370
  • Country: us
  • USA
Re: K55 ceiling fan motor
« Reply #65 on: September 20, 2011, 09:29:12 AM »
A beauty to watch. I'd leave the plexiglass there for ALL to watch.
Do you build stuff like this for a living?
Thanks for the update!!
Bruce S
A kind word often goes unsaid BUT never goes unheard