First motorconversion; the start

[dinges]

Dear diary.

Today I've made my first steps on the road to conversion. Nobby, thanks for the motor! The specs of it:

3. HP (2.2kW)
   
220. /380V (3phase)
     
9. A/5A (star/delta)
   
1415. RPM (4pole)
      
36. stator slots
    
    

rotordiameter 89mm

bearing 2x 6206-Z

The bearings will be replaced with new ones; was planning on doing this anyway, but one of them makes a sound like grinding sand between the teeth; not good. the motor will be refurbished, new bearings, repainted. The rotor will be completely renewed with a steel cage. Not quite sure what magnets to use, the 2"x1"x.5" ones or many smaller, round ones like Zubbly uses. I prefer to use the large rectangular ones, since there will be fewer to use and thus easier to fix them properly (loctite, perhaps even with an extra retainer screw through the magnets).

But these are all plans. Have just taken apart the motor, this is as far as I got today.

Here's the original motor, used to be part of an old lathe:

And the other side:

With the fan-cage removed, the fan becomes visible & accessible:

After removal of 2 inner hex-screws the fan can be removed; don't think it will be re-used on the generator, since the wind blowing by should cool the motor much better than this fan ever could.

Removal of this key (for the fan) was quite easy:

After the key was removed, I decided to take the backcover off for a quick peak. I was pleased. 36 stator slots. Big bearings (6206-Z). Windings seem to be in a good condition.

Then I put the backside back on and started working on the key at the front. This was not nice. 2 hours later and lots of cursing later, I got it out (drilled a 5mm hole, tapped M6, put in an M6 screw and hammered on the screw till the key came out). Then I ended up with the free rotor. Tried pulling the bearings off with a 'poelie trekker' (whatchamacallit in English? bearing or wheel puller?). Didn't work. So will have to grind it off or whatever other method I come up with.

The rotor is a little under 90mm diameter and about 120mm long. Thus, 2 2"x1"x.5" magnets should fit lengthwise, skewed.

Right. That was the easy part. Next comes figuring out what & how many magnets to use (Acad is great for making quick sketches and figuring this out, not just for drawing technical drawings). And then building the new rotor. If it works satisfactory, work will begin on 'freshening up' the motor itself (repainting).

We've only just begun...

[SparWeb]

Dinges,

Like you, I stand on the brink of converting an old 3HP motor by adding

permanent mags.  Your motor specs are remarkably similar to the one I

have, except for speed rating.  With a 415 RPM motor, doesn't that make

it a 16-pole rotor?  Or did the Auto-Format on the text garble your number?  

Your photos also didn't come thru on the screen, and I can't find them among

your photos.  I can help out with those issues, if you like.

Please stay in touch, as it will be very informative if we compare results

at the end.

[SparWeb]

Re-sized

(free of charge)

If you were to take the re-sized photo inserted here, and replaced your

original among your user files, then your original posting would have

smaller photos.  Accessibility for dial-up users interested in your

project would be improved.

[dinges]

Lots of swearing and cursing later...

The first time using batch-conversion with IRFAN; didn't work, so have manually reconverted and uploaded. Works out fine now.

And the numbers got garbled up; it's a 1415RPM motor; you can find out the number of poles by diving 6000/RPM (in case of 50Hz, like we have in Netherlands; when 60Hz grid you should divide 7200/RPM = # poles).

6000/1415 = 4 poles (note: 1415 RPM is under load; when freely running, it's 1500 RPM; the difference is called 'slip'.

Peter.

[dinges]

Steven,

Thanks. But as soon as I noticed they didn't come out properly I immediately removed them , reconverted (failed a few times) and uploaded again under the identical name. You managed to grab a file off this board while I was still convertying & trying to get it right. You beat me to it

BTW, the file size came out OK all the time (60-70kB), it's just that 'image size' was way too high. I HATE it when that happens with other posters, but now it happened to myself... Sorry. Immediately mended my ways, but you were faster...

[willib]

peter cant you just leave the bearings on there ?

it will save you some cash , and hassle from getting new nes

[kitno455]

i thought the 9A/5A is not star/delta, but low V/ hi V?

allan

[dinges]

Willib,

Don't know about you but I don't like the sound of sand grinding between my teeth. That's the sound one of the bearings makes...

I usually build in a way that comes as near to perfection as I can. Don't want to spend a lot of time building, converting, etc. only later to find out I should have replaced a bearing. This is annoying when the genny is still on the bench, but when it is mounted on top of a tower with a prop on it... You get the idea.

I think the proverb 'penny wise, pound foolish' applies here. Cost of bearings, as part of a complete windgenerator SYSTEM, is negligeable.

Peter.

[willib]

i missed the part of grinding sand. sorry

*  note to self read more thourghly  *

[dinges]

Hmmm...

I've just realized a VERY important thing w.r.t. motor conversions and using them for windgennies; haven't read about this before so I will explain here, as good as I can.

If you look at the bottom picture in my original post, you will see, near the right bearing, a small ring. This fixes the front-bearing to the front-plate of the motor.

Important note: motors are usually only loaded radially; only limited axial force is present. Also, ballbearings (of this type) are designed to mainly accept radial force. Will make some calc's about the maximum axial force that this bearing (6206-Z) will be able to withstand, but I think the bearing won't be the real issue.

What is the problem? That little ring, which fixes the front bearing. It's pretty small and made of cast-iron. It's fixed with only 2 M5 screws (5mm diameter, outside; 4mm core diameter). This is VERY little. And note that this ring, and the two tiny screws, will have to transfer the ENTIRE axial force of the prop to the tower...

BTW, it has to do with the particular way in which this motor is designed, where the front bearing is 'fixed' and the back one is 'loose'.

An important thing to notice!! Like I said, the bearings will stand the full axial force, I think (will confirm with calculations), but those two tiny screws most likely won't.

Was at first thinking of designing a new ring and mounting it with more screws. But there's probably an easier solution: turn the rotor 180 degrees. I think it will be possible with this motor without too much trouble. By reversing front and back of the motor, that ring will never see an axial force; this force will be taken up by the end-plate of the motor. Problem solved. A tiny, but very important detail...

Just a heads up for anyone else that intends to do a motor conversion.

Peter.

[SparWeb]

Quickdraw Steve here,

I don't know the exact shape of the ring, so I may be wasting breath, but here goes:

The axial load will be supported by whatever bearing is secured axially.  Any pair of bearings mounted on a shaft will have one secured and one loose (free to slide axially in the housing slightly).  This prevents the tightening of the motor casing from determining how tightly mounted the bearings are.  If the ring on the end secures the bearing OD axially, it will do so, whether it is on the front or the back.

You may find that tapping more (bigger) holes into the ring is all you need to do.  Then again, perhaps it's too dainty to do anything with it at all.

I'm going to check my motor and see if it has the same problem, or a solution to yours.

[WXYZCIENCE]

Peter, you list 6206Z what is the shaft size? What size of blade do you think you will be putting on the unit? I am working on one very similar.Joe.

[willib]

You should check with mr Zubbly, but IMHO heating the bearing with a torch is preferable to grinding them off.

[dinges]

Thanks Steve.

BTW, I'm mech.engineer so thanks for the explanation on loose & fixed bearings

This link gives some explanation too, for anyone interested:

http://www.skf.com/portal/skf/home/products?paf_dm=shared&maincatalogue=1&lang=en&newlin
k=1_0_79

Just checked it out again and yep, that thin ring would take ALL the axial force. The two screws are way undersized too for this application. If I loosen the two screws, I can move the axle almost 2 mm backward before the back bearing starts taking up the axial load.

My original idea of reversing the rotor I won't do, because then 4 other bolts would be stretched (also M5 bolts). These are the bolts that hold the endplate to the house.

So, the plan is now: to turn a new ring, much more solid; drill it with 6 or 8 holes, M5. And put in high-grade bolts to push this ring to the front endplate. Highgrade bolts like 12.9 (Rm=1200 N/mm^2; Rp0.2=900 N/mm^2). That should do the trick but will check with calc's too. Plus, I never intend to use this genny above 1kW, not near to the max. 2.2kW it can (electrically) handle. This way I've got some reserve. BTW, I'd never load any of my gennies over 50% of the maximum design load, but that's another discussion.

So I'd be using the original method of locating the bearings, just beefed-up a bit.

In this configuration, the original 4 bolts that hold the front-endplate to the housing won't be loaded, since the pushing axial force is transfered via 'push' to the main housing; the bolts aren't loaded (neither push nor pull).

Easier to see on a drawing than to explain in words; maybe I'll make a quick drawing of it someday and post it here. It's important enough.

[dinges]

I don't know yet what size of blades. It depends a bit on the voltage that this genny starts putting out at differnt RPMs. I'm guessing at the moment 14ft, but that's just guessing. Basically the rest of the system will be designed around this motorconversion. And before further design work can start, I need some measured results. And for that, the genny should be finished first But I'm very interested in anyone's opinion on the matter of prop size, TSR, etc!

The axle diameter is 30mm (you can easily check yourself for different size bearings on www.skf.com)

I really hope I will not have to rewind the stator. No matter how good Zubbly explains it... When I look at the stator I get a tad confused No doubt I could do it, if I put my mind to it, but still.

Looking forward to hearing more about your conversion.

Peter.

[dinges]

Please explain. I'm not sure what I wrote is correct, but it seemed logical.

Are there countries where they use 3phase 220V (as opposed to 3phase 380V in NL, and single phase 220V)?

Peter.

[sahlein]

Peter,

In some small shops and light industry in the U.S. one can find

motors being run on 3 phase 220.

sahlein

[dinges]

Zubbly, if you read this:

When I use 2"x1"x.5" magnets, should I maintain minimum airgap between rotor and stator, i.e. as small as is physically and technically possible? Or can airgap be too small and cause problems (cogging, too low cut-in, ?).

I'm right now designing a new rotor and am trying to keep airgap as low as possible, but am not 100% sure whether that's a good idea at all.

Peter.

[vawtman]

Hello Peter isnt this fun

 When you say steel cage for the rotor what do you mean by that?

 The mag size is determined by coil space and must be skewed

 The nice thing about skewing the stator is you can have more poles since you have to rewind.

 Ive asked the experts so many questions zubbly and flux their probably sick of me.

 Hang in there.

 Keep the airgap tight

 Ive decided to play follow the leader

 Good Luck with ____.

 

[SamoaPower]

Yes, but wouldn't you still be stuck with a bearing doing a job it wasn't designed for? Has anyone tried adapting a tapered roller bearing to a motor? Seems you would need only one.

[WXYZCIENCE]

Peter, I got lots of the answers from Zubbly for my conversion.
Here is the Diary Post:

http://www.fieldlines.com/story/2006/5/11/2116/17183


Joe.

[dinges]

Just made a quick sketch, see below. These are 8 magnets N40, 2"x1"x.5". Machining the rotor will be no problem for me, but:

I'd rather use 16 magnets, but they don't physically fit. So, 8 is the way. Yet, this means my rotor is only partly filled with magnets, whereas Zubbly's conversions are usually almost completely filled with smaller, round magnets.

I'd plan on having a round rotor with pockets milled for the magnets. The magnets will be Loctited in (youse think this will be enough, or should I EDM a hole in the magnets and keep them down with a stainless retainer screw as well?) Have absolutely no experience using Loctite.

And, what does seem the best solution to youse: going this route (8 rectangular magnets), or doing what Zubbly usually does, using lots of smaller, round magnets? I prefer 'my' route, but I think I'm seriously limiting output by using a relatively small part of the rotor for magnets.

Very interested in hearing your opinions!

[dinges]

Picture resized to 100% for better readability:

[dinges]

Now that I think about it; if you use two big magnets like this, as opposed to many small round ones, they probably should be skewed just 5 degrees?, not the 360/36slots=10 degrees.

If you make a mental picture, I think that in the drawing above it would still cogg terribly. At an angle of 5 deg, I think it wouldn't anymore. In my thought-experiment it should be 5 deg, not 10...

Not sure. Anybody has experience using these kinds of magnets in this situation, for a motor conversion? I'd hate to waste lots of money and effort, only to find that it still cogs terribly...

[Countryboy]

Hello Dinges,

  You should be able to press the shaft out of the bearings.  A press will work much better than the gear puller "a 'poelie trekker' (whatchamacallit in English? bearing or wheel puller?)".

  I don't know how well "press" translates.  It can be a noun (thing) or a verb (action).  A press (noun) is a hydraulic jack mounted in a frame.  As the hydraulic jack is pumped, the hydraulic ram moves toward an unmoving part of the frame.  The bearing would be placed against the frame, and the hydraulic ram would be placed against the hydraulic ram.  

  Be sure that the shaft is aligned with the hydraulic ram preceisely.  If it is not, you may bend the shaft if you are not careful.

  You also said,

"If I loosen the two screws, I can move the axle almost 2 mm backward before the back bearing starts taking up the axial load."

  Is it possible to make a 2mm shim to place behind the back bearing to eliminate the 2mm shaft endplay?  If you move the back bearing 2mm, the shaft will have zero endplay before the back bearing absorbs axial loads.

[Countryboy]

Dinges,

  Look at the stator armature right now.  You will notice that the metal is angled.  Measure that angle, and place the magnets on the same degree of angle.  

  Zubbly recently had a topic in which he discussed skewing the laminations, so that he would not have to skew the magnets on the rotor.

[dinges]

Hopefully this drawing explains my cogging problem better:

To be honest, I think that, the way Zubbly de-coggs, using many small magnets, that it is the magnets that are in the middle do the real decogging; the idea is that, as the rotor rotates, that the magnets see, on average, the same amount of metal from the stator. If you mount the magnets at 10deg, as theory tells to, both magnets are closest to (different) fingers of the stator. As it wants to turn a bit more, BOTH magnets will 'see' more air than steel and thus resist. When the magnets are skewed by half, i.e. 5 degrees, as the rotor turns the amount of air and steel that both magnets will see together (as a sum) remains the same. Hope I'm at least a little bit clear...

Very interested in your opinions.

[vawtman]

Peter the 5hp 4pl i did earliar had enough room in the span for 6 mags but there was no way they would sit that close together so ended up using just 4 with good results according to the open voltage readings i sent Zubbly.

 If you only skew 5 degrees you might run the risk of heavy cogging the difference between 2 fingers and a pipewrench to turn the shaft.

 I tried drilling a hole in a ceramic mag once i not only changed colors but the mag broke just has i got through.I know Zubbly can do it.

 Has long has you mill your pockets very shallow you will be ok dont embed them in the rotor.

 Hang in there

[ghurd]

I like the idea of a tapered roller bearing since I changed to them in an old VW.

The wind-thought is they have a lot more rolling resistance.

Would a mech. engineer put the one roller bearing at the blade end or the tail end?

G-

[dinges]

Personally I'd put the bearing in the place that the pushing-load of the prop is transferred to the motor-house from the end-plate via pushing force. In my case, that would mean mounting such a bearing on the front plate. And I would still need to reinforce that retainer ring and screws, as I explained before. If I mounted such a bearing on the backplate, the prop and the axle would transfer the axial force to the backplate by trying to 'push' that plate away from the main motor house. This would mean that the 4 M5 screws that fix the back-plate would see a pulling load.

It's just my idea on the matter, if anyone else has other ideas, I'm interested.

But, Ghurd, you know:

'The only thing 2 engineers agree upon is that the 3rd one is an idiot'  

[ghurd]

I also figured the front plate, partly because it will be where there is more radial force.(?)  Unlike the ball bearings, I never had to change roller bearings in the old Beetles.

Lately I feel like the 3rd one...

Did you ever get my email?

G-

[dinges]

Sorry, haven't received any mail from you; just checked again.

Peter.

[ghurd]

Check again?