Help Understanding Stator Design

[jwells02]

I've been working on the design of a stator and have been trying to redesign what I have to make it smaller, simpler and cheaper to build. But the design I want to go to doesn't seem to be as effective as the original. The original uses (2) 1/2" steel bars bent into a U almost like a triangle with a corner cut off, welded across each other with (4) coils. The one I want to go to uses (4) pieces of the same 1/2" bar welded to a 3/16 steel plate with the same (4) coils. I'm trying to understand the reason for the difference in magnetic strength. I've speculated that the 3/16 steel plate is not thick enough? The coils are butt up against each other and working against each other? The bars need to be angled more directly at the rotor so the flux engages it better? I get some loss due to the right angles the flux makes from bar to steel plate?

I've attached a drawing. Any help would be greatly appreciated.

 

[IRstuff]

One would have thought that someone would have at least attempted to duplicate the functional interface. Seems to me that you need to find a qualified designer.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[jwells02]

Thanks for the response, Sheldon. That was quite helpful.

 

[Skogsgurra]

Hi, Sheldon 2 here.

It would help a lot if you showed us the rotor design as well. Is it cylindrical, flat, conical, sperical or some other shape? Around which axis is it turning?

All that means a lot to the air gap between stator and rotor and that, in its turn, is the major factor in determining the flux per ampere-turns. If you angle the butt ends so they are positioned and angled like the original's ends, there will be a much better flux density and a better function.


Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[jwells02]

Thanks. This is a simple mag drive that does a number of things but it's main function is similar to a magnetic stirrer you'd see in a lab. It engages a magnetic bar and rotates it around what you would picture as the center vertical axis of the stator. The gap can be anywhere from 1/2" to 2". I need to reduce the overall size and I'm finding myself outside my comfort zone when it comes to the shape of the core and how the magnetic flux will be effected.

I appreciate you looking at this and pointing out what I have wrong. You answered the question about the angle of the ends of the core.

The pairs of coils are wired in series and operate as a single larger coil. Because of the reduced size the coils are up against each other. Does that have a significant negative impact?

The cores are welded perpendicular to the plate as opposed to a rounded path for the magnetic flow. Anything significant?

Should I be concerned about the thickness of the plate or the shape?

Would there be a significant difference if the cores are screwed onto the plate instead of welded?

Is there anything else you see that's working against me?

Again, I appreciate you looking at this and giving your advice.

 

[EdStainless]

Does the incline angle matter?
Is the plate thick enough to carry the flux?
Dose flux leakage from the edges of the plate matter?

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[electricpete]

The shape and path length within the iron aren't very important (unless you are deep into saturation). The path length in air is limiting / critical (needs to be made as small as pratical). Similar to what skogsgurra said: the original tilted design is better if it gets the stator iron closer to the rotor (minimizing the airgap). Your design would be greatly improved if you could attach another core piece on the end going radially inwards as far as you can.

It doesn't look like a great design to begin with. Ideally the airgap between stator poles and rotor should be far less than the gap between adjacent stator poles (I assume you have alternating polarity poles). If not (as appears in your design), a significant portion of the stator flux will go directly between stator poles (contributing nothing useful) without even going through the rotor.

I'm still not understanding the rotor. "magnetic bar". Do you mean permanent magnet, or just high magnetic permeability (like steel). Given that the axis of the stator is vertical, is the axis of the magnetic bar horizontal or vertical?

By the way, is your core laminated? What is the core material and excitation frequency? (eddy currents can reduce your flux).


=====================================
(2B)+(2B)' ?

 

[jwells02]

The rotor (magnetic bar) is a permanent magnet and the core is plain carbon steel It rotates on the stators vertical axis typically 1" above the stator and it is horizontal. The frequency varies from 1hz to about 20hz. I had looked into lamination but since the original design worked perfectly well and my goal is to reduce size and cost I shelved that idea.

Since the original works well and the new prototype not so much with all other things being equal (coils, power, frequency, materials, etc.) I was focusing on what was different about new design that works against me.

It appears from the responses that none of the questions I posed a few posts back are of any concern, right? It appears the focus should be on the gap, which I can't do much about, but making sure the distance between poles of the stator is set to where they most closely align with the rotors poles. Does the angle of the stator poles matter as long as the gap is still the same?

 

[IRstuff]

"The pairs of coils are wired in series and operate as a single larger coil."

How? Are they opposing poles at the interface or same poles? Are you sure you wired them the same way?

I think your baseplate approach is a non-starter. The magnetic field lines are supposed to travel along the cores to the opposing poles; the flat plate makes the field lines spread out and potentially forces some fields to be emitted from the inactive poles, weakening the field strength.

The angling of the pole pieces matters, since that determines the direction of the field lines and determines the total air gap between the poles.



TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Skogsgurra]

So, it is some kind of stirrer? The kind you see in chem labs? If that is so, the air gap is what it is and not much to do about that.

I think that we all assume that the currents in the coils are phase-shifted. Like sine/cosine. If they are not - the thing will not work. From what are the coils supplied? If it is a transformer for one coil pair plus a capacitor for the other pair, then the coil impedance is important. That also applies if you have an electronic source for the two currents. We have been guessing a lot. Wouldn't it be better to reveal the whole truth? I, for one, does not feel very inspired when we do not know much about what we are guessing on.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[IRstuff]

Op stated that it was a stirrer on the 28th

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[jwells02]

Skogsgurra - As I said, the original core works well so I've have got the phases and timing worked out with the coils. The problem has be the core. Thanks for responding.

IRstuff - "I think your baseplate approach is a non-starter. The magnetic field lines are supposed to travel along the cores to the opposing poles; the flat plate makes the field lines spread out and potentially forces some fields to be emitted from the inactive poles, weakening the field strength."

I think you nailed it.

So if you don't think the coils butt up against each other is also an issue then I think I have enough info to make some changes. Thank you for spending the time to help me out.

 

[Compositepro]

The magnetic properties of steel are variable due to alloying and heat treatment. There are special grades for magnetic applications.

 

[jwells02]

Compositepro - what I'm using works well but you make a great point that I could still be using something better. I'll look into sourcing that. Any recommendations?

 

[djhurayt]

make sure you anneal it after forming/machining to allow the magnetics poles to align.

 

[MagBen]

This is simply a polarities changed magnet underneath the plate that keeps the permanent bar magnet spinning over the plate.
Did you see weakened force, or lower speed?
If lower speed, pay attention to the connection of your 4 coils, you need two pairs of N-S, and the flux should not cancel each other.
To answer some of your Qs:
If less force/torque,
Rounded path is better since this decrease the path length of flux.

No. with that widely open circle, that would make limited difference. same reason, using fully annealed does not make a whole lot of difference. using cheap carbon steel shall be good enough.

 

[jwells02]

MagBen - thanks. Do the coils being against each other cause a problem?

 

[MagBen]

what do you mean by against each other?
a correction: i was talking about magnetic circuit, NOT circle!

 

[MagBen]

by the way, the idea is to create a rotating mangetic field, so the permanent bar magnet will spin to stir.
one can home make such field: use a fan with a magnet sticking on it. When the fan rotates, it creates a so-called rotating field.

 

[electricpete]

Do the coils being against each other cause a problem?

No problem, as long as:
1 - there is sufficient electrical insulation (magnet wire strand insulation might be enough)
2 - there is sufficient cooling.
3 - don't get the iron /steel parts from one core too close to the other. For example if you wound your core onto a steel rod with large steel washers on each end (to help form a guide while you're winding), you don't want the washers coming close to each other because that may "short circuit" the magnetic flux path. Probably better to avoid steel washers on the end altogether though.

=====================================
(2B)+(2B)' ?