top

WhipMag JavaScript applet

Post new topic Reply to topic FizzX.org Forum Index | WhipMag Discussion/Development Goto page Previous  1, 2, 3, 4, 5  Next   Page 3 of 5

Wed Nov 05, 2008 5:50 pm PostPost subject:
Harvey
Major Contributor
Major Contributor


Joined: 16 Jan 2008
Posts: 1927

Reply with quote

(Moved from LC thread)

@korkscrew,

Does FEMM provide 3D force vectors? Or is it 2D only?

I was thinking of redoing the WhiPMag in Vizimag and outputing the force table but it is only a 2D slice. Perhaps we could do a series of layers to assemble an MRI type matrix. Just thinking more about this.

Also, after I got your model to run like Al's rig I never heard any more about that. Was it a glitch in your code or something unexpected?
Back to top
View user's profile Send private message
 
Wed Nov 05, 2008 6:07 pm PostPost subject:
Harvey
Major Contributor
Major Contributor


Joined: 16 Jan 2008
Posts: 1927

Reply with quote

@korkscrew,

Somewhere in my stuff here I have a program named MathCad. It will do eigenvalues and does have a solver. But IIRC I was disapointed with all the stuff it needed placed in order to solve.

Personally I would like to see your model running with the FEMM table replaced with calculations. Unfortunately, I think we would be reinventing the wheel - iow, FEMM and Vizimag already use those algorithms to produce their tables.

I think I could write a small program to take slices (by shrinking the rotor magnet width) in vizimag. Then if we could find a way to reassemble them in a 3D space for visualization we may really have something to look at. Essentially the density lines would follow a normalized curve on the Z axis between layers. With the new transparency features of directx we could give each density a true gradient with each boundry as a different color and an overall transparecy from opaque at the surface of the magnets to clear beyond the stators.

Food for thought.

Cool
Back to top
View user's profile Send private message
 
Wed Nov 05, 2008 6:18 pm PostPost subject:
korkskrew
Regular Contributor
Regular Contributor


Joined: 25 Jan 2008
Posts: 106
Location: Longmont, CO

Reply with quote

(also moved from LC thread)

FEMM only does 2D. 3D costs big bucks.

Before I did my recent analysis, I would have said that I thought my model had a bug that caused it to accelerate. Now though I'm not so sure.

What I do know is that FEMM definitely has a positive bias when it calculated torque. Also, the finite element calculation method produces a small amount of "noise". Both are easily dealt with but annoying none the less.

As for my recent uncertainty, I've been messing with modeling the non-linear parts of this system. It's really difficult, and I doubt I'll get it done this year, but I've seen some very strange things.

You know that average torque chart I recently posted in the reports from the front thread? Well, the non-linear behavior of the system causes the lines on that chart to twist and shift in ways I have yet to understand. It looks like I will be able to obtain Matlab soon. That should help figure it out, but it will still take time. (It's an annealing algorithm to find the minimum in a 3 dimensional non-linear energy space. Yikes! I honestly don't know if I can do it. I may have to cave and finally build my own rig, but it would bear little resemblance to the rigs you guys are building.)

My point is that for the first time, I'm thinking that maybe the secret IS in the steady state torque charts. That would be almost as surprising as the anomaly itself.
Back to top
View user's profile Send private message Send e-mail
 
Wed Nov 05, 2008 6:34 pm PostPost subject:
korkskrew
Regular Contributor
Regular Contributor


Joined: 25 Jan 2008
Posts: 106
Location: Longmont, CO

Reply with quote

Harvey wrote:
I think I could write a small program to take slices (by shrinking the rotor magnet width) in vizimag. Then if we could find a way to reassemble them in a 3D space for visualization we may really have something to look at. Essentially the density lines would follow a normalized curve on the Z axis between layers. With the new transparency features of directx we could give each density a true gradient with each boundry as a different color and an overall transparecy from opaque at the surface of the magnets to clear beyond the stators.


I understand what you have in mind, but I'm not sure it would work. The interactions in 3D space would be at least as complex as those in 2D, so at best you would end up with an aproximation that may or may not represent reality better than the 2D.

I guess I'm not really shooting for an exact model of the Whipmag. I gave up on that when I priced the 3D finite element anaysis packages. Instead I'm focusing on a generalized model and looking at how the forces interact with the MOI of the rotor and stator (and to a lesser extent the friction and drag).

The goal of my modeling is to try to guide myself to a physical model that will be more useful in perfoming anayses of the system than what most replicators are building. The more questions you ask, the more pointed your questions become. I want to build something that will enable me to make measurements and investigate the interactions quantitatively.

I agree that tables are a lot less useful than a continuous function. I don't have the resources to try and figure out that function though.
Back to top
View user's profile Send private message Send e-mail
 
Thu Nov 20, 2008 8:19 pm PostPost subject:
Harvey
Major Contributor
Major Contributor


Joined: 16 Jan 2008
Posts: 1927

Reply with quote

I hope sometime before year end I can help find an algorithm you can plug in rather than using the tables.

I have another question that I asked over here:
http://www.fizzx.com/viewtopic.php?p=7253#7253

Thanx,

Cool
Back to top
View user's profile Send private message
 
Sat Nov 22, 2008 6:05 pm PostPost subject:
korkskrew
Regular Contributor
Regular Contributor


Joined: 25 Jan 2008
Posts: 106
Location: Longmont, CO

Reply with quote

Harvey wrote:
I hope sometime before year end I can help find an algorithm you can plug in rather than using the tables.


Good luck Harvey (for real!). I spent some time last week, while I was in Israel, trying to do it with matlab to no avail. Sad

Harvey wrote:
I have another question that I asked over here:
http://www.fizzx.com/viewtopic.php?p=7253#7253


My javascript model does not include the effects of "stationary" stators. I don't call them idlers because it it seems a misnomer to me. I tend to regard spinning stator (both GW and AGW) as "idling" so the term tends to confuse me. It also does not include the effects of the MKJD's because I have no idea how to model those.

It would not be very difficult to add the effects of stationary stators to the model. Basically you would add a couple of functions that would get their torques from the same table as the AGW stator does, but it would just use a single stator angle and an angular offset for the rotor.

******

I had a lot of down time last week on my trip and spent most of it (when I wasn't e-chatting with my wife) working with the non-linear analysis and came up with some very startling results. I used an excel spreadsheet to do a similar model using my hi-res torque charts.

Guess what? Oh come on, you know what's coming...

The damn thing accelerates! So that eliminates math errors in the javascript model. And it was a simple matter of making sure that there were no biases in the torque tables using excel. So at this point I have to assume this is real.

But the real stunner was that the positive residual torque does not vary with phase offset (at least not from +20 to -20 degrees). It is simply a function of the non-linear speed relationship between the rotor and stator. As the thing speeds up the relationship approaches a more linear one and the positive residual torque diminishes until it matches the friction and drag losses where the system stabilizes.

I'm frustrated because I've been working on this for a year to prove an external energy source is required. I ended up proving (at least to myself) that it's a bloody perpetual motion machine which is, of course, absurd.

So here is what I think is important based on all my work so far.
- The higher the forces the better. Stronger / bigger magnets should produce a larger result. Magic magnets are not needed.
- The MOI relationship between the rotor and stator is a big deal. The MOI of the stator should be a very small fraction of the rotor MOI.
- The residual torque is pretty small compared to the force of the magnets so minimizing friction and drag is a pretty big deal.
- Magnet shape matters but I have yet to figure out what is optimal. The cylindrical diametrically magnetized stator really enhances the residual torque. I am trying to figure out the optimal rotor magnet shape (more of those week long simulations).

Things I can't explain:
- Why have those replications with large differences in MOI and low friction not seen a measureable difference in spin down between GW and AGW? (edit) Misstated: Should say "no stator vs AGW stator"
- Why would only some of Al's rotors and stators work?
- (edit) Why would Al's rig have two AGW modes (one that accelerates and one that doesn't)?

All my modeling suggests that a measureable effect should be easy to produce as long as there is AGW sync and the stator MOI is much less than the rotor MOI.

Now I have to do what I have been trying all this time to avoid. I have to build one.

I will not be building the same thing you guys are. I will make one with a 1 to 1 polar relationship instead of the 4 to 1 that you guys are doing. I will be making it with large magnets (like 3/4 inch N52 neos) to minimize how much I have to worry about friction, but I will also turn the thing on its side to put the load on the bearings in the direction they are designed to give the least friction. I will be using tiny bearings too (1/4 inch OD) and I will probably use the lube cleaning trick you guys have been discussing.

Iíll provide more info as I get the design further along. It will probably take a while to build because Iíll have to get the machining done by someone so it will be constrained by available funds.

Bother!


Last edited by korkskrew on Sat Nov 22, 2008 9:11 pm; edited 2 times in total
Back to top
View user's profile Send private message Send e-mail
 
Sat Nov 22, 2008 6:24 pm PostPost subject:
lostcauses
Major Contributor
Major Contributor


Joined: 02 May 2008
Posts: 871
Location: NM

Reply with quote

"It is simply a function of the non-linear speed relationship between the rotor and stator."

Agreed on this one, LOL Yet I suspect the small gain from the AGw is not in reality enough to cause this to run as we have seen it. This is due to models don't get all them variables as bearing clearance causing vibration loss, heating causing friction loss etc.
Yet it is the differential of the velocities in this thing are the cause and effect of it.

I am suspecting a bouncing ball effect from this on them Gw things also. it is how I found an area that such may occur in. It does exist, and so far I suspect it is the cause of both GW stator idler variations in speed and even the abnormal ratios of turn to rotor.
Back to top
View user's profile Send private message
 
Sat Nov 22, 2008 6:39 pm PostPost subject:
overconfident
Major Contributor
Major Contributor


Joined: 10 Feb 2007
Posts: 1121

Reply with quote

korkskrew wrote:

So here is what I think is important based on all my work so far.
- The higher the forces the better. Stronger / bigger magnets should produce a larger result. Magic magnets are not needed.


I think it's more the field intensity vs. weight relationship and that size has very little to do with it.

korkskrew wrote:

- The MOI relationship between the rotor and stator is a big deal. The MOI of the stator should be a very small fraction of the rotor MOI.


Been saying this from day one, it's in the transcript.

korkskrew wrote:

- The residual torque is pretty small compared to the force of the magnets so minimizing friction and drag is a pretty big deal.


Magnetic drag plays a part as well.

korkskrew wrote:

- Magnet shape matters but I have yet to figure out what is optimal. The cylindrical diametrically magnetized stator really enhances the residual torque. I am trying to figure out the optimal rotor magnet shape (more of those week long simulations).


I've been puzzling this one too, thoughts of pole pieces to shape the field. I have tried using 1/4" ball bearings as pole pieces and I do get slightly longer rundowns with them, just a couple seconds but it is consistent. Not sure though, might just be the additional weight.

korkskrew wrote:

Things I can't explain:
- Why have those replications with large differences in MOI and low friction not seen a measureable difference in spin down between GW and AGW?


I get radically increased spindown times in AGW compared to GW, at least with certain parameters. I'm afraid I don't have logs to back it up. I'd have to redo the experiments.

korkskrew wrote:

- Why would only some of Al's rotors and stators work?


That's a BIG unknown!

korkskrew wrote:

All my modeling suggests that a measureable effect should be easy to produce as long as there is AGW sync and the stator MOI is much less than the rotor MOI.


Something I have said all along and the reason I tried with a stator on a dowel, without the additional mass of stator housing or bearing. It's much more difficult to spin that way, though. I could never get it to sync.

korkskrew wrote:

I will be using tiny bearings too (1/4 inch OD) and I will probably use the lube cleaning trick you guys have been discussing.


I think nonmagnetic bearings might be better. The ferromagnetic ones Al is using might be introducing some asymmetrical field distortions. It might be what's so special about the stators that work.
Back to top
View user's profile Send private message
 
Sat Nov 22, 2008 7:04 pm PostPost subject:
lostcauses
Major Contributor
Major Contributor


Joined: 02 May 2008
Posts: 871
Location: NM

Reply with quote

"Magnet shape matters"

I suspect from what i am doing this would be an angled face to a curved body. An strange design, have not seen such and suspect it would be a problem to make on.

Like this. Angles curved...

/ curved, center down, curved \
Back to top
View user's profile Send private message
 
Sat Nov 22, 2008 7:13 pm PostPost subject:
overconfident
Major Contributor
Major Contributor


Joined: 10 Feb 2007
Posts: 1121

Reply with quote

lostcauses wrote:
"Magnet shape matters"

I suspect from what i am doing this would be an angled face to a curved body. An strange design, have not seen such and suspect it would be a problem to make on.

Like this. Angles curved...

/ curved, center down, curved \


Pole pieces!
Back to top
View user's profile Send private message
 
Sat Nov 22, 2008 8:59 pm PostPost subject:
korkskrew
Regular Contributor
Regular Contributor


Joined: 25 Jan 2008
Posts: 106
Location: Longmont, CO

Reply with quote

overconfident wrote:
I get radically increased spindown times in AGW compared to GW, at least with certain parameters. I'm afraid I don't have logs to back it up. I'd have to redo the experiments.

I'm refering to the spin down time of no stator vs the spindown time with and AGW stator. Clanzer's measurements (the only ones I can remember specifically, but I'm sure I've seen others including Harvey's) show very similar run-downs there.

The reason there is such a large difference in run-down between GW and AGW is that most of the chaotic behavior in GW mode ends up multiplying the frictional losses to the stator.

My work suggests that even if the system has too much friction to spin up, there should be an easily measured difference in the spin-down time of no stator vs AGW stator. All of the measurements to date suggest very similar run-down times.

This is inconsistent with my model and therefore indicates that my model is invalid. But as I've said before, I can't for the life of me figure out what I'm doing wrong. I'm squarely in Al's camp in terms of perpetual motion machines being impossible. It's just that I've put a lot into making sure these models are accurate, and still they spin up (unlike the real devices other than Al's).

When a math model differs from observed reality it's pretty clear which is wrong. It doesn't mean the math model isn't useful, it just means it isn't modeling the whole. By making the math model I know what I've put in. Now I build a real model that enables me to measure the various things going on in the math model, and I can see what's different.

I have no expectation that what I build will spin up. But I want to know where I went wrong in the model. I'm terribly curious about how the energy is getting into Al's system.
Back to top
View user's profile Send private message Send e-mail
 
Sat Nov 22, 2008 9:32 pm PostPost subject:
korkskrew
Regular Contributor
Regular Contributor


Joined: 25 Jan 2008
Posts: 106
Location: Longmont, CO

Reply with quote

One other little curiosity.

In the linear model the PE minimum is at zero degrees phase offset. In the nonlinear model, not only is the PE minimum quite a bit deeper, but it shifts a few degrees negative (stator lagging rotor).

BTW, I'm sorry I'm contradicting myself so much today. I'm still a bit jet lagged and not thinking very clearly. I was up for 28 hours yesterday, and then slept nearly 12 hours last night...it just has me a little messed up. I proabably should have waited before I said anything about the nonlinear analysis.

(edit) Another unexpected result...phase offsets from about -19 to +16 degrees produce a net positive torque on BOTH the rotor and stator. That is to say, in the observed operational state, one is not "powering" the other. They are both "producing" energy.


Last edited by korkskrew on Sat Nov 22, 2008 9:46 pm; edited 1 time in total
Back to top
View user's profile Send private message Send e-mail
 
Sat Nov 22, 2008 9:45 pm PostPost subject:
lostcauses
Major Contributor
Major Contributor


Joined: 02 May 2008
Posts: 871
Location: NM

Reply with quote

I suspect you have some control in timing of your modle that does not exist in reality.

such as your saying the rotor need the mass with the stator as little.

Yet it is such if the stator does not have enough mass, it will be pushed and pulled by the rotor.
same for the rotor, if to much mass, it will not allow a transfer to motion, and two little well it will also be caught in the push and pull of the stator.

As you say folks get easy AGW lock, yet no real action on the rotor. question is does the rotor or idler have to little or to much mass??

LOL it is hard to speed up a train in motion up as well as slow it down.
It is a balance of the magnet strength to the rotor mass that it working with this. Some control is du to the MKJD of Al's I believe. It not much and again a push pull effect on the mass of the rotor, as well as possibly effecting the stator.

What can I say, I had to go back over every thing, rethink and then step back and stop looking at the individual tree and take a look at the forest. Some of this then makes for reason.
Hmm On the perpetual thing, So far I have not heard of Al,s device running forever.
It appears to be motion causing an offset of the normal equalization of the magnets. Simply put a pendulum. some were it will reach a velocity that causes this unbalance to fall out. a pendulum shorted its throw each swing unless force is added. the whip mag runs out, and has to be reset... Think of it as a very low drag pendulum. It would last a very long time, but still runs out of the unequal motion to force. You have already said this thing does such.
Back to top
View user's profile Send private message
 
Sat Nov 22, 2008 9:50 pm PostPost subject:
korkskrew
Regular Contributor
Regular Contributor


Joined: 25 Jan 2008
Posts: 106
Location: Longmont, CO

Reply with quote

@lostcauses

When I say it's a perpetual motion machine, I'm refering to my model.
Back to top
View user's profile Send private message Send e-mail
 
Sat Nov 22, 2008 10:03 pm PostPost subject:
lostcauses
Major Contributor
Major Contributor


Joined: 02 May 2008
Posts: 871
Location: NM

Reply with quote

"I'm refering to my model."

LOL it may more to a digital to analog thing happening. It uses compartments per say it is setting up the next frame.
Old game of the tortoise and the hare game..
Back to top
View user's profile Send private message
 
Sat Nov 22, 2008 10:22 pm PostPost subject:
overconfident
Major Contributor
Major Contributor


Joined: 10 Feb 2007
Posts: 1121

Reply with quote

@korkskrew

I timed my rotor with no stators and got repeated spin downs of 4:52 to 4:58 seconds from 500 rpm. With GW sync, it was less than 2 minutes. With AGW sync it was slightly less than without any stators. But when I placed 1/4" steel balls on each pole of the rotor magnets, I achieved AGW spin downs of 5:04 to 5:07. I don't know if it was due to the additional mass or some reduced friction or what.

Do you know anyone who is capable of checking your model and math? Or do you know any way to verify the tables (FEMM?), cross check them for validity?

Controlling the phase angles of a real device will be much more difficult than doing it in a sim, just ask Cloud Camper. Wink

ETA: Maybe you should provide Cloud Camper with some phase offsets he can plug into his rig, see if things behave anything like your predictions?
Back to top
View user's profile Send private message
 
Sat Nov 22, 2008 10:55 pm PostPost subject:
lostcauses
Major Contributor
Major Contributor


Joined: 02 May 2008
Posts: 871
Location: NM

Reply with quote

"But when I placed 1/4" steel balls on each pole of the rotor magnets, I achieved AGW spin downs of 5:04 to 5:07. I don't know if it was due to the additional mass or some reduced friction or what."

Field change and mass.
Back to top
View user's profile Send private message
 
Sun Nov 23, 2008 5:20 pm PostPost subject:
korkskrew
Regular Contributor
Regular Contributor


Joined: 25 Jan 2008
Posts: 106
Location: Longmont, CO

Reply with quote

overconfident wrote:
I timed my rotor with no stators and got repeated spin downs of 4:52 to 4:58 seconds from 500 rpm. With GW sync, it was less than 2 minutes. With AGW sync it was slightly less than without any stators.

AFAIK, those results are similar to most of the replicators. As I said...very different from my model so my model is f****d up. But I haven't the faintest idea how.
overconfident wrote:
But when I placed 1/4" steel balls on each pole of the rotor magnets, I achieved AGW spin downs of 5:04 to 5:07. I don't know if it was due to the additional mass or some reduced friction or what.

I gotta go with lostcauses on this one. It's looks like it's just a result of the added mass. More mass means that at a specific speed it has more energy.
overconfident wrote:
Do you know anyone who is capable of checking your model and math? Or do you know any way to verify the tables (FEMM?), cross check them for validity?

If I'm not mistaken, Harvey is putting some effort into this. The tables are generated from FEMM simulations and then corrected for tiny positive offsets using Excel. I'm sure they are flawed, but I'm also sure the general morphology of the PE (torque) surfaces is right. The rest is just F=MA. Not exactly chalenging math.
overconfident wrote:
Controlling the phase angles of a real device will be much more difficult than doing it in a sim, just ask Cloud Camper. Wink

Cloud Campers rig is good at controlling things like this, but it isn't very good at measuring tiny residuals. He and I have discussed this a little and decided there is not much to be gained. Actually sophisticated angular control is not the problem at all. The angles pretty much take care of themselves. The question I can't answer is why in my model the naturaly ocurring angular relationships result in residual positive torque, but in the replications it doesn't.
overconfident wrote:
ETA: Maybe you should provide Cloud Camper with some phase offsets he can plug into his rig, see if things behave anything like your predictions?
He asked for a simulation from me at one point (after my recent post to "Reports from the Front"), but I haven't heard from him since.
Back to top
View user's profile Send private message Send e-mail
 
Sun Nov 23, 2008 5:44 pm PostPost subject:
overconfident
Major Contributor
Major Contributor


Joined: 10 Feb 2007
Posts: 1121

Reply with quote

korkskrew wrote:
...very different from my model so my model is f****d up.


Maybe not so f****d up if you consider Al's device. Wink
Back to top
View user's profile Send private message
 
Sun Nov 23, 2008 6:22 pm PostPost subject:
lostcauses
Major Contributor
Major Contributor


Joined: 02 May 2008
Posts: 871
Location: NM

Reply with quote

Hm I wounder if it is the difference in friction on them rotor bearing that folks do see that your sim is not???

I keep going over all them other things looking for a possible gain some were in them (idlers, mkjd), and do not find much. If this is not what they do, could they be reducing drag by controlling the friction on the rotor bearings.

In other words I suspect your sim has a smother friction action than the torque being placed on the bearings in the sim.

Oh you should at least in the magnetic torque, see a gain.

I hope this may show you the difference with sim over real.
Back to top
View user's profile Send private message
 
Sun Nov 23, 2008 8:06 pm PostPost subject:
Harvey
Major Contributor
Major Contributor


Joined: 16 Jan 2008
Posts: 1927

Reply with quote

Quote:

The question I can't answer is why in my model the naturaly ocurring angular relationships result in residual positive torque, but in the replications it doesn't.


Noise? (and other unmapped losses). In Al's rig there is a relationship of losses that conserves some of that energy (iow moves the losses around inside the system). So far in the replications that energy is lost to the environment.

In your model, we don't have harmonic losses, just friction and drag. By reducing friction and drag to compensate, the model produces positive torque Very Happy

We never did pinpoint the x10 issue either did we Confused

Cool
Back to top
View user's profile Send private message
 
Sun Nov 23, 2008 9:55 pm PostPost subject:
korkskrew
Regular Contributor
Regular Contributor


Joined: 25 Jan 2008
Posts: 106
Location: Longmont, CO

Reply with quote

overconfident wrote:
korkskrew wrote:
...very different from my model so my model is f****d up.


Maybe not so f****d up if you consider Al's device. Wink

The source of all our torment...
Back to top
View user's profile Send private message Send e-mail
 
Sun Nov 23, 2008 10:31 pm PostPost subject:
korkskrew
Regular Contributor
Regular Contributor


Joined: 25 Jan 2008
Posts: 106
Location: Longmont, CO

Reply with quote

@Harvey and lost

Ok, so I just deleted an hour of typing. I really can't answer your questions or ideas because I just don't know. The effect I'm seeing is not trivial or near the noise of the system. One simulation (significantly different from Al's rig and the replications) shows 0.003Nm residual torque at perfectly reasonable speeds and MOI's (or is that M'sOI).

I don't think I'm going to work on the models anymore. To me they have served their purpose. I know what I want to measure and how I can measure it. The models are flawed, and I don't know how to un-flaw them. But they have told me what I need to know to build an informative test rig. As I said it will be very different from what you guys have been playing with.
Back to top
View user's profile Send private message Send e-mail
 
Sun Nov 23, 2008 10:38 pm PostPost subject:
lostcauses
Major Contributor
Major Contributor


Joined: 02 May 2008
Posts: 871
Location: NM

Reply with quote

korkskrew
I answered in my thread. It is I can not find were the kinetic transfer from stator to idler can happen. I suspect your sim and reality is were this is the proplem. I can see were the stator will gain in this, but not the rotor. I may just be missing the mass relations a bit.

Hey I know the delete the writing. I wish you would have posted, LOL Understood why you did not.
Back to top
View user's profile Send private message
 
Mon Nov 24, 2008 3:44 pm PostPost subject:
korkskrew
Regular Contributor
Regular Contributor


Joined: 25 Jan 2008
Posts: 106
Location: Longmont, CO

Reply with quote

@Harvey

By the way, I've tripple and quadruple checked all the math to I've pretty much concluded that the 10X problem is due to me mis-estimating the MOI of the stator.

I did the best I could and nobody else will even try...so screw it.

@lostcauses

If you like I can make my excel spreadsheet model publicly available...but I'm saying right now, I'm not available for tech support.
Back to top
View user's profile Send private message Send e-mail
 
Mon Nov 24, 2008 4:06 pm PostPost subject:
overconfident
Major Contributor
Major Contributor


Joined: 10 Feb 2007
Posts: 1121

Reply with quote

@korkskrew,

I believe you said you stripped some of the calculations (friction and such) out of your spreadsheet. Would you be willing to put those back in so we can compare different versions, with and without?

Thanks,
OC
Back to top
View user's profile Send private message
 
Mon Nov 24, 2008 4:10 pm PostPost subject:
lostcauses
Major Contributor
Major Contributor


Joined: 02 May 2008
Posts: 871
Location: NM

Reply with quote

Sounds good to me. I understand the no tech support.

Yet with your sim, we are watching one work in which we are trying to replicate, LOL all info is welcome.

When I get thinking on the idlers, with best guess on them other things, It seems it just takes areas to a balance of close to just motion and friction with them. again it may just be a slight gain due to the Agw effect with it. It seems to reduce the drag time when them idlers (GW) are stopped. In other words the proper offset puts it to the torque is an area of weak drag over the rotor magnets between the two. When stoped it is simply the shear.
The MKJd are switched due to magnet length, etc. It seems at least in Al,s that it is an area of needed action upon rotor to most likely keep the stator in timing.

Best ideas I can come up with all the extras on Al,s..
Back to top
View user's profile Send private message
 
Mon Nov 24, 2008 4:18 pm PostPost subject:
korkskrew
Regular Contributor
Regular Contributor


Joined: 25 Jan 2008
Posts: 106
Location: Longmont, CO

Reply with quote

overconfident wrote:
I believe you said you stripped some of the calculations (friction and such) out of your spreadsheet. Would you be willing to put those back in so we can compare different versions, with and without?


The spreadsheet is designed to examine energy transfer between the rotor and stator. Friction just get's in the way. Instead there are energy constants that you can alter to look at the effects of adding or subtracting energy to/from the system. If you make the constant negative it's like putting friction/drag in the system.

I'll see if I can get the spreadsheet up tonight...or maybe the next night. *shrug* (edit) OK, fine..."soon" then.
Back to top
View user's profile Send private message Send e-mail
 
Mon Nov 24, 2008 4:32 pm PostPost subject:
lostcauses
Major Contributor
Major Contributor


Joined: 02 May 2008
Posts: 871
Location: NM

Reply with quote

Thanks.
Back to top
View user's profile Send private message
 
Mon Nov 24, 2008 8:25 pm PostPost subject:
Harvey
Major Contributor
Major Contributor


Joined: 16 Jan 2008
Posts: 1927

Reply with quote

korkscrew wrote:

I did the best I could and nobody else will even try...so screw it.


I thought I calculated the stator MOI in the physics thread here on FizzX...

http://www.fizzx.com/viewtopic.php?p=4667#4667

They were arbitrary starting numbers but as close as could be expected for the assembly. How did they compare to yours?

EDIT: Those were the calcs I used when I got your model to self run.
Back to top
View user's profile Send private message
 
Post new topic Reply to topic FizzX.org Forum Index | WhipMag Discussion/Development
View previous topic
View next topic
Display posts from previous:   




You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum