Hi all,
I found this on the internet: http://www.geocities.com/k_pullo/SMOT15_PM3-3.htm (http://www.geocities.com/k_pullo/SMOT15_PM3-3.htm)
At least the first motor is actually putting out positive torque in 95% of the samples done (I haven't tested the other one yet). The original LUA script has a resolution of 0,05 degrees in 24 degree measurement. However, there is 16 magnets in the rotor which should give 22,5 degree measurement, but a modification down to 22,5 degrees also shows the same result. Even a 360 degrees measurement, with 0,5625 degree resolution (0,5625 = 9/16 - the number of outer and inner magnets) give 95% positive samples, with an average of 2,5Nm with peaks up to10Nm and dips down to -3Nm. No matter how I calculate the samples I'll get a significant positive result.
Vidar
The link is not working for me, is the link is now down?
Yes Vidar, the site is gone,
would it be possible for you to post what you have on the project here so we can have a look at it?
Greetings
Hans von Lieven
Hi,
avoid to make the mistake to add discrete momentas insted of integrating the functions of tourque and add them.
My opinion without seeing this device: sumM=0.
Why: the peaks contain more area between the lines than the lower values.
See eden project magnetic therory.
Good day
kaRLfunkel
Yet another "working" device and yet another link gone? I think I am starting to see a pattern here. I am not saying MIB but, after dealing with the info on xpenziv's design, this is getting old. for once, I would like to see a design that works, and can be replicated, and, the designer/inventor says..."here is how you build it.' Am I asking for too much?
I don't mean to jump to any conclusions because, maybe the link is bad or a server is down....that happens. I still have hope.
Bill
let's have a look what Vidar has, If he has anything on the device other than a link I am sure he'll share it with us. Until then PATIENCE
Hans
Quote from: Pirate88179 on November 15, 2007, 12:56:13 AM
Yet another "working" device and yet another link gone? I think I am starting to see a pattern here. I am not saying MIB but, after dealing with the info on xpenziv's design, this is getting old. for once, I would like to see a design that works, and can be replicated, and, the designer/inventor says..."here is how you build it.' Am I asking for too much?
I don't mean to jump to any conclusions because, maybe the link is bad or a server is down....that happens. I still have hope.
Bill
As I have already mentioned many, many times, these magnet-to-magnet motors such as the one discussed here (a typical motor of this type) are very difficult to tune up. Mechanically they seem symmetric but their fields, and that's what matters, are not because the seemingly equal in shape magnets have quite differing forms of their magnetic fields. The FEMM simulation is of fields with idealized form which can never be achieved in practice as easily as one may think. Because of the unevenness of the magnetic fields in these seemingly equal in form magnets, cooperative phenomena take place and super poles are formed which destroy the nice pattern seen in the simulation and always bring the rotor to a stop at the overall potential minimum caused by the mentioned cooperative phenomena.
Therefore, as I said many, many times, the way to go is towards constructions whereby magnet-to-magnet interaction causing the above-mentioned cooperative phenomena is avoided and is replaced by magnet-to-steel interaction of the SMOT type. A very creative application of the SMOT principle is found in @xpenzif's rendition of the problem. No wonder why @xpenzif's device is the most promising so far. The magnet-to-magnet motors can also be made to work but the adjustments are Sisyphus labor. Who needs that? Of course, a straightforward practical solution of the problem is by applying brute force as in the Butch-Sprain-Honk type of device. As a scientific experiment, however, so far @xpenzif's device is unbeatable in its simplicity and potential to be reproduced.
I have unfinished work on magnetic motor on my web:
http://free-ri.htnet.hr/Branko/04.html (http://free-ri.htnet.hr/Branko/04.html)
It is just linear movement principle. I didn't finish that work, because of some other investigation of Nikola Tesla's work in energy device:
http://free-ri.htnet.hr/Branko/02.html (http://free-ri.htnet.hr/Branko/02.html)
For now, investigation in 'cosmic ray' energy capturing system is in my mind.
If you have enough magnets, good luck! Left/right mechanical movement is only mechanical problem (and gap in that part).
I must say that link disappeared quite quick. I was initially looking for the FEMM-site in Google, and came across this link. Well, I downloaded the whole thing so here it is:
Anyone can try to run as precise script thay want - there is two suggestions in the RAR-file - you'll every time come out with positive torque. Use Notepad to alter the step values in the LUA-scripts.
The sum will be very high, so remember to devide on number of samples used. I get allways an significant positive average no matter how I look at it.
Vidar
Quote from: Low-Q on November 15, 2007, 03:01:19 AM
Anyone can try to run as precise script thay want - there is two suggestions in the RAR-file - you'll every time come out with positive torque. Use Notepad to alter the step values in the LUA-scripts.
Vidar
Don't rely to FEMM simulation results ! It's far from real world. Though FEMM is the best freeware simulation software available today.
Quote from: Nostradamus2 on November 15, 2007, 03:23:14 AM
Quote from: Low-Q on November 15, 2007, 03:01:19 AM
Anyone can try to run as precise script thay want - there is two suggestions in the RAR-file - you'll every time come out with positive torque. Use Notepad to alter the step values in the LUA-scripts.
Vidar
Don't rely to FEMM simulation results ! It's far from real world. Though FEMM is the best freeware simulation software available today.
You're right, but some excitement doesn't harm anyone :) Just download it and see that it works in FEMM. What ideal "bug" in the software is making this device working? I have no answer except FEMM is not a 3D simulation software. However, to make two magnetic lines interact with eachother, they have to be in the same dimention, so 2D will do for the simulations.
Cheers
Vidar
Thanks Vidar
:-)
Hans
Quote from: karl on November 15, 2007, 12:54:23 AM
Hi,
avoid to make the mistake to add discrete momentas insted of integrating the functions of tourque and add them.
My opinion without seeing this device: sumM=0.
Why: the peaks contain more area between the lines than the lower values.
See eden project magnetic therory.
Good day
kaRLfunkel
I have done this error many times, but are learning - eventually ;D
I use the Area-tool in FEMM to measure the torque. And that function allways measure in the correct direction. Other ways is to make a circle around the rotor and mark it with the Contour-tool, but if you mark the circle in oposite order some times, the measurements will fu..k up.
Note! Using the Area-tool requires that the rotor is allways rotating around the 0,0 point in FEMM - with Contour-tool this is not necessary if you remember to mark the two half parts of the circle in the same order every time.
I have ran the LUA-script on many other projects I have in FEMM, but the result is then allways 0.
To be sure, make measurements for one complete revolution.
Vidar
Quote from: Low-Q on November 15, 2007, 03:45:51 AM
Quote from: Nostradamus2 on November 15, 2007, 03:23:14 AM
Quote from: Low-Q on November 15, 2007, 03:01:19 AM
Anyone can try to run as precise script thay want - there is two suggestions in the RAR-file - you'll every time come out with positive torque. Use Notepad to alter the step values in the LUA-scripts.
Vidar
Don't rely to FEMM simulation results ! It's far from real world. Though FEMM is the best freeware simulation software available today.
You're right, but some excitement doesn't harm anyone :) Just download it and see that it works in FEMM. What ideal "bug" in the software is making this device working? I have no answer except FEMM is not a 3D simulation software. However, to make two magnetic lines interact with eachother, they have to be in the same dimention, so 2D will do for the simulations.
Cheers
Vidar
Simulation has to be done in 3D because the nearest neighbors must not be neglected and not all of them are in the same plane. Try using Maxwell3D. Probably, you may ask @Jdo300 here to help you in that. He's an expert in Maxwell3D.
Quote from: Pirate88179 on November 15, 2007, 12:56:13 AM
I don't mean to jump to any conclusions because, maybe the link is bad or a server is down....that happens. I still have hope.
Bill
The site has just exceeded its free bandwidth (for the day, month?) It should be back when ever this time period is up.