Author Topic: The Thomas Motor (Read 21060 times)

gyulasun · « **Reply #15 on:** March 12, 2014, 11:47:14 AM »

Good Morning Ben,

The smallest OD core available in the 3R1 material is TN14/9/5-3R1, see here http://www.elnamagnetics.com/?s=tn14&type=spec but perhaps it is too small already? Not only small for winding the thin wire onto it but may go into saturation just by the core magnet, not to mention the bigger rotor magnets when in the near vicinity of the core.

Gyula

k4zep · « **Reply #16 on:** March 12, 2014, 12:19:55 PM »

Quote from: gyulasun on March 12, 2014, 11:47:14 AM

Good Morning Ben,

The smallest OD core available in the 3R1 material is TN14/9/5-3R1, see here http://www.elnamagnetics.com/?s=tn14&type=spec but perhaps it is too small already? Not only small for winding the thin wire onto it but may go into saturation just by the core magnet, not to mention the bigger rotor magnets when in the near vicinity of the core.

Gyula

Thanks Gyula,

They are pretty darn close to what I want. What I am using now is a 21mm OD X 13mm ID The saturation of the core by the core magnet and the rotor magnet is a physical problem.....So many variables here. With the core I am using now, I get about 600 gauss leakage (Gauss through the core) when touching the rotor magnet and 0 gause leakage through the core with 1/8"spacing which I am using right now. That could be brought closer to optomize the demo motor but no need to do it right now. With my core magnet, I am getting about 140 gauss leakage through which means I am starting to saturate it but room there for optomization too. But building this motor was a learning experience to me and getting up to speed on switching magnetics! I think my cores are very hi u material around 10,000. I don't remember for sure. That's what you get for not keeping good notes from 3 years ago. That 3r1 material is much lower u but has a hell of a square switching curve. If you can get right up to the knee, would take very little energy to switch. Hummmmmm. Anyway, thanks again for all the tips.
Will call the place today and get a quote on some cores.

Ben

k4zep · « **Reply #17 on:** March 12, 2014, 01:19:55 PM »

Quote from: k4zep on March 12, 2014, 12:19:55 PM

Thanks Gyula,

They are pretty darn close to what I want. What I am using now is a 21mm OD X 13mm ID The saturation of the core by the core magnet and the rotor magnet is a physical problem.....So many variables here. With the core I am using now, I get about 600 gauss leakage (Gauss through the core) when touching the rotor magnet and 0 gause leakage through the core with 1/8"spacing which I am using right now. That could be brought closer to optomize the demo motor but no need to do it right now. With my core magnet, I am getting about 140 gauss leakage through which means I am starting to saturate it but room there for optomization too. But building this motor was a learning experience to me and getting up to speed on switching magnetics! I think my cores are very hi u material around 10,000. I don't remember for sure. That's what you get for not keeping good notes from 3 years ago. That 3r1 material is much lower u but has a hell of a square switching curve. If you can get right up to the knee, would take very little energy to switch. Hummmmmm. Anyway, thanks again for all the tips.
Will call the place today and get a quote on some cores.

Ben

Gyula,

Found the packing slip in the bottom of the bin where I keep the cores. The cores I am using is a Fair-Rite core, PN1N014PD Code 1N017. As seen here:
http://www.westfloridacomponents.com/IN017/Ferrite+Core+ID%3A+13mm+OD%3A+21mm+Fair-Rite.html, of course that is the only one there that
has no Fair-Rite catalog number.
I downloaded the Fair-Rite catalog from ELNA magnetics but can't find the cross to it to get the Spec. sheet, I'm pretty sure u of 10,000. Have a request into
Fair-Rite for help.

Ben

gyulasun · « **Reply #18 on:** March 12, 2014, 03:34:30 PM »

Hi Ben,

Perhaps if you have an L meter you could check your core (with the core magnet attached as in the motor setup) as follows: just wind 10 turns of a probe coil onto the top of the original winding, and check its inductance when the original coil is floating. Then excite slowly the original coil from your power supply and see how the inductance of the 10 turn coil reduces as the core goes towards saturation.
To approach the 160mA current draw from 19V DC input (if that is what you meant) then adjust that current value with a series resistor of a 120 Ohm (19V/160mA), though you may have to consider the peak current value and reduce the resistor accordingly. A series resistor is a must though because otherwise the supply's internal resistance would short the original coil so the 10 turn probe coil could not be measured correctly (unless the power suply has a 'current source' mode). This way you could figure out from the percentage reducements of 10 turn coil's inductance values in the function of the input voltage or current that at what input level the original coil is able to saturate the core, it would be interesting to explore the 11-12V voltage area.

I found a Fair-Rite core with your OD/ID/h sizes made from material #75, u=5000 Part Number: 5975000601 but in the u=10,000 (material #76) there was not any such sized toroid, (perhaps it is not produced as standard size any more?) The 21x13x6.4 is produced with a u=2000 and less too.
If your core has indeed u=5000 only, then its A_L value is 2950nH so 10 turns on it should give 10x10x2950nH=295uH inductance if you have an L meter.

Gyula

k4zep · « **Reply #19 on:** March 12, 2014, 04:13:43 PM »

Quote from: gyulasun on March 12, 2014, 03:34:30 PM

Hi Ben,

Perhaps if you have an L meter you could check your core (with the core magnet attached as in the motor setup) as follows: just wind 10 turns of a probe coil onto the top of the original winding, and check its inductance when the original coil is floating. Then excite slowly the original coil from your power supply and see how the inductance of the 10 turn coil reduces as the core goes towards saturation.
To approach the 160mA current draw from 19V DC input (if that is what you meant) then adjust that current value with a series resistor of a 120 Ohm (19V/160mA), though you may have to consider the peak current value and reduce the resistor accordingly. A series resistor is a must though because otherwise the supply's internal resistance would short the original coil so the 10 turn probe coil could not be measured correctly (unless the power suply has a 'current source' mode). This way you could figure out from the percentage reducements of 10 turn coil's inductance values in the function of the input voltage or current that at what input level the original coil is able to saturate the core, it would be interesting to explore the 11-12V voltage area.

I found a Fair-Rite core with your OD/ID/h sizes made from material #75, u=5000 Part Number: 5975000601 but in the u=10,000 (material #76) there was not any such sized toroid, (perhaps it is not produced as standard size any more?) The 21x13x6.4 is produced with a u=2000 and less too.
If your core has indeed u=5000 only, then its A_L value is 2950nH so 10 turns on it should give 10x10x2950nH=295uH inductance if you have an L meter.

Gyula

Hi Gyula,-
With 10 turns over plain core, it is 54.63uH, therefore I suspect it is a lot less than u=5000 , possibly u=750 with a small magnet on the core, it drops about 5uh to 49uh, within 1/8" of the large magnet it drops to about 4uH....Big magnet on rotor is definitely saturating the core! I have to wind another coil as I don't want to lift the magnets off the video motor as it might break the wires as they are glued down. I'll play with the constant current power supply and see how much the big coil does around 12VDC. Thanks for the input. I'll get back to you.

Ben

gyulasun · « **Reply #20 on:** March 12, 2014, 11:22:41 PM »

Hi TinselKoala,

Thanks for the video links on your earlier Orbo (oops Orbette...) builds and tests. I went through on them and also on some earlier posts of yours when the Steorn-craze raged and I believe you must have gained a huge amount of knowledge on such setups. If you do not mind to share some chips like which is the best core material to use for instance? Is it having the square loop B-H curve or it is not a must?
[Somewhere in one of your pictures I saw 'W' material which has a u=10,000 (here is a link to it if that is indeed the core you used: http://www.mag-inc.com/products/ferrite-cores/w-material ) or you tested other core material types?]

Another question would be, could you recall what efficiency you actually measured, with the methods you have mentioned? Have you attempted to recover the collapsing field energy of the core after the current switch-off and is it included in the efficiency estimations (or that energy is so small that no worth to collect?).

Thanks, Gyula

P

synchro1 · « **Reply #21 on:** March 13, 2014, 05:28:26 AM »

@K4zep,

Really cool project. I wrapped a spiral magnetic haywire coil with magnet wire, so it looked like a wound up spring when finished. I stuck two tube magnets to each end in attraction. When I charged the magnet wire, both magnets released and dropped off. The magnets were attracted both to the magnetic haywire spiral core and each other.

Now imagine two magnet rotors, north poles on one and south poles on the other with the spiral spring coil in between. The rotor magnets on each rotor will be attracted to the magnetic wire core and to each other as well. When they're at TDC, the coil charges, masking both the magnetic material and the attraction between the rotor magnets. One pulse driving both sets of rotors!

Furthermore, a ferrite toroid, standing on end could handle four such rotors, one set for the top and the other two for the bottom with the result that one pulse would drive all four rotors. Both rotor sets attracted to the ferrite, then masked to both the magnetic ferrite and their mutual magnetic attraction by one pulse!

k4zep · « **Reply #22 on:** March 13, 2014, 05:42:04 AM »

Quote from: synchro1 on March 13, 2014, 05:28:26 AM

@K4zep,

Really cool project. I wrapped a spiral magnetic haywire coil with magnet wire, so it looked like a wound up spring when finished. I stuck two tube magnets to each end in attraction. When I charged the magnet wire, both magnets released and dropped off. The magnets were attracted both to the magnetic haywire spiral core and each other.

Now imagine two magnet rotors, north poles on one and south poles on the other with the spiral spring coil in between. The rotor magnets on each rotor will be attracted to the magnetic wire core and to each other as well. When they're at TDC, the coil charges, masking both the magnetic material and the attraction between the rotor magnets. One pulse driving both sets of rotors!

Furthermore, a ferrite toroid, standing on end could handle four such rotors, one set for the top and the other two for the bottom with the result that one pulse would drive all four rotors. Both rotor sets attracted to the ferrite, then masked to both the magnetic ferrite and their mutual magnetic attraction by one pulse!

Sounds good to me, do you have any pictures to show the basic motor? Drawings?

Ben

synchro1 · « **Reply #23 on:** March 13, 2014, 03:45:58 PM »

@Ben,

No, I just dreamed it up thinking about your project.

k4zep · « **Reply #24 on:** March 13, 2014, 04:55:19 PM »

Well, build it and "they" will come! What you got to loose

Ben

synchro1 · « **Reply #25 on:** March 13, 2014, 05:27:05 PM »

@Ben,

That would help keep your shop from capsizing!

k4zep · « **Reply #26 on:** March 16, 2014, 08:33:21 PM »

Hi Gang,

You might want to watch my latest video of my miniature Thomas motor.
Same configuration, very quiet, very fast. Just a step towards my next
totally different motor, hope it is very fast, very strong, and most of all
interesting!

Yes I know, I could build it in plexiglass, make it pretty but I just happen to have
a bunch of little wooden boxes around that I want to get rid of and using them for the bases for the
motor are ok for now. Plastic will come later! Mechanics are from a $10.00 R/C motor that I gutted
to get the rotor and bearings. For now its a lot easier than turning all that stuff on a lathe.

http://www.youtube.com/watch?v=Su-v45BsFpI

Respectfully
Ben K4ZEP

gyulasun · « **Reply #27 on:** March 16, 2014, 11:24:32 PM »

Hi Ben,

Very cool, I like it and indeed no need for fancy plaxiglass in this experimental phase.

I think that not winding the toroidal core fully up with the coil (but to about 80 to 90% only) helps the operational principle be valid more thoroughly than filling up the full winding area 100%. If I can make out properly, you use a multilayer coil on both cores, maybe forced to do so to help core saturation happen at as a relatively low input power level as possible.
I wonder how the current draw increases in this setup when you apply a mechanical load to the rotor. The prop you are going to obtain will surely make this manifest.

Thanks for showing it!

rgds, Gyula

k4zep · « **Reply #28 on:** March 30, 2014, 04:29:41 AM »

Hi All,

http://youtu.be/PklYUbv_dqw

New version of motor, lots of problems mostly with super strong ball magnet! But waveform is absolutely clean. Something is happening in the core/coil with heating and I think it has to do with the ball rotor strength. Notice large distances between core/hall effect, etc..........lots to learn here......

Also working on Wesley W. Gary magnetic motor. Patent as written will NOT work. Don't know if it is intentional but I suspect so. Fields are reversed in the patent from what they actually are at one point.

Respectfully

Ben K4ZEP

k4zep · « **Reply #29 on:** March 31, 2014, 07:52:00 PM »

Quote from: k4zep on March 30, 2014, 04:29:41 AM

Hi All,

http://youtu.be/PklYUbv_dqw

New version of motor, lots of problems mostly with super strong ball magnet! But waveform is absolutely clean. Something is happening in the core/coil with heating and I think it has to do with the ball rotor strength. Notice large distances between core/hall effect, etc..........lots to learn here......

Also working on Wesley W. Gary magnetic motor. Patent as written will NOT work. Don't know if it is intentional but I suspect so. Fields are reversed in the patent from what they actually are at one point.

Respectfully

Ben K4ZEP

Coil had heated/shorted out........
Must make new coils.

Ben K4zep

News:

User Menu

Author Topic: The Thomas Motor (Read 21060 times)

gyulasun

Re: The Thomas Motor

k4zep

Re: The Thomas Motor

k4zep

Re: The Thomas Motor

gyulasun

Re: The Thomas Motor

k4zep

Re: The Thomas Motor

gyulasun

Re: The Thomas Motor

synchro1

Re: The Thomas Motor

k4zep

Re: The Thomas Motor

synchro1

Re: The Thomas Motor

k4zep

Re: The Thomas Motor

synchro1

Re: The Thomas Motor

k4zep

Re: The Thomas Motor

gyulasun

Re: The Thomas Motor

k4zep

Re: The Thomas Motor next step

k4zep

Re: The Thomas Motor next step