Storing Cookies (See : http://ec.europa.eu/ipg/basics/legal/cookies/index_en.htm ) help us to bring you our services at overunity.com . If you use this website and our services you declare yourself okay with using cookies .More Infos here: https://overunity.com/5553/privacy-policy/If you do not agree with storing cookies, please LEAVE this website now. From the 25th of May 2018, every existing user has to accept the GDPR agreement at first login. If a user is unwilling to accept the GDPR, he should email us and request to erase his account. Many thanks for your understanding

Custom Search

### Author Topic: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)  (Read 364259 times)

#### synchro1

• Hero Member
• Posts: 4720
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #120 on: November 20, 2015, 03:56:47 PM »
Hi all

I have tried the 10 ohm scr between the entry of the cap and the assistant coil (X4), the trace is much better but is at no place at zéro volt, so i think perhaps my probes or my scope should be recalibrated, so i stop measuring so small current with my scope to avoid junk datas. If somebody has a better equipment, feel free to follow MH recommendations for the measurement.

So this morning i have tried to get some feeling with my hands, and i decided to place a big ventilator fan on the rotor to significantly increase the mechanical load. So i decreased the voltage to 1.9 volts and i noticed that i had to increase a lot the cap in the flyback circuit (from 0.3 up to 10 uF ) to get the best rotation speed. For info, in this case the " parent " pulse duration in the main coil, is almost the same as the "child" pulse duration in the assistant coil.

So some results on this setup

1- when the main coil and the assistant coil are working together, the input voltage is 1.9 V and the average current is 0.110 A. that is 0.21 Watts and the fan spins at 500 rpm

2- i disconnected completely the assistant coil and flyback circuitery and put away the assistant coil in order to not influence the rotor magnetism.
- i put a freewheeling diode across the main coil and put the power on. 1.9 volts and average current at 0.12 A that is 0.23Watts to get only 450 rpm.
- than same setup and i disconnected the diode so the reed switch is strongly arcing  so 1.9 volt at around 0.12 A , that is 0.23 Watts and only 400 rpm

3- I remounted the flyback circuitery but put away the assistant coil from the rotor - in open magnetic assistant C core = 1.9 V and 0.130 A that is 0.247 Watts with 420 rpm
-and than i magnetically closed the C core and 1.9 V and 0.125 A that is 0.24 Watts and 416 rpm

So it seems that the assistant coil bring an strong torque addition in comparison with the main coil alone and for the same or less input power, and it is what is important to me at this stage.

Just for info

Laurent

@Woopjump,

Another video would be appreciated.

#### gyulasun

• Hero Member
• Posts: 4117
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #121 on: November 20, 2015, 04:03:27 PM »
Hi Laurent,

Thanks for all your efforts and sharing the results.

Gyula

#### gotoluc

• elite_member
• Hero Member
• Posts: 3096
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #122 on: November 20, 2015, 04:28:42 PM »
Hi everyone,

here is a new demo video of the GTL Flux Gate with a better comparative test then my first attempt which I deleted.

The test starts with the scale showing the pull force of 2.6Kg from the permanent magnets which are imbedded in the MOT (GTL Flux Gate)

Test 1: I power only the Low impedance coil (0.4 Ohms) with pure DC to establish a baseline to have a comparative for the 2nd test.
Results are: with an input of 2.20vdc @ 4.3a = 9.5 watts, the low impedance coil can alleviate 2Kg of the PM pull force.

Test 2: I power both the low impedance and high impedance coil connected in series (93 Ohms) with the flyback diode connected across the series coils.
Results are: with the pulse circuit at 235Hz / 50% duty cycle with 60vdc @ 0.151a = 9 watts the high impedance coils can alleviate the same 2Kg of the PM pull force.

The results are not as spectacular as I would of hoped for but how can we explain Test 2 with such a high coil resistance to perform the same task with 1/2 a watt less input by adding components that represents only losses. pulse switch, high impedance coil and diode?

Luc

#### gotoluc

• elite_member
• Hero Member
• Posts: 3096
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #123 on: November 20, 2015, 04:49:56 PM »
@ synchro1 and anyone else wanting to debate inductive-kickback related effects.

please stop posting your debates in this topic. citfta has started a topic for anyone interested to do that:  http://overunity.com/16203/inductive-kickback/msg466238/#msg466238

I'm deleting all the new posts that were added after citfta started the topic relating to this debate.
It's filling up the topic with too many posts that will make it difficult for future replicators.

Luc

#### MileHigh

• Hero Member
• Posts: 7600
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #124 on: November 20, 2015, 05:57:00 PM »
@Milehigh,

Based on your analysis increasing rotor magnet strength would not improve the "Flyback Motor". What effect do you think placing magnets on the auxiliary coil's ferrite U core would have?

The vast majority of experiments where people place magnets in magnetic circuits as part of a pulse motor setup with the belief that they will do something special or add to the output power are simply wrong.  In the majority of cases they don't test their setup with and then without the magnets in place so they never know one way or the other.  You are also subjecting the magnets to a lot of changing flux and for some magnets you will start to demagnetize them.

#### gyulasun

• Hero Member
• Posts: 4117
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #125 on: November 20, 2015, 06:02:15 PM »
Hi everyone,

here is a new demo video of the GTL Flux Gate with a better comparative test then my first attempt which I deleted.

The test starts with the scale showing the pull force of 2.6Kg from the permanent magnets which are imbedded in the MOT (GTL Flux Gate)

Test 1: I power only the Low impedance coil (0.4 Ohms) with pure DC to establish a baseline to have a comparative for the 2nd test.
Results are: with an input of 2.20vdc @ 4.3a = 9.5 watts, the low impedance coil can alleviate 2Kg of the PM pull force.

Test 2: I power both the low impedance and high impedance coil connected in series (93 Ohms) with the flyback diode connected across the series coils.
Results are: with the pulse circuit at 235Hz / 50% duty cycle with 60vdc @ 0.151a = 9 watts the high impedance coils can alleviate the same 2Kg of the PM pull force.

The results are not as spectacular as I would of hoped for but how can we explain Test 2 with such a high coil resistance to perform the same task with 1/2 a watt less input by adding components that represents only losses. pulse switch, high impedance coil and diode?

Luc

Hi Luc,

The only explanation to your question is the much more number of turns for the 93 Ohm MOT coil than the 0.4 Ohm thick wire MOT coil has.  The high number of turns can insure a similar excitation for the core at a lower current level  compared to a coil with low number of turns at a higher input current level.

The 93 Ohm coil resistance dissipates 2.1 W power (0.151A*0.151A*93).
The 0.4 Ohm coil resistance dissipates 7.4 W power (4.3A*4.3A*0.4).
This may seem surprising but shows 2 things: loss in a wire increases with the square of the current and high current can be circumvented by many number of turns in favor of getting less loss in a coil.  Of course in any particular case a trade-off should be chosen on the number of turns and the wire diameter.

Gyula

#### MileHigh

• Hero Member
• Posts: 7600
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #126 on: November 20, 2015, 06:02:19 PM »
Hi all

I have tried the 10 ohm scr between the entry of the cap and the assistant coil (X4), the trace is much better but is at no place at zéro volt, so i think perhaps my probes or my scope should be recalibrated, so i stop measuring so small current with my scope to avoid junk datas. If somebody has a better equipment, feel free to follow MH recommendations for the measurement.

It's entirely possible that when you switch over to the 10-ohm resistor that you have a continuous current flow through the secondary coil.  Therefore there may be no problem at all with your probes and your scope calibration.  If you check the voltage across the 1 uF capacitor you may also notice that it never goes to zero volts.

The explanation for this is that the current pulses from the drive coil keep filling up the 1 uF capacitor with charge so that it never goes down to zero volts.  Therefore there is always current flowing through the secondary coil.

#### gotoluc

• elite_member
• Hero Member
• Posts: 3096
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #127 on: November 20, 2015, 07:08:24 PM »
Hi Luc,

The only explanation to your question is the much more number of turns for the 93 Ohm MOT coil than the 0.4 Ohm thick wire MOT coil has.  The high number of turns can insure a similar excitation for the core at a lower current level  compared to a coil with low number of turns at a higher input current level.

The 93 Ohm coil resistance dissipates 2.1 W power (0.151A*0.151A*93).
The 0.4 Ohm coil resistance dissipates 7.4 W power (4.3A*4.3A*0.4).
This may seem surprising but shows 2 things: loss in a wire increases with the square of the current and high current can be circumvented by many number of turns in favor of getting less loss in a coil.  Of course in any particular case a trade-off should be chosen on the number of turns and the wire diameter.

Gyula

Thanks Gyula for keeping the explanation simple enough for me to understand.

Even though today's tests are not so bad with the GTL Gate, yesterdays tests reveled I could achieve 2Kg. for pull force from a MOT (without PM)  using just the primary with only 3 watts of input. So for now the GTL Gate idea is shelved until something new come.
So I'm going back to my original idea of building a bucking MOT motor with the assisting flyback coil separated from the primary low impedance coil.

Luc

#### Magluvin

• Hero Member
• Posts: 5884
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #128 on: November 20, 2015, 07:12:55 PM »
Running out of time here at lunch so dont have time to find the post by Brad I read earlier to quote.

Brad was talking about 1kv jumping across 1mm gap, suggesting that there wasnt enough to jump the reed at 60v.  What is missing there is the fact that as the reed opens, the tiniest fraction of a mm, the spark starts and continues to flow across the gap as it goes to fully open. Like if you set up a standard relay to buzz, 12v in usually can give 90v out. But while it buzzes, there are sparks across that gap. The initial spark when the contacts just open ionizes the air allowing the spark to continue even when the gap is opened more to its fully open state, then dies out.

Not knocking you brad. Just providing info that I know on the subject.

Mags

#### gyulasun

• Hero Member
• Posts: 4117
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #129 on: November 20, 2015, 08:12:59 PM »
Hi Luc,

In the meantime Hob showed a paper he and his colleague made on just electromagnets to increase their force, see attachment and then in his next post a video :
http://overunity.com/15796/elementary-physics-revisited/msg466245/#msg466245

They introduced the force/power ratio, tested and compared 6 different coils and the conclusion is that

"The amount of copper in an electromagnet determines the force per power
ratio, not the number of turns or the wire thickness in the coil,  the more copper the greater force."

Well, interesting approach, I have not seen such comparison before, and "to my rescue"  I think the amount of copper is also increased when you increase the number of turns...(what I said)  and of course there are other factors to be considered.

Gyula

#### gotoluc

• elite_member
• Hero Member
• Posts: 3096
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #130 on: November 20, 2015, 09:31:35 PM »
Thanks Gyula for bringing Hob's experiment to my attention.

It's all these small details we need to know and combine in or design that may lead us to the goal.

Luc

#### MoRo

• Jr. Member
• Posts: 80
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #131 on: November 21, 2015, 05:35:12 AM »
...  What is missing there is the fact that as the reed opens, the tiniest fraction of a mm, the spark starts and continues to flow across the gap as it goes to fully open. Like if you set up a standard relay to buzz, 12v in usually can give 90v out. But while it buzzes, there are sparks across that gap. The initial spark when the contacts just open ionizes the air allowing the spark to continue even when the gap is opened more to its fully open state, then dies out...

Good points. Newman's book discusses ways to achieve fast opens with less sparking. Maybe try parallel reed switches.

MagnaMoRo

#### tinman

• Hero Member
• Posts: 5365
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #132 on: November 21, 2015, 06:21:22 AM »
Hi Luc,

In the meantime Hob showed a paper he and his colleague made on just electromagnets to increase their force, see attachment and then in his next post a video :
http://overunity.com/15796/elementary-physics-revisited/msg466245/#msg466245

They introduced the force/power ratio, tested and compared 6 different coils and the conclusion is that

"The amount of copper in an electromagnet determines the force per power
ratio, not the number of turns or the wire thickness in the coil,  the more copper the greater force."

Well, interesting approach, I have not seen such comparison before, and "to my rescue"  I think the amount of copper is also increased when you increase the number of turns...(what I said)  and of course there are other factors to be considered.

Gyula

But it is also true that by decreasing the number of turns while increasing the size of the wire,can also result in more copper used. The advantage of this is less resistive losses-ohms law rules.

The ultimate inductor would use square copper wire to reduce the air gaps between windings

I am half way through my Woopy replication-should be up later tonight.

#### Magluvin

• Hero Member
• Posts: 5884
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #133 on: November 21, 2015, 06:31:54 AM »
Good points. Newman's book discusses ways to achieve fast opens with less sparking. Maybe try parallel reed switches.

MagnaMoRo

Hey MoRo

Thanks

That would be series on the reeds I believe. 'If' they open at the same time, the total gap possible is twice the distance as just 1 reed, to fully open in the same time frame. So the total gap gains distance 2 times as fast.  Would probably be best to make a spinning commutator switch in which many opening series contacts open at once, crating a super fast close to fully open situation.    Seems like a Tesla thing. Have to look that up.

The reason for the commutator version is that reeds are not all exactly the same even from the same batch. There will be issues of magnets(of many on a rotor) not having the exact same field strength and or orientation as the others.  best to use tiny ones to avoid pole position discrepancies and just move the weaker ones out further to make up for field strength differences. But I would be way more confident in a well designed commutator with say 8 series switch points than 8 series reeds.  Dont get me wrong. Reads a very good. But they also can only work up to certain freq as resonance of the reed contact arms does interfere with keeping control of the on off times. Got a vid of that i think where the rotor slows in acceleration till it gets past that resonant point then surges further in rpm. Also happens at harmonics of the reeds resonant freq. So you get more than 1 resonance break through, as I remember I got at least 3 on my setup.

Good points MoRo. Never thought of the increase in speed of the contacts gaining distance from each other, only thought of the fact that the distance would be more.  Now that interests me more that the speed of the distance is increased with multiple switches in series.  Thanks for bringing that up.  Was probably about 6 to 7 years ago i played with the series reeds but didnt consider the speed till you brought it up.

Should probably make a thread on the subject.

Mags

#### Magluvin

• Hero Member
• Posts: 5884
##### Re: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)
« Reply #134 on: November 21, 2015, 06:45:46 AM »
But it is also true that by decreasing the number of turns while increasing the size of the wire,can also result in more copper used. The advantage of this is less resistive losses-ohms law rules.

The ultimate inductor would use square copper wire to reduce the air gaps between windings

I am half way through my Woopy replication-should be up later tonight.

I suppose the idea of having 2 coils of the same dimensions and volume of copper but each having 2 diff sizes of wire, that they could produce the same field strength as long as the (P)ower dissipated into each is the same, where the smaller wire coil gets higher voltage input to overcome higher resistance and the larger wire coil gets less voltage to calculate the power as the same.  Makes sense.

The 24 coils on my lasersaber motor are in series. each are 42 awg 3200 turns at 15k ohm all in series. Seems like a large inductance would be had having a high rise time, but due to the high resistance the rise time seems almost instantaneous. Strange to experience.

They do make square enameled magnet wire.  Would probably do wonders for the capacitance of a bifi coil.

Mags