Announcement

Collapse
No announcement yet.

For Peter Lindemann and energenx "Zero Force Motor"

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Thanks Gyula! I'm looking forward to fully exploring that thread - appreciate the reference!

    Hehe - ok, I have a confession - a friend and I asked Paul if we could go see his stuff right after last year's convention, and he said yes! He spent 2 days with us, never asked for a thing, and wouldn't even let us buy him lunch! (in fact, he bought lunch for us!!!) I have nothing but awesome to say about Paul Babcock - a truly giving, amazing individual. When I asked if I could share all that he shared with me, he said "Knock yourself out!", so I am... lol

    So most of the stuff I'm experimenting with is his ideas, and I'd like to give him full credit right here.

    I searched for his patent stuff, and I cannot find it anywhere. Can you link your source? I think I understand what you're mentioning about only one pole facing out - you're saying that it is identical to my drawing, minus one of the permanent magnets, correct? That's an interesting idea! Well worth experimenting with the waveform of this.

    I understood from Paul that he uses the full horseshoe config - 2 neos and an iron core - but I could have mistaken him. The patent application may be misleading too. Unfortunately, it seems like no matter how many pictures you take of something, there is always more that could have been taken to clarify more.

    I need to review it again, but I'm pretty sure Paul says in his presentation that he fires the coils one way, then the other. I'm not sure if you would need to do that with the one-neo horseshoe config you mention. Have you done any testing to confirm that you can fire the same polarity both on approach, and retreat?

    Here's another pic - you can actually see one magnet marked N for North.

    Click image for larger version

Name:	S044.jpg
Views:	1
Size:	91.7 KB
ID:	45294

    As an afterthought here, in the picture of Paul's motor in post #14, you can clearly see magnets on our side of the yellow arm, and the red arm. Logically thinking, if we are observing the retreating side of the yellow arm, then we are observing the advancing side of the red arm. Since we can see magnets on both the advancing and retreating sides, I would say it's almost certain that the one-pole configuration is not one that Paul used on this particular motor...

    Thanks for the input!
    Daniel
    Last edited by emfimp; 02-14-2013, 11:01 PM. Reason: Afterthoughts in green

    Comment


    • #17
      Hi Daniel,

      When I wrote about the circularly arranged row of magnets with one pole towards the coil/core (I quote: "BUT those circularly fixed row of magnets face the coil/core uniformly with only one pole") I did not mean a horseshoe shape with one magnet facing out, sorry that you got it like that. Just imagine a nonmagnetic toroidal ring with say 6cm OD, 4cm ID, 1cm thickness and you make radially oriented holes from the outside (OD) towards the inside (ID) and insert cylinder magnets into these holes with say all N poles facing towards the center of the ring. So let's forget the one-neo horseshoe configuration I did not imply it.

      Okay on the 'origin' of that excellent quality picture on Paul's motor... It must have been a fascinating experience to 'peep in' the birthplace of such unique motor setups. Did you ask some questions on input output power measurements I wonder...

      It is okay that Paul uses a horseshoe config with 2 poles and also okay that he fires the coil one way then the other: with two poles on the horseshoes he has to do so. BUT if you consider the circularly arranged row of magnets with one pole uniformly facing the inside coil/core (as per I described above for the nonmagnetic toroidal ring i.e. magnet holder), then there is no need for firing twice with changing current direction, first you attract in the one pole of all the circular magnets that face towards the coil, and repel out them at the other end by firing the coil once again with the same current direction. I have not done testing on this, I have become rather limited in mechanical construction at home, I had to go on an early retirement and do not have a mechanical shop, only a corner of the kitchen table, lol

      What you wrote in green text, I can agree with it because the horseshoe structure inherently involves two magnet poles, I did not want to suggest at all that Paul used one pole. From the good quality picture you kindly showed I can see at least 6 horseshoe structures embedded circularly into one end of the rotor arm to encircle the coils and this is repeated for the other end of the rotor arm. Paul surely had a reason to choose the two pole structure in horseshoe shape versus the circularly arranged row of cylinder magnets with one pole facing towards the inside coils. The 'Why?' was not discussed with him during your visit?

      Normally I use the European patent office ( Espacenet - Home page ) to look up for patents and found Paul's patent application(s) by searching for his name. This is how I found US2011156522 in the Bibliographic data: Espacenet - Bibliographic data and I did not notice yesterday that inside the actual patent application (when you open it as Original document from the left hand side menu) there is an extra zero character just after the starting 2011 numbers, so the actual patent number is US20110156522. This is why you have not found it, the extra 0 was missing. Now you can use the simpler US patent retrival site, PAT2PDF - Free PDF copies of patents: Download and print! and copy and paste this application number: 20110156522

      Regards, Gyula
      Last edited by Gyula; 02-15-2013, 04:12 AM. Reason: spelling

      Comment


      • #18
        Did he ever get the microcontroller working?

        Tom C


        experimental Kits, chargers and solar trackers

        Comment


        • #19
          Hi Gyula!

          I see what you're talking about now! Thanks for the clarification, and the patent document lends the visuals to go with the words! (BTW, a huge thank you for the patent! I'm still reading through it, and it is awesome!)

          So when you're talking about all N poles facing the center of the ring, you are looking @ page 4 of the patent, where the top figure (Fig 5a) only shows the North pole. This is a front view, and Fig 5b shows the top view of the same rotor arm. I agree that it would be far simpler to create a motor with just North poles - you are correct that in this orientation, you could fire the same way to attract it in to the stator electromagnet as you do to repel it out on the other side.

          I was showing in post #14 that in using this configuration (all North poles), it appears that the generated voltage from rotor movement is far greater than in Paul's configuration - meaning that there is more back emf or Lenz's Law interaction than when you have the double horseshoe config. (both N & S poles facing the core, in a ring around the coil) It appears to be a significant amount, too, so the extra switching required by the horseshoe config is probably well worth it.

          Hope I got it this time!

          It was amazing to see the genius at work! So many awe-inspiring projects, and many completions! He has a lot of new products that are going to be coming to market through his company, Fly-Back. One such neat product is a device that will run AC motors on pulsed DC, and collect the flyback energy in the process!

          I didn't ask output power, but it could be interpreted from the numbers that he did give me. His coils he said were 1500' of 20 gauge, 300mH, 15 ohms, and in the presentation (video), he says he drives them to 3.5 amps. This would suggest that his voltage is between 48 and 60 volts (probably 48). You can assume that 4 coils are always on during rotation, meaning your overall amp draw would be around 14 amps continuously. Power consumption would clock in around 48 x 14, or 672 watts continuous. Normally, this would output somewhere under 672 watts of shaft power, but being his Lenz force is around 10% of normal (I believe this was stated in the presentation or video), I would guesstimate his output at around 1200 watts of shaft power. This would leave 500 to 600 watts of excess power. This, again, is pure guessing, really, and I haven't taken into account the flyback energy. He said flyback capture halved his input, which might mean that he could generate 1200 watts of shaft power using around 350 watts of 'personal' input. If this is true, then there would be around a 850 watt excess from this. Quite significant!

          I had a lot of questions to ask him, but definitely missed some of the pertinent stuff! One of those was also the 'why' of the horseshoe config! Oops... Just another excuse to go back & see him, right?

          One other thing Paul mentioned that may be of interest to you is that he said the coils should have 2 - 3 times the copper that they do in steel. (2 - 3 times in weight) He uses steel shot, like is used in reloading shotgun shells. Impossible to get up here in Canada!

          @ Tom C
          He mentioned something to me about a couple of the coil circuits being down last summer, so he couldn't run the motor for us. Maybe that was it - the microcontroller? (I presume the microcontroller is the 'conductor' of the whole timing operation?)

          I hope that he has it up now - I am going to send him some correspondence in the next couple weeks, so maybe if he has time he'll update me...

          :-)
          Daniel
          Last edited by emfimp; 02-18-2013, 01:27 PM.

          Comment


          • #20
            @ Tom C
            He mentioned something to me about a couple of the coil circuits being down last summer, so he couldn't run the motor for us. Maybe that was it - the microcontroller? (I presume the microcontroller is the 'conductor' of the whole timing operation?)

            Yes that is what I am refering to, He mentioned at the conference the motor was NOT functioning, and he never got the microcontroller working properly, could not ge the timing right or something like that....... so its interesting proof of concept but a long way from a functional device.

            I enjoyed his presentation on modifying physics laws and how they can cause gains in circuits, especially his welding coil experiment.

            Tom C


            experimental Kits, chargers and solar trackers

            Comment


            • #21
              I'll have to watch his presentation again - I don't remember what Paul said exactly on the unit not running... Thanks for the heads up!

              I was just chatting with some friends on the transient voltage spike during coil collapse, and they brought up some very interesting points. They mentioned Aaron's analysis on this at the Energetic Forum (Back EMF vs. Collapsing Magnetic Field Spike), and while I don't agree with everything he says, he has some very valid points. The one that interested me most is that the higher the voltage that the transient spike has to overcome, the quicker the coil collapse.

              I had always wondered why Paul would seemingly use 48 volts as his input to this motor, when in his presentation it seemed like he was advocating that it was better to bias towards low voltage, high amperage. I am wondering if this is why - the higher the supply voltage, then if you're dumping the transient spike into the supply side to re-use it, the transient spike must also be a higher voltage. This causes a steeper gradient - causes the coil to discharge faster, which can possibly open the door to a greater quantity of radiant energy. Less input would possibly be required directly from the operator then. Not to mention that a faster coil discharge allows for a higher frequency of coil operation, which equates to a higher rpm capability.

              Just thoughts...
              Daniel

              Comment


              • #22
                Hi Daniel,

                Okay on your understanding now the all N poles facing the center of the ring as I had tried to express it, you got it.

                In connection with your output power estimations that are based on data from Paul (15 Ohm coils resistance and the 3.5A current etc), I wonder whether this current was meant to be the average current taken from the power supply (for he pulses the coils)? This is why it would be good to learn about output power versus the input, only this reflects truly what a performer this motor is.

                You wrote in your last but one post: "...he (Paul) said the coils should have 2 - 3 times the copper that they do in steel (2 - 3 times in weight) He uses steel shot, like is used in reloading shotgun shells."

                Well, this is a very interesting notice indeed, I have not come across any pulse motor application where this ratio is considered like that. How Paul has come to such a ratio I would like to know for sure. Probably by trial and error but there must be some explanation for it too.

                You wrote in your previous post:
                "I was just chatting with some friends on the transient voltage spike during coil collapse,.... The one that interested me most is that the higher the voltage that the transient spike has to overcome, the quicker the coil collapse."

                I think the speed of the coil's flux collapse depends mainly on the switch-off speed but I do not think it depends (like you wrote) on the initial amplitude of the supply voltage, this is just the opposite way... I mean the spike received from the collapse is always higher than the supply voltage (unless you use a very low speed switch) and the higher the difference between the supply voltage and the received spike's amplitude, the quicker the process during which the spike's energy is pumped back to the supply (if it is directed to the supply which is the case here). Higher voltage difference inherently involves higher current launched by the higher voltage amplitude i.e. by the difference so the recovered energy can diminish more rapidly.)
                So if you use a relatively high supply voltage and you do not wish to spend a fortune on HV quality switches to handle the very high spikes (as a result partially due to the high supply voltage), then you have to choose a trade off supply voltage level somewhere and the 24-48V range may come as a good choice. OF course this also depends on the planned output power of the motor.
                One more thing with this: I think it is also better to use higher supply voltage and less current versus low supply voltage and higher current to achieve the needed performance because any copper or switching loss can be less when the actual current is lower (this is the I*I*R power loss) and peak current does count in this loss too.
                (I basically agree with Aaron's post on the back emf vs collapsing field topic (I have an issue with what he quoted from WikiAnswers but this is not important with this topic), I mean this post: Back EMF vs. Collapsing Magnetic Field Spike but the quote in it was written by someone at WikiAnswers.)

                Of course it is ok that the voltage spike from the flux collapse can be higher when your DC supply voltage is higher but this mainly depends on the switch-off speed, on the coil's inductance and on the coil's current value in the moment of switch-off. (You can get several hundred Volts for a spike amplitude when your supply voltage is a mere few Volts.)

                By the way I am also still trying to digest Paul's patent application on his motor and his other applications on his power supplies because the motor coils are directly switched from the supplies if I see it correctly, so the overall consideration seems to be in order.

                Regards, Gyula






                Originally posted by emfimp View Post
                I'll have to watch his presentation again - I don't remember what Paul said exactly on the unit not running... Thanks for the heads up!

                I was just chatting with some friends on the transient voltage spike during coil collapse, and they brought up some very interesting points. They mentioned Aaron's analysis on this at the Energetic Forum (Back EMF vs. Collapsing Magnetic Field Spike), and while I don't agree with everything he says, he has some very valid points. The one that interested me most is that the higher the voltage that the transient spike has to overcome, the quicker the coil collapse.

                I had always wondered why Paul would seemingly use 48 volts as his input to this motor, when in his presentation it seemed like he was advocating that it was better to bias towards low voltage, high amperage. I am wondering if this is why - the higher the supply voltage, then if you're dumping the transient spike into the supply side to re-use it, the transient spike must also be a higher voltage. This causes a steeper gradient - causes the coil to discharge faster, which can possibly open the door to a greater quantity of radiant energy. Less input would possibly be required directly from the operator then. Not to mention that a faster coil discharge allows for a higher frequency of coil operation, which equates to a higher rpm capability.

                Just thoughts...
                Daniel
                Last edited by Gyula; 02-21-2013, 03:21 PM. Reason: correction

                Comment


                • #23
                  Hey Gyula,

                  You're right - the 3.5 amps is an average - RMS average as far as I understand. Paul mentioned in his presentation (shows up on the video too) that he could pulse the coils at 120 Hz with 3.5 amps. It seemed like that was 3.5 amps per coil, & would probably come out to an RMS of sorts. I also presumed that this was not including any collection of flyback when I did my calculations. All the power numbers are definitely more guess than fact...

                  I'm not really sure how he came about the copper to iron ratio either - it seemed important to him though, so I'm thinking it must have improved his flyback capture or torque generation. Maybe both?

                  Thanks for the clarity on the coil flux collapse - I should have mentioned that I am assuming a high switching speed. Assuming that, the recovery side voltage definitely has a large effect on the time that it takes to collapse the coil magnetic field. I have run numbers on this, and I'll throw a few in this post for all.

                  The following were collected with a classic Bedini bicycle wheel SSG, with an MJL 21194, bifilar coil, running 24vdc, and differing resistors were placed across the recovery side of the circuit.

                  0.1 ohm resistor: ~0.1 volts generated across it, approx. 8 millisecond discharge
                  1.3 ohm resistor: ~1.0 volt across it, approx. 6 millisecond discharge
                  50.7 ohm resistor: ~48 volts across it, approx. 1.6 millisecond discharge
                  90 volt neon bulb: ~250 volts seen across it, approx. 0.4 millisecond discharge

                  These numbers are not very precise - the overall idea is just to show two things. One is that using V=IR, you'll find that the current remains fairly constant. Any variance here is probably more my measuring than the circuit actually creating any difference. The second is just that the coil discharge time varies widely with differing recovery voltages. This can also be seen in that with the 1.3 ohm resistor, the circuit only gets 1 coil fire in per magnet passing. With the 50.7 and higher, it fires 3 times per magnet sweep.

                  An interesting sidenote - using pure feel (not fancy instruments), the thermal energy in the recovery side seems to appear where the most resistance is, and strangely cool temperatures seem to appear where the lower resistance is. So with the neon bulb, the transistor seems to run cooler, and the coil feels cold, while the neon runs warm. With the 0.1 ohm resistor, the resistor feels super cool, while the transistor and the coil feel warmer. Interesting!

                  Paul makes a great argument for lower voltage in his presentation - if done well, it should afford higher motor power outputs using lower power inputs, I*I*R power losses notwithstanding.


                  Daniel
                  Last edited by emfimp; 02-24-2013, 01:01 AM.

                  Comment


                  • #24
                    Hi Daniel,

                    Would like to comment your observations with the different resistors placed across the recovery side of your SSG.
                    With the 0.1 Ohm you found the resistor super cool while the transistor and the coil felt warmer: this is due to the fact that the 0.1 Ohm killed the total collector (inductive) impedance what the coil represented so the collector current hence the coil current inherently got increased and caused those dissipations you observed. There was no chance for the recovery coil to be involved in the induction so the resistor did not have useful AC pulse across itself (0.1V arcoss 0.1 Ohm).
                    The other end of the this 'temp scale' is the neon bulb which has but a little loading effect on the coil, this way the coil inductive
                    impedance can be higher, reducing the collector current hence dissipation in the transistor and the coil. The neon got warm because it had no series current limiting resistor and whenever the spike was higher than the 70-90V breakdown voltage of the neon, the current went high in the lamp, causing dissipation.

                    I assume that on the 'cool or super cool temperature'of a component you meant a comparison to the normal ambient room temperature you were in. This comparison when done just with our skin on the face or by our fingers can be misleading because depends very much on the heat conductivity of a particular component. Mainly metal surfaces are prone to let us in because they can easily siphon off heat from our skin, albeit their surface is just on room temperature say 25°C versus our 35-36°C skin temp at our fingertips.
                    I would have to go through Paul's video to get what you referred to on the lower voltage argument, how could it be done well to still get higher motor outputs at lower power inputs: can you tell approx 'minutes' he mentioned this. If not, no problem.

                    Greetings, Gyula

                    Comment


                    • #25
                      Babcock Alternative Magnet Configuration

                      So yesterday my good friend Gestalt helped me take a few more scope shots (thanks G!) - we were trying out the alternative magnet configuration as shown in Paul's patent, so helpfully donated by Gyula... (thanks Gyula!)

                      In post #14, we explored the configuration shown in Fig. 6a in that patent (patent 20110156522) - that's the configuration where the magnets are all pointing towards the center of the coil, or perpendicular to the core direction. We compared the horseshoe design to just a single magnet oriented also perpendicular to the core direction. This time around, we were trying the config shown in Fig. 6b, as well as some of our own 'home brew'. This configuration has the magnets facing parallel to the core of the coil, or axially.

                      We were using a different magnet setup this time - using ferrite magnets instead of neo's. The neo's in post #14 measured 0.700" dia by 1/8" thick, and were on a much smaller steel bar. This time we were using the 1/8 x 7/8 x 1 7/8" ferrites that JB uses on his bicycle wheel. Just what we had on hand... I'll post a pic in my next post of these 2 magnet setups.

                      Here's a drawing of the four configurations we tested this weekend:

                      Click image for larger version

Name:	scan0025.jpg
Views:	1
Size:	73.9 KB
ID:	45311

                      So test #1 we put in here for a comparison - especially since we had changed our magnets to the ferrite, and changed the amount of steel involved. (this is Fig. 6a in the patent) Here's the waveform:

                      Click image for larger version

Name:	90 deg mag.jpg
Views:	1
Size:	93.1 KB
ID:	45312

                      Numbers for test #1:
                      - initial & final voltages - 0.55 volts
                      - central peaks - 3.75 volts
                      - initial & final periods (each) - 33 ms
                      - total wave period - 150 ms

                      - initial volt/central peak - 14.7%
                      - initial period/total wave period - 22.0%


                      In #2, we tried the actual Fig 6b configuration - or at least a config that's close to it...

                      Click image for larger version

Name:	axial mag.jpg
Views:	1
Size:	94.5 KB
ID:	45313

                      Numbers for test #2:
                      - initial & final voltages - 0.21 volts
                      - central peaks - 1.4 volts
                      - initial & final periods (each) - 32 ms
                      - total wave period - 115 ms

                      - initial volt/central peak - 15.0%
                      - initial period/total wave period - 27.8%


                      In #3, we tried a modified 6b configuration. We wanted to see what a 'solid' magnet with a steel backing would do...

                      Click image for larger version

Name:	axial solid mag steel.jpg
Views:	1
Size:	96.8 KB
ID:	45315

                      Numbers for test #3:
                      - initial & final voltages - 0.43 volts
                      - central peaks - 3.2 volts
                      - initial & final periods (each) - 40 ms
                      - total wave period - 125 ms

                      - initial volt/central peak - 13.4%
                      - initial period/total wave period - 32.0%


                      Finally, the #4 test was all about seeing if the steel or iron backing, making this into an actual horseshoe configuration, is actually needed or not.

                      Click image for larger version

Name:	axial no steel.jpg
Views:	1
Size:	96.2 KB
ID:	45314

                      Numbers for test #4:
                      - initial & final voltages - 0.38 volts
                      - central peaks - 3.5 volts
                      - initial & final periods (each) - 40 ms
                      - total wave period - 130 ms

                      - initial volt/central peak - 10.9%
                      - initial period/total wave period - 30.7%


                      This time around, I can see that sometimes small variations can make significant changes in the percentages generated, so my margin of error here is probably fairly significant. These are by no means highly controlled experiments - however I am satisfied that they generally give a nudge in the correct direction. I'm also realizing that the spacing between the magnets may play a large role in the periods generated (when there is a gap between magnets), so I've tried to keep the spacing similar, as well as the pendulum velocity during coil transit. I may not have succeeded in this. The voltage generated is probably the most significant value here anyways, so I'm not too worried about the period.

                      That said, all the numbers compare quite closely in these trials, which for me was a pleasant surprise! They also compare favorably with the initial trials done in post #14.

                      It appears that using the axial configuration of permanent magnet does not require a horseshoe shape at all to produce a very favorable low voltage during the initial & final peaks, nor does it appear to require steel or iron at all. The waveform generated in test #4 really is basically the same as any of the waveforms here. That said, I haven't exactly recreated either of the figures 6a or 6b. Note, however, that Paul Babcock himself did not create those exact shapes in his test motor.

                      All things considered, the rotor is one of the most difficult pieces to manufacture because of the magnet & iron configurations, and the need to have a nonmetallic rotor. (consider how you would physically put it together, looking at the photo in post #14, when magnetic fields are interacting, glue is drying, etc...) These results are exciting to me, as they are pointing me towards the possibility of a very simple rotor assembly that potentially would perform as well as Paul's prototype motor.

                      One could cut the hole a little larger than the coil diameter, and then drill holes for cylindrical neo magnets all around the periphery of that initial large hole, parallel to the axis of the initial hole. In placing the neo's, all the North poles would be placed at the same end, and all of the South's would be on the other end. An end cap could be glued over each end, and clamped as the glue around the neo's and end caps dries. This would take probably 10% of the time that Paul's prototype motor took to built the rotor. My next post will also have a drawing of this design.

                      Cheerio!
                      Daniel
                      Last edited by emfimp; 02-24-2013, 05:07 PM.

                      Comment


                      • #26
                        So I'd promised a picture of my magnet & steel configurations - the left hand setup is my neo configuration (post #14), and the right hand setup is the current setup (for post #25)

                        Click image for larger version

Name:	mag setup 008.jpg
Views:	1
Size:	84.7 KB
ID:	45316

                        Here is the drawing of how I picture what is possibly the easiest construction for the rotor assembly... The optional cover plate would have to be nonmetallic, as is the whole rotor assembly.

                        Click image for larger version

Name:	scan0026.jpg
Views:	1
Size:	74.4 KB
ID:	45317

                        Gyula, in the video, if you start at the beginning he really lays out the groundwork for the whole low voltage argument, but some points of interest are around 5:02, 6:54, 8:29, and 9:40 especially.

                        Cheers,
                        Daniel
                        Last edited by emfimp; 02-24-2013, 04:43 PM.

                        Comment


                        • #27
                          Hi Daniel,

                          Thanks for showing the nice tests, and sorry for a late response, I have been busy.

                          I understand the conclusions from the tests can be drawn as you wrote but choosing the setup in test #4 has the most severe stray flux able to spread everywhere in the vicinity of the rotor area, while using steel (i.e. suitable core) for guiding and focusing most of the magnets flux to where it should mainly go is also a consideration. Also, you seem not to bother with the presence of the center peak waveform (+/-3.5V) which is ok if you do not wish to utilize it but if you do, then Lenz drag is going to happen. One more notice here is that these test setups produce 'conventional' waveforms, not like the one coming from the 'zero force' coil orientation. Of course this is not criticism, just my observations and I know your goal has been to test and find the most 'attractive' setup.

                          All in all, you then finally settled for the 'all_single_pole_in' magnet arrangement for the rotor assembly as you drew it in scan0026.jpg so I assume your to-do list includes a test for that too.

                          I watched the video as per your minute indications and now I know how you meant Paul's points for achieving higher motor power while using lower power inputs. BUT earlier you also wrote: "I had always wondered why Paul would seemingly use 48 volts as his input to this motor, when in his presentation it seemed like he was advocating that it was better to bias towards low voltage, high amperage." and that is why I mainly started to answer like you saw it in reply #22.
                          Also, in general, by just lowering supply voltage level and increasing the current is not normally the preferred method just due to the I*I*R loss issue (what you also mentioned in #23). Of course, by cleverly combining and applying the 'tricks' shown by Paul a good 'compromise' setup in the positive sense can surely be accomplished. In our everyday routine practice we do tend to stick with say 12V operational supply voltage, probably just because the wide availabilty of the 12V car batteries so a 36V or 48V voltage source sounds already an 'overkill' perhaps for the mere cumbersomeness of handling 3 or 4 batteries and we just fail to consider possible advantage of the higher supply voltage levels.

                          Greetings
                          Gyula

                          Comment


                          • #28
                            Hey Gyula,

                            It's true, #4 should have the most stray flux of all of them, but it still seems to have the very same voltage generation characteristics that I'm looking for. In fact, the voltages generated are so similar to the tests including steel that anyone would be hard pressed to pick out which scope shot is which, if they weren't labelled. With this magnet configuration in particular (axially oriented magnets, as compared with the core), the steel seems to really only affect the side of the magnet that is away from the coil anyways, so stray flux is maybe not an issue.

                            It's true - I don't mind what the central peaks are, since I'm not interested in using them. This is Paul's idea, and I think it is a good one.

                            I would contend that these waveforms are not, in fact, 'conventional' waveforms. They are actually two zero force waveforms combined into one. The first half of the waveform is what you'd get from say a single North pole magnet approaching the coil center, and the last half of the waveform is what you'd get from a single South pole magnet leaving the coil center, and exiting the area. Thus the initial and final (small) peaks are opposite one another - one positive, one negative. This is what causes the 'conventional' waveform as well - since there must be a flux change when the magnet or magnet assembly is directly centered over the coil - the main influence on the coil is changing from the N pole end to the S pole, or vice versa.

                            Somehow in producing this 'conventional' waveform, the initial & final voltages are lowered by a huge amount. From my testing, it seems that it is fairly easy to lower these voltages by a factor of about 3. That is a huge reduction in counter emf that must be overcome! I've tried a lot of single magnet configurations a week ago, including all kinds of different size magnets (sometimes combined), different strengths, shielded in different manners with steel, etc., and I've had a hard time finding even one configuration that approaches the benefits that I see in the 'axial' magnet configuration. I did manage to lower the generated voltages with steel shielding on a single magnet, but the ratio of the initial voltage to central peak did not change. This makes me wonder if the shielding I was doing was actually reducing the amount of flux available for motor effect also. (could be wrong on that) In order to get these results I had to shield in ways that made me think that I was reducing the flux available for motor interaction. With the configuration #4, I feel that there is lots of flux available for this motor interaction, while the voltage generated in the motoring area is some of the lowest I've seen, and the construction simplicity is at an all time high. Seems like a winning situation to me!

                            You were mentioning a test for what I drew in scan0026 - I hadn't considered it as I was fairly content with testing the 'single magnet' version of it in test #4, but thanks for the recommend! I am planning on building a replication of Paul's motor anyways, but I'll try to work in a full scale test of that configuration when I have time.

                            Thanks for the comments & perspective!
                            Daniel

                            Comment


                            • #29
                              Hi Daniel,

                              Your stance on the waveforms (...not conventional ones.. actually two zero force waveforms combined into one) is interesting, nevertheless. They certainly look like as you wrote they could be created: The first half of the waveform is what you'd get from say a single North pole magnet approaching the coil center, and the last half of the waveform is what you'd get from a single South pole magnet leaving the coil center, and exiting the area.

                              I editied your earlier "axial no steel.jpg" scopeshot picture and inserted a moving magnet and a coil from your drawing. I did this because it is easier to ask. So I wonder if you were to fire the attract-in pulse in the area of the waveform I show with a red line on the left and consider also the position of approaching magnet to the coil.
                              And likewise I wonder if you were to fire the coil again to repel out the leaving magnet at the area of the waveform I indicated with also a red line on the right side? (Because I assume you wish to fire the coil twice, and this would involve a change in the input current direction of course in this setup.)
                              I hope I did not misunderstand: in your previous post you meant the axle of the coil (which is horizontal in the drawing) is parallel with the heading direction of the magnet movement, right?

                              I do think that there is a certain relationship between the axial length of the coil and that of the magnet and this ratio may influence the full duration of the waveform and its shape in a certain extent. Have you considered or found this during your pendulum tests? Also the question of using a core in the coil or not?

                              Greetings and thanks,
                              Gyula

                              Click image for larger version

Name:	axial no steel_mod.jpg
Views:	1
Size:	87.5 KB
ID:	45339

                              Comment


                              • #30
                                Nice diagram, Gyula! Thanks for drawing that!

                                Paul says he fires the coil about exactly where you show - attract in, repel out. The core of the coil does indeed need to be parallel with the direction of magnet movement - I was noting especially that when the coil is parallel with the axis of the magnet, or said another way a line between the poles of the magnet is parallel to the coil core, then the steel possibly doesn't have much effect on the flux pattern. Not that would interfere with our desired operation, anyways... (maybe a better way to say it is that fig. 6b in the patent doesn't seem to see a difference between steel & no steel)

                                The waveform doesn't seem to be hugely influenced by the magnet length, other than the waveform being more compressed with a smaller magnet. All the ratios seem more or less the same when changing magnet lengths. (initial to central peak voltages, etc seem to be the same.) Paul seemed to use a magnet assembly that is about 80% of the length of the copper coil (guesstimate) - I've been wondering how critical this is too. The only thing that comes to mind is that around the length of magnet assembly that Paul uses, probably the coils on either side of the rotor magnet assy will come on at the exact same time. This would create a bucking field between the electromagnets (coils) - not sure how strong this bucking would be, or if it would have any benefits or detractors? He has a great picture of what I'm thinking of in Fig 7 in the patent. I'm thinking all 4 of those coils might actually be on at the same time in his prototype. If one uses a shorter magnet assembly, then maybe only 2 coils would be on at the same time - 180 degrees from each other, and both pointing a mag field in the same direction around the toroid. Does it matter? I don't know...

                                I haven't really tried many experiments without a core - I think it is important to have the core in order to create a larger field and generate more rotor torque. It could be tried without a core, though. John B has a great video on the energetic forum of a zero force motor, in which he removes the iron core, and inserts it again. The amp draw goes up, and the rotor slows down when he removes the iron. In fact, the ammeter almost goes to zero when he inserts the iron. I haven't been able to find this post again - if anyone knows where it is, can you post it please?


                                Daniel

                                Comment

                                Working...
                                X