Muller Dynamo

[hartiberlin]

you need to see the schematic to understand the spikes.  then you could read the wave form diagram.
there is 1500 volts of emf coming out of this motor.

@toranarod

where in your circuit do you measure exactly the green and the yellow waveform ?
Could you please explain it in more detail please ?

Do you also have a video of your motor already ?

Many thanks in advance.

@Clanzer :
Fantastic build ! Well done !

Looking forward to see the first run tests.

@Woopy ( Laurent)

Could you please post a new video with many scopeshots across
the parts, so we can have a look into your device what is going on ?
Why is the voltage of one coilpair much bigger in your
device than the other coilpairs ?

I would first optimize for one coilpair output and find there the optimal
output so you know the distances of the rotor versus stator and versus
stator magnets and washers, etc.. and only then switch in the other
coils.
As Romero said, it took him almost one month to finetune
this thing, so you have to have a lot of patience.

Also a Gaussmeter would probably help very much to see,
where you are in your BH curve working point inside the airgaps.

So it is great that Clanzer made all the coils reverse switchable to
see, how they will behave with a different polarity.

Good luck.

Regards, Stefan.

[hartiberlin]

On my youtube chanel under "romero test 3 " a participant (USER 127---)made lot of informations regarding the splitting of the charge on all the coils.

Here are the things this user posted.
Please read it from bottom to top, as his last postings are at the top.

Laurent, thank you for having taken my advice about using a variable load resistor and removing the capacitor so that you could see the waveforms from the generator coils + FWBRs properly. The paradox about this pulse motor + generator is that it is all conventional and your testing, and everyone else's results will show the same thing. That still doesn't mean that you all can't have fun. However, there is a darker issue related to Romero's motivations but not to discuss here.
User2718218 vor 44 Minuten

2nd: You said that you got about 35% efficiency and that sounds about right. Just for fun let's look at how we can improve efficiency.

Let's start by looking at K1, the efficiency of the transferring of the electrical power you pump into the drive coils becoming the mechanical power that is pumped into the rotor.

The way to maximize the value for K1 is to pulse the coils at the perfect "sweet spot" where you get the maximum torque on the rotor when current flows in the drive coils.
User2718218 vor 39 Minuten

3rd: Perhaps you have already experimented with the position of the Hall sensor to find the best sweet spot? My assumption is that the further the hall sensor is away from the rotor magnet the shorter the ON time for the pulse. And of course the angle of the Hall sensor relative to the rotor disk will advance or delay the start of the pulse. For the sweet spot, there is one specific angle between your drive coil and the rotor mag that is the best. So you energize before and after that angle.
User2718218 vor 27 Minuten

4th: I can suggest a simple manual test for getting a feel for where your sweet spot is:  Just energize your drive coil with DC power. Then with your hand hold onto the rotor and manually change the position (angle) of the rotor magnet relative to the drive coil. Simply feel what the torque is like at different angles. You should be able to easily feel where the sweet spot is and by eye know what the start and stop angles are to center your pulse around the sweet spot.
User2718218 vor 25 Minuten

5th: Once you know where the sweet spot is you could temporarily put a LED in parallel with your drive coil and use that as a strobe. Then mark a white line on the edge of the rotor. Then run the motor and move the hall sensor around while observing the strobe illumination of the white line.

Of course there is Plan B:Move the hall sensor around and listen for the maximum RPMs and at same time monitor the current.

CAUTION: When energizing the coil with DC make sure it doesn't overheat.
User2718218 vor 24 Minuten

6th: So Laurent, you could find the sweet spot for each of your two drive coils. Do separate tests for each drive coil. You can do it with the fancy LED strobe illumination and the white line. Like I said just listening to the RPMs and monitoring the current consumption would work also. The wider you make the pulse for each coil (centered around the sweet spot angle) the more power you put into the rotor and the higher the RPMs.
User2718218 vor 21 Minuten

7th: You have to decide how much power you want to put into the rotor. If you make the sweet spot pulse too wide then your efficiency starts togo down. Don't forget, there is only one angle that is the sweet spot angle. The wider your pulse the farther the starting and ending edges of the pulse are away from that ultimate sweet spot angle.

The bottom line is that making careful experiments with the positions of your Hall sensors will maximize the K1 value for your motor. K2 for another day
User2718218 vor 17 Minuten

2nd: I suspect that when you removed 3 coils that the RPMs went up a bit. This means that the 4 remaining output coils outputted a higher voltage into your load, therefore more output power per coil. I am assuming that you are using a resistive load with a true-RMS multimeter. Your power output power measurement is your true-RMS voltage squared divided by the value of your resistor load. So why are you measuring almost the same power?
User2718218 vor 1 Tag

3rd: Part of the answer is understanding how the power actually flows in the motor-generator. Your two drive coils pulse energy and that makes the rotor speed up. The rotor stores the electrical pulse energy in the form of mechanical rotational energy. Here is the key: Suppose in Case A the rotor is turning at 500 RPM and in Case B the rotor is spinning at 1000 RPM. In both cases the drive coils might be on 20% of the time and off 80% of the time.
User2718218 vor 1 Tag

4th: So, if the coils are on 20% of the time for 500 RPM and 20% of the time for 1000 RPM, what can you conclude? The conclusion is that the average power that the drive coils can put into the rotor is almost independent of RPM. That's interesting in itself.

Now look at the pick-up coils. We now know that the power available to draw from the spinning rotor is fixed.  If you have 7, 6, 5, or 4, or 3 pick-up coils, you know that the rotor power that you can transfer into the coils is fixed.
User2718218 vor 1 Tag

5th: So with 7 coils you have 7 instances or Lens drag as you drive the load. With 4 coils at first you only have 4 instances of Lenz drag as you drive the load. That less drag and the rotor speeds up until the 4 instances of Lenz drag draw the same amount of power from the spinning rotor as the 7 instances of Lenz drag.

With 4 pick-up coils the motor establishes a new balance point where the drive coil power source is equal to pick-up coil power drain. It always finds a balance.
User2718218 vor 1 Tag

6th: Finally, what else can you say when the rotor speeds up to the new balance point? You know that the air friction and the bearing friction increases. Therefore the increased friction "steals" some of the available power that would normally go into your output coils and that's part of the explanation for why you measure slightly less output power when you have 4 pick-up coils.

The key to understanding is there is always balance: the drive coil power = the output coil power + friction
User2718218 vor 1 Tag

7th: I also must clarify one important thing here:

Rotor power = (drive coil power x K1) where K1 is the efficiency factor for the coupling of the coils to the spinning rotor. In the REAL WORLD K1 is always less than 1.

Output coil power = (Rotor power x K2) where K2 is the efficiency factor for the coupling of the spinning rotor to the output coils. In the REAL WORLD K2 is always less than 1.

So, its more like: output coil power = (((drive coil power x K1) x K2) - friction power)
User2718218 vor 1 Tag

8th: Laurent: Let me bring it all home for you now with just a hypothetical example:

A really good Romerouk replication might consume 12 watts of input power and with the best possible tweaking and adjusting you will be lucky if you can get 8 watts of output power from the drive coils into the best possible value for the load resistor. You cannot escape the K1 and K2 coupling factors and you can "tweak" forever and you will never make K1 and K2 greater than one. The REAL WORLD.
User2718218 vor 1 Tag

"i get some output on the 7 gen coil all together, and then if i disconnect for instance 3 of those coils (that is to say about 43 % of the generative power ) , i see only a very small decrease of the output power and almost nothing in the input power"

Look at what I said below: "as you add generator coils the available power starts go get shared between the coils." The same thing applies if you remove generator coils.

What happened to your RPM when you went from 7 coils to 4 coils?
User2718218 vor 1 Tag

5th: For you Litz wire do a basic test. Compare one of your current coils made with regular wire and rewind just one coil with Litz wire. If the number of turns are different then that will affect the voltage (or EMF) generated by the coils. Otherwise you will observe no difference in the outputs. You would be making a big mistake if you just blindly rewire all of the coils with Litz wire without making the test I just described.
User2718218 vor 5 Tagen

6th: The simple fact is that the only reason you would want to use Litz wire is if you had very high frequencies in your generator circuit. But the truth is that there are only very low frequencies in your generator circuit. Therefore there is no reason to use Litz wire. You and all of the members on OU have to use your critical thinking skills and stop just blindly believing something because somebody told you. Do the test YOURSELF and prove it to YOURSELF, don't take my word for it.
User2718218 vor 5 Tagen

@User2718218 Yes there is no reason for the use of Litz wire. There is a downside also....to make a Litz bundle to be the same resistance as the solid wire, the overall dia. will be larger and thus less turns possible on the form. On our Muller model, one drive coil gives us 700+ rpm with a current draw of 1.35 amps.

Richard
hhoforvolts vor 2 Tagen

Ok Laurent, I saw your scope shots. You are making slow progress. Permit me to help you.

For starters I have to talk tough about Bolt. He said, "Impedance mismatch!" and "POWER FACTOR CORRECTION" and "incremental standing waves" and "increase overall power by TEN times." This is all ridiculous nonsense and you should ignore it. Bolt doesn't know what he is talking about and he does more harm than goo. Ignore him.
User2718218 vor 5 Tagen

2nd: You put a small DC electric motor as a load on your generator output. This was a big mistake. A DC motor is a non-linear and very complicated load to be testing with. You have to start with the basics and learn from very simple tests. One more time, you should try to understand how your setup works with a basic resistive load, NO FILTERING CAP. You want to see the undisturbed waveforms from the generator FWBR outputs. No filtering capacitor, no DC motor, KEEP IT SIMPLE.
User2718218 vor 5 Tagen

3rd: Try different resistors and look at the waveform. Do not makes any measurements with no load at all, you need the resistor load to "pull down" the voltage when the voltage output from the FWBRs drops. You absolutely must have a simple resistive load.

If you did a god build you should see a regular waveform with spikes that are almost all the same amplitude less the "missing teeth." You want to confirm that your generator output is working properly before you do anything else.
User2718218 vor 5 Tagen

4th: "I am investigating why i get 0.52 watt on one coil and why i do not get 7 time those 0.52watt when all the coils sets are connected??"

You are making an incorrect assumption. In very simple terms, as you add generator coils the available power starts go get shared between the coils. Think about it. Each coil creates Lenz dragon the rotor. You can't keep in adding coils and adding up the output power. It doesn't make sense, the Lenz drag increases as you add coils.
User2718218 vor 5 Tagen

great idea very cool. i need to find some of that magnetic screen.
Magneticitist vor 5 Tagen

Stefan also asked you to try different load resistors across the FWBR but with the capacitor in place. You can also do this but it is not a "pure" measurement like you can get without the capacitor.

Here is the reason why: Let's imagine that your spikes have a peak value of 10 volts and with the filter capacitor and a load resistor you measure 8 volts.

That means that the output from the coil is only active for when the voltage is above 8 volts + the diode voltage drop.
User2718218 vor 5 Tagen

2nd: Can you see this in your mind? If the capacitor is at 8 volts then the voltage spike from the coil only starts to conduct into the load when the voltage is 8.6 volts (including diode drop). Therefore about 80% of the time the coils is "locked out" and it is not driving the load, the charged capacitor is driving the load. Do the test with your scope and you will see what I mean.

If you remove the filter capacitor then the coil starts driving the load when the voltage > 0.6 volts.
User2718218 vor 5 Tagen

The other thing Laurent is that all of the replications are going to fail to self-run. I know this ahead of time, I used to be an electrical engineer and I know what I am talking about.

After all of the self-running tests fail I have a deal with Fausto/Plengo and he is going to post my recommendations for what tests to make on OU after everybody discovers that they can't make a self-runner. The tests will help everybody understand how their replications work and why they don't work.
User2718218 vor 5 Tagen

Congratulations Woopy, you are one of the first ones to have completed your replication. Permit me to give you some advice. In your pictures it looks like you have the filtering capacitor connected to the outputs of your FWBRs. When you make measurements with no load, all that you are doing is looking at the peak voltage because you have effectively made a "peak detector." In other words your FWBRs charge up the capacitor and you only see a DC voltage, you miss seeing the waveform.
User2718218 vor 6 Tagen

2nd: The first thing you want to do is make a basic measurement with a small load like a 1K resistor WITHOUT the filter capacitor. You will then see the actual voltage waveform being produced by the output of the FWBRs. You can check if all of the pulses are about the same amplitude and the same shape, and also see the "missing teeth" where there are missing pulses because of the two drive coils. Try switching the 1K resistor for a 500 ohm resistor or perhaps a 100 ohm resistor.
User2718218 vor 6 Tagen

3rd: So, when you chance resistors (again NO FILTER CAPACITOR) do you see the amplitude of the spikes change? Does the RPM change? What if you use a 20-ohm resistor, what happens then? You do this to understand how your generator operates with different resistive loads.

You said, "This means that the Stator magnets are really important" However, I assume you did this with no load and filter capacitor. Therefore you have almost no information, just a peak detector again.
User2718218 vor 6 Tagen

4th: Please repeat the test where you try with the extra stator magnets in place and without the stator magnets in place, NO FILTER CAP, and with a 1K load resistor. How do the waveforms look with the extra stator magnets in place and without the stator magnets in place? Can you see where the high-voltage pulse is that was responsible for charging up the capacitor to 27 volts? Are the waveforms almost the same or are they very different? A good test you might want to post the results.
User2718218 vor 6 Tagen

5th: Laurent, these are logical tests to do. You say that the the stator magnets are really important. I challenge you and I don't believe that the stator magnets are important at all. The stator magnets can have some influence on the cogging. However, for a constant rotor RPM, with or without stator magnets in place, the output waveform from the FWBRs should be approximately the same. I can't explain how you got a 27-volt peak but I am quite sure that the output waveforms should be the same
User2718218 vor 6 Tagen

The only thing that matters here Woopy is to know which coil produces the most EMF as the rotor turns. The coil that produces the most EMF is the coil that is seeing the highest rate of change of magnetic flux with respect to time. Which coil that is depends on the speed and the angle of the rotor. When one coil is producing the most EMF all of the other coils are 'locked out' and no current flows through them. At any given time only one coil is driving the load.
User2718218 vor 1 Woche

2nd: And that's the whole story. There is no "backwards rotating" patten to study and analyze. You saw with your paper demo how the pattern changes. On the paper demo it's the coil that sees the highest rate of change of the area of the black half-moon shape that is equivalent to the coil that is seeing the highest rate of change of magnetic flux with respect to time.

To repeat, the coil that generates the highest EMF "wins" and drives the load and all of the other coils are "locked out."
User2718218 vor 1 Woche

3rd: There are nine coils where two are drive coils and seven are pick-up coil. Each coil produces two spikes for a magnet pass after the output passes through the FWBR. So you have 14 spikes for seven coils and four "missing teeth" non-spikes because of the two drive coils.

All of you forget about the "pattern" and connect up your scope to the output of all of the FWBRs connected together. Use a 1K resistor as a load and look at scope waveform.
User2718218 vor 1 Woche

4th: Again, to repeat, the 8/9 magnet/coil arrangement is just a way of making the timing of the coils fire one after another where each coil has a time slice (and there are two coils that represent the "missing teeth" in the waveform)

THAT'S WHERE THE ACTION IS - ON YOUR SCOPE. There is no "magical pattern" for over unity associated with the 8/9 configuration, that's all bullshit. Use a 1K resistor as a load and look for the "missing teeth" on your scope.

[conradelektro]

Hi Bill

I am using the good old faithfull RC miniature flange bearings.
These are tough little buggers that are good for 30'000 RPM and take the weight of a rotor brilliantly.
Really low friction and long wind down times.

I go down to as small as 1mm internal size for smaller rotors. But with this one I went bigger to 3mm.

All that is needed is a 6mm hole to mount then bearings in and a shaft lathed down to 3mm on the ends to fit.

I use a Gear shaft to mount the rotor on and again i has a perfect tight 6mm hole for a 6mm shaft.

Here are some pictures, that show how simple it is.

Cheers

Sean.

@Sean: I am following your excellent build with great interest because my mechanical skills are limited.

When looking at the photos in your latest post I wounder why the bottom bearing does not slip of the bottom plate when the rotor is pressing down on it (via the shaft which is lathed down to 3 mm)?

I understand that the bearing fits tightly into the 6 mm hole, but when vibrations start it eventually could fall off the plate nevertheless?

I always think that I have to fix the bearings somehow to the top and bottom plate (more tightly than just pressing them into a hole)?

I like the idea of making the shaft slimmer for the bearings (and having it thicker where the rotor is fixed). But still, what keeps the shaft from sliding a bit up and down? The shaft could "sit" on the lower bearing because it becomes thicker, but at the top bearing it must have a longer slim part, otherwise one can not adjust the distance between the plates?

Greetings, Conrad

[hartiberlin]

ZeroFF has a new update video:

http://www.youtube.com/watch?v=I5vklDDPC0A

[conradelektro]

Concerning the teachings of User2718218:

User2718218 can teach us a lot about careful measurements.

But if one denies flatly that the "backwards rotating" occultation pattern, the biasing magnets on the generator coils and the "pulse driving" have any effect, one misunderstands the purpose of all the replications.

The purpose of the replications is to find out by experiment (and not by words)

- whether the unequal number of coil pairs and magnets,

- whether the biasing magnets and

- whether the "pulse-drive-method"

have an unexpected effect or not.

One could discuss this for ages, but only a real experiment can show what will happen.

Conventional theory allows to "speculate" that nothing special will happen, but only an experiment can "proof" that nothing special will happen.

All new things were found because the inventor/discoverer/tinkerer did not heed the advice of specialists concerning a crucial aspect or detail.

Specialists like to extrapolate their knowledge to areas were one can not be sure that conventional knowledge still applies. Most often they are right, but not always.

Greetings, Conrad