Quote from SqueezingSparks on February 9th, 2018, 06:24 PM

Russ,

Here is a decent little LC resonance calculator:  http://www.ham-radio.com/lc.html

I punched in 4400 Henries and 10 mrofarads and got 0.75 Hertz.

So the joke is on me because I suggested 1000 uF and 100 uF.   Clearly the resonant frequency would be so low that presumably the resistance of the coil would dampen the resonance a lot and prevent even one full cycle from being captured on the DSO.

It's just another reminder about how big 4400 Henries of inductance really is.

SS

Haha,  well,  at least you see where I was trying to go,  seems it worked.  And it also backed up the Inductance calulations that work for the most  part.  (Thanks for that!!!)

Here was the " we finally got it"

https://youtu.be/8ExRNNHbgSg

4200-4400 H and about 480pF

SRF ~110-120 hz dependent on what's around it. 

That's an air coil!  Wow!

Might be one of the lowest SRF most of us have ever seen!  Lol

~Russ

"Finally, I am wondering if by chance you got your units off by powers of 10 when you did the time constant calculation.  The time is in seconds and the resistance is in ohms."

I verry much could have,  becuse I came up with ~45,000H and we are at 4400H.  I'll meed to check that! 

~Russ

Okay Richard & Russ, while you have this process relatively fresh in your minds, how about creating a spreadsheet for doing these calculations.  Keep things top down in the order that you need to find the variables and add some side column notes of what needs to be done and how you connect things up.  Seems to me like it might be useful in not just this project but for those playing with Stan Meyer setups too.  Get something rough and I'll work on sweetening things up, maybe even add Excel's solver and some macros to make this whole process see-spot-run.

"Seems to me like it might be useful in not just this project but for those playing with Stan Meyer setups too."

---YES---  :clap:

Step 1. try try again
Setp 2. Beat head agenst table (at lest 3 times)
Step 3.....

Ok good idea.

Will work on that. 

~Russ

Quote

Matt:

Get something rough and I'll work on sweetening things up, maybe even add Excel's solver and some macros to make this whole process see-spot-run.

Actually I've written an entire database program from the core. It can use all common .dbf formats. I have to start compiling some of my own data.
At the phase of development this database can quickly create and manipulate multiple open tables. It does not however have yet the structured query language or proprietary query and extended data formats yet. I had to quit working on it for reasons 4 years ago it need minor fixes from windows XP to windows 10.

In the short term a spreadsheet works fine and is ok for small immediate need. In the long term a database will do much more and save time.

       

https://www.youtube.com/watch?v=dAntpkLwsw8

https://www.youtube.com/watch?v=Xn9zEs9gKNE

~Russ

Here is an idea for you Russ:

Wrap say 30 turns of wire around the outside of the big coil.  It doesn't have to be tight, in fact you want it to be a bit loose.  So perhaps temporarily tape a pencil to the coil for the wrapping, and then when you remove the pencil you have a loose fit between the big coil and the 30-turn coil.  Then use some tape to firm up the form of the 30-turn coil.  What you want is for the 30-turn coil to be a movable ring that you can move up and down the big coil.

The 30-turn coil is your sensor coil that you can connect to a scope channel.  It will show you the rate of change of magnetic flux with respect to time through the sensor coil.  So that means it will show you a combination of the influence of the rotating magnet and what is happening when you energize the coil.  By moving the sensor coil up and down the shaft of the big coil you will see how the rotating magnet affects different parts of the big coil.

It may be able to show you how smooth or rough the commutator contacts are but that is to be determined.  Don't forget the main coil has so much magnetic inertia that it may smooth out all of the rough patches when the commutator arcs.  In other words, the current flow rate in amps though the coil will barely change when there is a commutator plasma spark.

Since the coil is showing you the rate of change of magnetic flux with respect to time, it would be nice if you could see the actual amount of flux in the coil also.  In technical terms the sensor coil will be showing you the first derivative of magnetic flux with respect to time through the coil.  Therefore if you integrate on that signal from the coil, the output of the integration will be the amount of flux "flowing" through the coil.  So if your DSO has an integration function you can play with that.  It would be cool to see the amount of flux in the coil versus time.

Integration functions are tricky though in the sense that they ideally need a zero reference when there is no flux flowing through the sensor coil before you start integrating.  Also, they tend to drift upwards or downwards over time because your DSO is probably sampling with 8-bit quantization and the associated quantization errors will accumulate as you integrate over time.

However, if the stuff above sounds like gobbly-gook to you I have good news.  What it effectively means is that if you just do a one-shot capture of a waveform including the "integrating of the signal from the sensor coil to show you the actual flux value in the coil versus time" then the one-shot capture waveform should look just fine.  But if you keep the capture running continuously then the subsequent waveforms for the integration are going to start looking wonky.

SS

Russ,

Let me explain the above about the DSO integration function with a concrete example.

You know that when the commutator energizes the coil with short pulses the current in the coil ramps up slowly and linearly, and then drops quickly to zero if there is an "open" on the commutator and you get a plasma burn.   Well, if the integration works well, then the integrated signal will show you just that, ramp waveforms when the commutator makes contact where the ramp represents the actual amount of increasing flux in the coil.

Note also that the regular signal from the sensor coil should also show you the flat-topped pulses that represent the energizing voltage signal from the commutator applied to the coil.  So that is also really good information.

In the case of the rotating magnet, the sensor coil shows you the rate of change of flux from the magnet and the integrated version of the sensor coil signal will show you the actual amount of flux passing through the sensor coil.   In simplified terms, if the sensor coil outputs a sine wave from the influence of the rotating magnet, then the integrated signal will be a cosine wave (the sensor coil signal effectively shifted by 90 degrees.)

SS

After Russ is done scoping that auxiliary coil, he might think about hooking it up in parallel with his commutator and moving the auxiliary coil up and down the main coil until he gets the least amount of commutator sparking... the voltage induced in the auxiliary coil will spike voltage at the commutator at just the right time (if it's moved to the correct point on the main coil) such that the commutator arc is 'blown out' quickly or possibly even prevented from occurring. He'll need a diode between the auxiliary coil and the commutator to prevent the external voltage pulse applied to the main coil going through the auxiliary coil.

With precise enough positioning, you'd in effect get a very sharp commutator off-time slope, spiking main coil voltage even higher. We want fast on and off times.

Another way of getting that is to insert small neodymium magnets around the perimeter of the commutator... a spark gap with a strong magnet will 'blow out' easily... the commutator when the brush leaves one of the commutator segments is acting as a spark gap. The quicker that arc 'blows out', the more the voltage will spike.

Russ,

I hope this sensor coil stuff is something you will consider.  Whether you are looking at the "regular" signal from the sensor coil or the "integrated" signal from the sensor coil, this is like taking the motor's vital signs.  You will see the combination of the effects of the spinning magnet rotor and the pulsing coming from the commutator.  And the fact that you can slide it up and down the big coil means that can "look at the action" in different parts of the coil.  That is not really relevant for the commutator pulses, but it is relevant for seeing how the rotating rotor magnet influences the big coil.

For your DSO if you have say two channels that are 10-bit and two channels that are 8-bit then you have to put the sensor coil on a DSO channel with 10-bit sampling to give you a cleaner integrated signal.

The other issue about the sensor coil is how many turns will it need to give the DSO a decent signal.  My gut feel is telling me 30 turns should be fine, but that is to be determined.

Final comment to bring this all back to hand-sketched timing diagrams.  You can do the integration manually by doing a rough approximation in your head, you don't actually need the DSO math function to do it.  I am just mentioning this to reiterate that a hand-drawn (or done with some drawing software) timing diagram has great value.  This is where you sit down and digest the information and then sum it all up in a timing diagram that explains what is going on.   One properly analyzed and explained and annotated timing diagram is worth more than 100 DSO captures.

SS

A special bonus comment about the sensor coil.   There is a concern that the changing flux from the rotating magnet is going to "drown out" the pulsing signal from the commutator.  Think about it, if the rotating magnet is generating a steeply rising waveform, then when the a commutator pulse fires during that steeply rising waveform it might be almost impossible to see.

Fortunately we have a "secret weapon."  The secret weapon is that we move the sensor coil to the center of the big coil and line it up so that it is exactly at the height of the rotating axis of the rotor magnet.  If we are lucky we should be able to nearly eliminate picking up any changing flux from the spinning rotor magnet such that the only thing the sensor coil picks up is the action from the commutator.

P.S. to this comment:  I am wrong and did not properly visualize the full rotation of the rotor magnet relative to the sensor coil positioned in the middle of the big coil.  In theory when the rotor magnet is horizontal there will be very little net flux cutting the sensor coil and hence the rate of change of flux will be very low.  But clearly when the rotor magnet is north-up there will be flux going down through the sensor coil and when the rotor magnet is north-down there will be flux going up through the sensor coil.  Therefore by definition there will be changing flux inducing an EMF into the sensor coil and no "secret weapon."  So at best you can say when the rotor magnet is near horizontal, then with minimum changing flux from the rotor magnet, you should be able to see the commutator pulses.  Note when the rotor is horizontal the assumption is that this region is the sweet spot for torque imparted on the rotor from the commutator pulses.

What is for sure is the waveforms induced by the rotating magnet will be different if the sensor coil is at the top, middle, or bottom of the big coil.

Whoops, I have to squeeze in one more comment.

We want the sensor coil to only be looking at what is taking place in the big coil.  But we have that big honking rotor magnet that is spinning and subtending this huge changing magnetic field everywhere.  And that big changing magnetic field will also induce voltage in the two wires coming back from the sensor coil.

So how do we mitigate this unwanted influence from the rotating magnet?

The answer to that is easy.   The two wires from the sensor coil are simply twisted, i.e.; a twisted pair.  It doesn't have to be a tight twist or super fancy, just make a twisted pair.

So let's say there is a 20-inch twisted pair wire that comes off of the sensor coil to bring the signal to the bench.  And you hook up your scope probe there.

EEVblog #662- How & Why to use Integration on an Oscilloscope

https://www.youtube.com/watch?v=Dh0xYu8YvaE

Just some side thoughts here.

Disclaimer...I am not an electronic engineer, but I am a pretty good googler.  ...and, I could have a completely wrong concept of how I think the engine, or process functions, so read at your own risk…lol

If you think of the coil as a wind chime, it will tell you where you have to open the coil based off of certain length measurements and type of material being used.   This wind chime concept probably applies to the coil too.

Also, both sides of the coil need to be opened at once so it will oscillate the maximum amount of time.  Similar to hanging a wind chime from a string so it can vibrate longer.

By using an aurdino or similar, to provide the ping timing you should be able to adjust the best time to apply the ping to a point when you reach the strongest and or longest magnetic attraction.  You really won't even need an Oscilloscope to tune it.

The aurdino would take the place of the commutator.

The question is can the little PC  and associated components switch fast enough to provide the resolution needed to open the coils at the precise time.  It would need to happen pretty quick, but if you hit it right it may ring for an extended period.  Form a sustained magnetic field?? Or what?

Hopefully the measurements Russ took will provide some insight as to how fast things have to perform the switching to catch the voltage spike at the exact right location within the coil.  That is what oscillates back and forth in the coil…probably.

The wind chime concept came to me as I was trying to design a circuit I've been calling the Pulse Cycle Generator Circuit, my electronic design skills are 100% google searches and reading...really slow process and there is a pretty high chance I may have things wrong.  I am about to begin Rev 1.3 now to try and figure out what is need when thinking in terms of Arduino type PC's or Raspberry Pi's  to provide the switch timing.  Seems like I read they can switch at 16 mhz and Russ's coil was ringing at around 117 Hz, so that gives a little time to play with.

Later,
Brad

Brad,

A wind chime is a bell.  And a bell is an LC resonator.  So the wind chime can be looked at like a coil in self-resonance, or more conventionally, like a coil connected in parallel with a capacitor, and the two components form the resonator.

Certainly some kind of microcontroller setup could ping the LC resonator and make it ring.  And like you said, for the big coil it self-resonates very slowly.  If you connect a capacitor to the big coil it resonates even more slowly.

But here is the key question:  What is that going to get you?  What is the next step?  You have to keep in mind that a resonating coil produces a magnetic field that changes direction periodically.  Normally, that would mean the rotor magnet would just wiggle and not turn.

Perhaps you were thinking that at the right resonance frequency the magnet would then turn continuously.  That's absolutely true, it seems reasonable.  But one more time, what is that going to get you?  What is the next step?

SS

2 thoughts, 

1. Using semiconductors...  It will fail if at all it sparks back...

2. If you ready my. Doc (you too SS :)  hehe)
You will see that the entire idea is to run it as its SRF

This will allow you to use resonance to:

A. Stop the flow of current through the coil.
B.  Let the coil gain in amplitude while the magnetic field grows in each direction.  Now ask how can we extract that stored energy?  If we do it electrically we will dampen the stored energy,  so we can't do that..  So we can however use that magnetic field to turn a rotor!

And that rotor can be set up to not disturb the coils resonance... 

Good thoughts brad,  good thoughts for not knowing to much electronics. 

A simple thought attached...  yes its not quite right but you get the idea...

~Russ. 

PS READ MY DOC SS ;) (this will also help you brad) HERE

Russ,

I have glanced through your doc, but still haven't read it in full.

This:

Quote

The 4500 lb coil only had 13 ohms of resistance!!!

That's a claim for a giant Newman build, right?  But doesn't it not smell right?  It has to be many miles of wire and I bet you if you looked up any AWG wire chart showing weight per unit length, resistance per unit length, etc, you will not find a "fit" for that claim.  Something to think about.

One easy test for you will be to see if you can "put current into a coil" like you are filling it up with current almost like a capacitor.  This is all over the beginning of your doc.  Simple test: something like a 5K current sensing resistor at one end of the coil and a 5K current sensing resistor at the other end of the coil.  I am assuming that you have differential probes or you can use your clamp-on current probes.  You want to see current go in one end of the big coil and not come out the other end of the big coil.

Yes, you might see a tiny smidgen of current go in one end that does not come out the other end and the explanation for that is you charge some parasitic capacitor in the coil.  But that capacitor will be minuscule.  When you say the big coil will "fill with current" you are suggesting a violation of Kirchoff's Current Law, and that is a very tall order.

I suggest that you devote a day to investigating this exclusively because your premise for the operation of the motor basically depends on this from what I can see from skimming through your doc.

SS

Russ,

Quote

You will see that the entire idea is to run it as its SRF

This will allow you to use resonance to:

A. Stop the flow of current through the coil.
B.  Let the coil gain in amplitude while the magnetic field grows in each direction.  Now ask how can we extract that stored energy?  If we do it electrically we will dampen the stored energy,  so we can't do that..  So we can however use that magnetic field to turn a rotor!

And that rotor can be set up to not disturb the coils resonance...

It will be great to see you get it running at the SRF.  It's too hard for me to visualize what will happen with confidence.  And I am not sure if you are talking about pulsing at the SRF or if you want the rotor to spin at the SRF.

A note about the magnetic field growing in each direction.  The volume inside a capacitor has an electric field and the volume around a coil has a magnetic field.  In both cases you can visualize a certain volume of space as representing a certain amount of stored energy.  To create that volume of space with a field inside it typically requires electrical work be done, i.e.; electrical energy.  You have to pay an electrical energy price to get that magnetic field created in that volume.  Keep that in mind when you say "the magnetic field grows in each direction."

The other thing I noticed is saying "Stop the flow of current through the coil" and "magnetic field grows in each direction" are incompatible statements.  Here is a Golden Rule:  Current and magnetic field are one in the same.  There are intertwined and you can't have one without the other.  That's why in most electronics calculations the magnetic field is ignored because they talk about current, and current is the magnetic field.

Quote

If we do it electrically we will dampen the stored energy,  so we can't do that..  So we can however use that magnetic field to turn a rotor!

Well, this is another thing to test, right?  You are suggesting that the magnetic field can turn the rotor without affecting the amount of energy stored in the magnetic field.  That's a very very tall order that goes against the conservation of energy so you are suggesting that there is some secret sauce here.

This is where you will find "invisible voltage drops" and stuff like that if you investigate.

SS

Quote from SqueezingSparks on February 11th, 2018, 09:26 PM

Russ,

I have glanced through your doc, but still haven't read it in full.

ok, take the time to read it and most importantly read ALL the references!

Quote from SqueezingSparks on February 11th, 2018, 09:26 PM

This: 
That's a claim for a giant Newman build, right?  But doesn't it not smell right?  It has to be many miles of wire and I bet you if you looked up any AWG wire chart showing weight per unit length, resistance per unit length, etc, you will not find a "fit" for that claim.  Something to think about.

actually, it matches almost exactly. i did  the calculations. 

( see attachment for higher rez)


i would not spend so much time on this stuff with out doing my homework first :)

the entire idea is to have miles of wire as to stop the current from going through! its part of the dezighn!

Quote from SqueezingSparks on February 11th, 2018, 09:26 PM

One easy test for you will be to see if you can "put current into a coil" like you are filling it up with current almost like a capacitor.  This is all over the beginning of your doc.  Simple test: something like a 5K current sensing resistor at one end of the coil and a 5K current sensing resistor at the other end of the coil.  I am assuming that you have differential probes or you can use your clamp-on current probes.  You want to see current go in one end of the big coil and not come out the other end of the big coil.

Yes, you might see a tiny smidgen of current go in one end that does not come out the other end and the explanation for that is you charge some parasitic capacitor in the coil.  But that capacitor will be minuscule.  When you say the big coil will "fill with current" you are suggesting a violation of Kirchoff's Current Law, and that is a very tall order.

I suggest that you devote a day to investigating this exclusively because your premise for the operation of the motor basically depends on this from what I can see from skimming through your doc.

SS

100% yes it dose violate of Kirchhoff's Current Law

This has already been discussed a lot in this thread.

to save you time, go here and read the "Limitations"

 https://en.wikipedia.org/wiki/Kirchhoff%27s_circuit_laws

we are hitting just about everyone one of those. lol

~Russ

Quote from SqueezingSparks on February 11th, 2018, 09:49 PM

Russ,

A note about the magnetic field growing in each direction.  The volume inside a capacitor has an electric field and the volume around a coil has a magnetic field.  In both cases you can visualize a certain volume of space as representing a certain amount of stored energy.  To create that volume of space with a field inside it typically requires electrical work be done, i.e.; electrical energy.  You have to pay an electrical energy price to get that magnetic field created in that volume.  Keep that in mind when you say "the magnetic field grows in each direction."

The other thing I noticed is saying "Stop the flow of current through the coil" and "magnetic field grows in each direction" are incompatible statements.  Here is a Golden Rule:  Current and magnetic field are one in the same.  There are intertwined and you can't have one without the other.  That's why in most electronics calculations the magnetic field is ignored because they talk about current, and current is the magnetic field.

Well, this is another thing to test, right?  You are suggesting that the magnetic field can turn the rotor without affecting the amount of energy stored in the magnetic field.  That's a very very tall order that goes against the conservation of energy so you are suggesting that there is some secret sauce here.

This is where you will find "invisible voltage drops" and stuff like that if you investigate.

SS

haha SS i love you brother! but most everything you are saying here is going agenst what Newman teaches... theses statements match the 180 degree reason why we are doing this this way.

buy the way, current will be passed between the parasitic capacitance and the inductance of the coil. This current will grow and at that point all you need is a small push to keep it going. so with that said, yes. you pay for the "start up" but after that its just a matter of that tinny tap...

so yes, the field will grow in each direction, while the current is stopped from going through the coil...!  let me explain again.

parall resonance = open circuit ( when looking from the outside) so this will stop the current flow as much as possible from the battery citcuit. ( high impedance)
parall resonance all ways = maximum current flowing ( between the cap and inductor ) with in the LC and the only resistance is the coil.

I nail this in my DOC.

I would encourage you to go over what i'm saying in my doc ( must read the references) with a fine tooth comb so you ca see why i'm doing the things i'm doing because i do believe that once you see the why, it ill make a lot more sense.

unfortunately your gonna need to take 50% of what you learned about EEE and re think it using physics!

"you cant learn what you think you already know"

we are doing physics. this is the key to our understanding.

"conservation of energy"
I go over this in video 6:

https://www.youtube.com/watch?v=Bo2I6mJ_ssE

Its a river of magnetic flux... put a paddlewheel in it, and let nature generate the river...

will get there, you'll see what i'm saying in due time.

~Russ

@ss come on SS read the doc, I'm writing you my version but its not doing justice to a few paragraphs and its coming out as long as Russ's. Russ gave you 3 great links...please read them in there entirety and slowly with a open mind.
@brad
encourage the thoughts brad stick with us..
Think of the 'shorted' coil more like a pendulum and think how the pendulum gets held at the far sides of the swing while the rotating PM keeps the current going, this gives the impression on the scope the circuit is resonating continuously, and it is but the induced voltage current comes from the PM and the field does not reverse because of it. I'm working on this way of seeing it...hope others see something too.
pictures I d/l from web but they help visualize...
The change in electric potential in the whole ‘short’ circuit must be zero (because it's a loop) - (minus) wire resistance over time. This is important,  but then we are still inducing current into the coil because the rotor is turning and the pendulum falls but not by using the energy of the gravity (potential) so its a net gain and no change in magnetic field.

Russ,

Okay, I see the wire calculations checked out.  I am surprised how low the resistance per kilometer is for the 5 awg wire.

I watched your clip, you did not give enough credit to the sun for the hydro power and the wind power.

Energy spec for the sun:

Quote

Every second, as we have seen, the Sun creates E = 3.9 x 10^26 joules of energy. To balance the books, every second the Sun destroys m = E/c2 = 4.3 x 10^9 kilograms of mass. This mass loss, equivalent to more than 4 million tons per second, is accomplished by fusing 600 million tons of hydrogen into 596 million tons of helium.

For the light bulb the light doesn't really "come from the load,"  Eventually as you increase the voltage the light bulb filament gets to the point where it starts to glow.  The higher the voltage the brighter the bulb.  But the whole time irrespective of voltage the filament is giving off heat by radiating it away as EM radiation within a band of frequencies.  The more power dissipated in the filament the band of frequencies gets higher.  Eventually the frequency band reaches the threshold where our eyes are sensitive to the EM radiation and we "see" it.  But the EM radiation was there the whole time even when we didn't see it.   And the light we can see, and the light we can't see, hits the walls and ceiling and floor of the room and ends up heating the room.

SS

Russ,

Okay, I am going to ask you to play devil's advocate for this one.

You state that you set up a parallel LC resonator and it blocks current flow at the resonant frequency which is fine.  Then you state that it just needs a tiny injection of energy per cycle to keep it topped off which which is fine.

Then you state that some kind of electrical load would drain energy from the LC resonator but you can take advantage of the magnetic field to push on a rotor magnet without draining energy from the LC resonator.

So, can you play devil's advocate and try to explain why pushing on a rotor magnet would in fact drain energy from the LC resonator?

SS

@SS
Can you explain why the rotor magnet keeps going when you know that a SRF LC circuit is going on in the coil during the short...
You have seen the rotor turn...where is the pixie dust that makes it keep going even when the current flow in the coil reverses...
Regards