Self accelerating reed switch magnet spinner.

[MileHigh]

TK:

On your new clip, you can see the limitations of your small relay-based sense coil.  The waveform is just too flat between magnet passes.  If I was in your shoes I would be tempted to try an alternative sense coil.  I would think of going to Radio Shack and buying one of those fist-sized spools of very light gauge speaker wire.  That should give you a much wider spread in the sense coil output waveform and get rid of the flat spots.  I am guessing you already have something comparable floating around.

MH, your AC voltage integrator is great... but will it work for a spiky, pulsed DC signal too? But I'm not sure if it is applicable here because any power taken off will be through a diode and put onto a big cap anyway, and there are losses in that process for sure.

The voltage averaging circuit that I described is just another piece of the "convert your cheapo multimeter into a watt meter" idea that I had.  It would do current sensing on the supply battery feed into the drive coil.  It would also do current sensing on the output from the drive coil into a charging battery, i.e.; back to a conventional Bedini-type setup.  There is not really any spiky-ness in those current waveforms and the bandwidth of the op-amp is 3 MHz.  So there should be no bandwidth issues.   Later on I will try to do a simplified schematic and description.

And your explanation of the torque point is good, for the max RPM case under no load. I'm not sure if the same timing and dwell parameters will apply to the loaded rotor though, since it will stabilize at a slower speed.

Not sure I get you here in the sense that the torque on the rotor as a function of the angle between the energized drive coil and a moving rotor magnet should be independent of load.

MileHigh

[synchro1]

MH:

I agree but,  (always a but) does the frequency not have to increase also or, in TK's set-up, does that happen automatically like in a Bedini motor?  In other words, on mine, I have to change the vr's to get it to accelerate.  This can be done continually until it reaches the max rpm for that configuration.

Perhaps you can explain this:

http://www.youtube.com/watch?v=LfprTzG5SY4

I never was able to figure out why this happened the way it did.  Some folks on Youtube offered their opinions which may, or may not have been correct.

This probably has nothing to do with TK's motor and for that, I am sorry.  (But it might)

Bill

PS  I had mounted the coil on a threaded rod assembly such that I could, with some precision, move the coil's position to the rotor.  I had already found the "sweet spot" in my earlier videos.

Acceleration coupled with increased resistance is relevant. I'm not asking anyone to rewatch any of my excorcist videos. My Bedini circuits behave the same way. What I demonstrated was that once the current tapering produces the maximum acceleration, a loop back to source circuit proves there is no longer any drain on the source battery.


You could try a trifilar wraped power coil to accomplish this. A bifilar, power and trigger coil for the bedini circuit and the third wrap for output. The output would need to pass through a rectifier, then connect directly back to the source battery. Once the reduced resistance produces the maximum acceleration, the loop back circuit creates a zinging effect because power can't run two ways in a wire silmultainiously. A push and pull results. Gadgetmall just wraps the third coil directly over the bifilar.

[MileHigh]

Synchro1:

What I demonstrated was that once the current tapering produces the maximum acceleration, a loop back to source circuit proves there is no longer any drain on the source battery.

One of the important lessons for anyone to learn when working on the bench is to question their own results when they seem strange or unusual.  That could mean double-checking what you just did or trying to make the same measurement in a completely different way as a means of confirming or denying your first measurement.  Also, some measurements are very difficult to do and it requires a lot of knowledge and skill.  How many clips have you seen where a digital multimeter display is jumping all over the place and the experimenter just decides to "pick" one as the measurement.  That's a huge mistake, and the right thing to do would be to ignore the crazy display jumping all over the place.

I can tell you with 100% certainty that there was still a drain on the source battery when your setup was running at high speed.  If you had a big capacitor in parallel with your battery when you thought you were measuring no current consumption, and then disconnected the battery and let the capacitor take over, then the voltage on the capacitor would drop which would prove that your setup was drawing current.

You could try a trifilar wraped power coil to accomplish this. A bifilar, power and trigger coil for the bedini circuit and the third wrap for output. The output would need to pass through a rectifier, then connect directly back to the source battery. Once the reduced resistance produces the maximum acceleration, the loop back circuit creates a zinging effect because power can't run two ways in a wire silmultainiously. A push and pull results. Gadgetmall just wraps the third coil directly over the bifilar.

One think to note about bifilar and trifilar setups around a core, and that is that they are all sharing the same available power.  So if you start drawing some power away with one coil, then there is less power available to the other coils.

MileHigh

[synchro1]

Pirate's turning the power down and his rotor's speeding up. I got the same effect from my simple Reed switch version. I went on to report that after the current dropped to zero on the amp meter, the acceleration went hypersonic. I would really love to see your Op Amp circuit run a high speed VCR rotor like Pirate's to see if TK could repeat that effect with scope shots.

[TinselKoala]

@synchro:
Pirate is turning a variable resistor. This doesn't necessarily mean he is "turning the power down" when the rotor speeds up. But even if it does, it could only be something like desaturating the transistor and allowing it to switch better _instead of heating up so much_ on the same or less power input.

We hear a lot of impressive sound from Bill's closeup microphone, but I didn't see any instrumental readings of RPM or frequency of operation in that video. I also don't know the operating voltages that are being sent to the drive coil. A VCR head has good bearings and is well balanced, but installing glue, magnets and tape around the outside will destroy the balance and add aerodynamic drag. There is a lot of aero drag on my very poorly balanced Folger's Instant Coffee jar lid too, but the pivot bearing system is as good as a ball bearing system when it is properly set up. My rotor, which has a relatively low mass, takes a surprisingly long time to run down, unpowered, from its normal 2100 RPM speed: 94 seconds is typical, in fact.
I have no doubt that Bill's rotor is turning faster than the MHOP rotor is. I sure would like to see what his scope shows on that system. I refuse to tear apart my good old VCR, though, and I doubt if I can even find one at a junk store any more. I'll ask around though.

@MH:
I've tried an assortment of different sense coils, short and fat, wide and narrow, high and low impedance, even a loopstick with the core removed. For the coreless coils the relay coil I am using works the best! But for all the coils the width of the magnet passage sinusoid is pretty much the same. I think that it is the magnet spacing that determines the width of the flat spots more than the coil geometry, and I found this to be the case with core-effect  toroidal coils during the Steorn Orbo investigation as well. If I had the ability to make an 8-magnet rotor of the same diameter I think that the flat spots would be gone. Perhaps even a six-magnet rotor using the magnets I used would do it.
But it's really not an issue. Even though the slope of the triggering signal is very flat at what would be a 50 percent duty cycle "start" position, the op-amp is still able to discriminate and trigger stably there. Most of the fluctuation you see in the videos is probably caused not by the flat slope itself but rather because the level of the flat signal is jumping around a bit due to the different magnet strengths and out-of-balance rotor. Still, I can get a 50 percent duty cycle, just not with the very sharp transition as is possible with say 60-40.