Tinman's Rotary Transformer

[tim123]

Fisher and Paykel Smart Drive Motors:
- http://www.theinventory.orconhosting.net.nz/smartdrives.htm
- http://www.watchtv.net/~rburmeister/smart.html

[tim123]

Test Using Pulsed DC

I tested my motor using pulsed DC:

- Timing controlled by signal generator
- Power switched by DC solid-state-relay (Teledyne 603)
- I included a smoothing capacitor in parallel with the power supply this time too.

Frequency
- I ran it at only one speed: 10,000Hz.
- 1000Hz brings the motor to a stop
- My SSR can do up to 25KHz max

Duty
- I used a duty cycle of 90% - which was the minimum that worked apparently without affecting performance at the 10KHz frequency.
- A lower duty-cycle slows it a lot.


Rotary Transformer Config

43 Hz - No Load
- 12.5v
- 1.28a

43 Hz - With Load
- 12.2v
- 1.24a

60 Hz - No Load
- 15.0v
- 1.33a

60 Hz - With Load
- 14.8v
- 1.27a


Standard Config

43 Hz
- 15.2v
- 0.57a

60 Hz
- 19.6v
- 0.6a


Conclusions

The results seem slightly better than with plain DC, but it's not a lot, and it's probably within the range of measurement error. The smoothing cap may well affect the readings.

The motor runs really badly on a low duty, or a low frequency.

The power from the load - as measured by LED brightness - didn't seem significantly different.

Given how the motor works, I still can't really see a way that pulsing the power - in any way - can help it's performance... But I see TM's still getting some interesting results I can't duplicate...

It's my 100th post apparently... 

[tim123]

Hi Folks,
  I've been thinking about this, and I really think that the powered-rotor is the Achilles-heel of the system, as far as OU goes... It is the thing that allows the Positive-Lenz effect to show-up though, by allowing switching...

1) Input power... We really want a system that doesn't need any.

2) The power available at the stator is basically equal to the change in flux.
   The change in flux is caused by the input current being re-routed by the brushes.
   So the input & output are too closely related...
   My feeling is they are * directly * related. Which means UU, not OU...

(Note: In the standard configuration of the motor there is Negative Lenz at work in the rotor, as it switches against the stator field.)

Also, I re-thought my comments about coil size, and I don't think they were right...


Field-Switching at the Stator-Face...

It's the * switching of polarity * of the magnet at the center of the stator that causes the effect...

Ideally we'd be able to switch the polarity of the rotor segment without the stator's field being able to affect it at all. In reality, who knows...

I also want to explore the possibility of using PMs for the rotor a bit more. I have a new idea...

The diagram below shows a stator / rotor arrangement.

- The idea is that the PMs * are * on the rotor this time - and in pairs.
- The stator core has 2 extensions: one aligned with the rotor's 'N' magnet, the other aligned with the 'S'.
- The rotor approaches the stator, with one of it's magnets near the extension.
- As the rotor passes thru the center of the stator, it 'switches track', the other magnet comes near the extension, and the stator sees the change in flux we're looking for.
- The stator field will do it's Positive Lenz effect, attract the incoming magnet & repel the outgoing one.


The Sticky Point...

The problem with this is that the switch-over point, where the 2 magnets meet the 2 extensions, might be really sticky.
- The magnets are making a circuit
- The coil field is pulling the magnet back.

There are solutions to this problem - i.e. timing the output instead of relying on a diode... It only gives a half the time to take power out though. I think.

The stickyness may be equal to, or less than, the thrust given by the +ve Lenz effect. In which case it's not a problem. A generator with no reaction force would be a great result...

[tim123]

Attached is a diagram showing a number of the above coils & cores in sequence, with rotor magnets shown too.

Coils are the dark circles, cores are rectangles, magnets are squares.

There are 2 sets of coils that would be active alternately. So it would be producing electricity and thrust 100% of the time.

It's a lot easier to build than my last idea.

But the sticky point might be really sticky... 

But it is counteracted by the +ve Lenz, and with this arrangement - with all the cores connected - it could all balance out...

Actually, I'm finding it hard to work out if the coils clash, or help. I'm a bit slow sometimes... I'll have to do a step by step set of diagrams.

[NTesla]

Here is my Replication test of Tinman's Rotary Transformer:

http://www.youtube.com/watch?v=IkH8jfGq0yM