Storing Cookies (See : http://ec.europa.eu/ipg/basics/legal/cookies/index_en.htm ) help us to bring you our services at overunity.com . If you use this website and our services you declare yourself okay with using cookies .More Infos here:
https://overunity.com/5553/privacy-policy/
If you do not agree with storing cookies, please LEAVE this website now. From the 25th of May 2018, every existing user has to accept the GDPR agreement at first login. If a user is unwilling to accept the GDPR, he should email us and request to erase his account. Many thanks for your understanding

User Menu

Custom Search

Author Topic: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated  (Read 429872 times)

listener191

  • Sr. Member
  • ****
  • Posts: 253
Just a Falstad simulation for half bridges simulating the coil 1 coil 2 period.

Green trace in second file is voltage across 1st MOSFET
There is some high frequency oscillation happening, which diminishes if I snub the source-drain.
Only clocking at 15Hz, but no large voltage peaks.

Current is across the 2nd inductor, which increases when the 2nd switch is turned on.

Regards
L192

r2fpl

  • Hero Member
  • *****
  • Posts: 744
listener191:  falstad has a lot of mistakes! ex. When changes to much anything does not work properly.  Help: select all, copy, reload page F5 and paste.

listener191

  • Sr. Member
  • ****
  • Posts: 253
Now this is the picture when you remove all of the lower diodes, similar to the Pierre relay case.
Now there is a 206V pk spike when a coil turns off, which also would be the case if it were switch 2 staying on and switch 1 turning off. If you turn off all coils that's a 388V pk spike.

So the problem with Pierre's waveform looking so spikey, is lack of diodes in parallel with the low side relay contacts. Basic he has not provided any solid current path for the coils to discharge.

L192

konehead

  • Sr. Member
  • ****
  • Posts: 462
hi Jeg
In message 249 you were asking about all the steps in voltage you see on scope between periods -
what I think all this is, is a "shorted" coil effect from his winding pattern and method, especially when the two opposing coils pass through the same slot in opposite=flow directions....this creating a bucking field, which slams the magnetic fields against each other, stopping the flow abruptly, and creating the bucking condition which also creates a shorted-condition more or less in both coils at once and this causes those hyper-spikes in voltage.....

these spikes in voltage also go "both ways" they are not like backemf/flyback spikes so much....rather ringing oscillations are created when coil is shorted; in Pierres case it is not so much a simple short of a coil by connecting coil leads together for instant with switch at the sinewave peak.......instead Pierre has more complicated (somewhat) = at least not so understandable method to create hyper-ringing spikes to his generating coils via his interweaving winding method and also passing coils through same stator slot in opposite directions
"perhaps" this has lots to do with the way it loops, and fills up the caps so well without lenz law or CEMF cancelling forces which normall occur in generators or transformers when loaded...

Normally you will see a gain in voltage into caps X20 when switch-shorting at sinewave peaks, a generator coil, X50 if all things perfect..
.
So this my konehead- theory it might have some merit might not but those spikes sure resemble peak-shorted spikes...

gmolina

  • Newbie
  • *
  • Posts: 8
Hi all,imho, in Pierre first video between minutes 5:09 and 6:19 i see that the core of the stator is very lightweight, out of normal, Piere can manipulate it of very easy way, the orange plastic that support winding insulators because aparently there are ausence of iron that can support it, and texture of the estator core make me think that this is a resin made core and not a iron or steel core. Maybe i could be wrong, but i hope i'm not.

Regards.

GM

listener191

  • Sr. Member
  • ****
  • Posts: 253
Hi Gmolina,

No sorry you are incorrect..see attached
You can clearly see the stator is laminated steel.
The orange tube is just to pack the airgap to stop the rotor vibrating, which I can confirm it does, unless the air gap is firmly packed.

Regards

L192



listener191

  • Sr. Member
  • ****
  • Posts: 253
Seems  my reported demise of a L298N after extended running at 4A was premature.

The device had actually entered thermal shutdown and it took some time to recover.

I tested the board the next day and all was well.

This demonstrates they are pretty robust.

Regards

L192

r2fpl

  • Hero Member
  • *****
  • Posts: 744
The current flowing through the relay is about 8A or more. The effect may not be below a certain value. It's better to use more robust layouts.

listener191

  • Sr. Member
  • ****
  • Posts: 253
The current flowing through the relay is about 8A or more. The effect may not be below a certain value. It's better to use more robust layouts.

Hi R2fpl,

Yes I totally agree.

I am waiting on one of the BTN7960B boards for evaluation.

It will have a 28V rail limit, so I want to see how much current I can get to pass through two sets of coils before deciding.

Also I think it wont sustain 10A on the heatsink supplied, with out thermal shutdown, so that's  another area of experimentation,

to see how the boards could be bolted on either a large heatsink plate or a thick aluminum angle rail.


Regards

L192

T-1000

  • Hero Member
  • *****
  • Posts: 1738
It will have a 28V rail limit, so I want to see how much current I can get to pass through two sets of coils before deciding.

Also I think it wont sustain 10A on the heatsink supplied, with out thermal shutdown, so that's  another area of experimentation,
It depends how far you would like to go on heating coil wires.

In Pierre's setup the input was 110V 1.5A which translates to 165W in total and supercaps was on 20V charge. Which translates to 8.25A consumption in total across all coils minus BEMF recovery.
And for a sake of test how much you can get from artificially created rotating magnetic field it does not need to consume so much power.

Cheers!

listener191

  • Sr. Member
  • ****
  • Posts: 253
It depends how far you would like to go on heating coil wires.

In Pierre's setup the input was 110V 1.5A which translates to 165W in total and supercaps was on 20V charge. Which translates to 8.25A consumption in total across all coils minus BEMF recovery.
And for a sake of test how much you can get from artificially created rotating magnetic field it does not need to consume so much power.

Cheers!
Hi T-1000

There are three current paths, so each high side and low side switch would be conducting 2.75A based on the assumption that all of the power is coming from the supply, which is well within a paralleled L298N's rating.

Pierre alluded to a much larger current per switch pair (not stated though), so perhaps the recovery is much higher than we can see at this time and hence the coil current is higher.
If he could give us a ball park coil current, that would be helpful in a number of ways.

Regards

L192

seaad

  • Sr. Member
  • ****
  • Posts: 311
Builders You need a gross value of 900 Watt to make your magnetic field,  if you don't find a method to recover power to the electromagets!

And Pierres OU ???

Regards Arne

listener191

  • Sr. Member
  • ****
  • Posts: 253
Builders You need a gross value of 900 Watt to make your magnetic field,  if you don't find a method to recover power to the electromagets!

And Pierres OU ???

Regards Arne

Hi Seead,

At such a low clock frequency DC resistance dominates so...

6 of my coils in series are about 6.41ohms, so two of these series chains are fed in parallel, so 3.205ohms  .. 21V/3.205ohms = 6.55A  x 3 paths = 19.65A,  closer to 412W.

Regards
L192   

onielsen

  • Newbie
  • *
  • Posts: 28
To prevent melting the insulation or the wires stay below 2000A/in2 = 3.1A/mm2. Source of this information:
http://what-when-how.com/electric-motors/current-density-electric-motors/
Here are more on designing stator windings:
http://what-when-how.com/electric-motors/design-equations-electric-motors/

There has to be enough inductance in the coils to limit the current at the frequency used. If using resistance to limit the current place external resistors rated for dissipating the required amount of power. This of course is a total waste of power. External inductors can also limit the current. Inductors don't dissipate the power as the energy is given back at demagnetization. Better design the windings for the job without dissipating power as heat.

Regards
Ole

Jeg

  • Hero Member
  • *****
  • Posts: 1532
IRF3205 / IRF4905

Max values
55V, 200W, continuous Id=52A/80A, Pulsed I=260A/390,  Rds(on)  8mOhm/20mOm

An interesting mosfet matching and cheap. With a capable heatsink it will run fine

http://www.irf.com/product-info/datasheets/data/irf3205.pdf
https://www.infineon.com/dgdl/irf4905.pdf?fileId=5546d462533600a4015355e32165197c

As for clamping diodes i was thinking about UF5408. 1000V/3A. Any other idea here?