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 429858 times)

listener192

  • Sr. Member
  • ****
  • Posts: 272
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1335 on: October 03, 2018, 10:04:23 PM »
So the stator cross sectional area must be greater than or equal to that of the rotor? Why aren't the pole ribs a flux limiting bottleneck if they have the smallest area?

If I have digested your posts correctly, then the coils at either end of the rotor are used for EMF collection. So can the same results be achieved with 120 degrees of stator (9 coils) and a half length rotor, or even 90 degrees (3 stator coils) with extra coils also at both ends of the rotor? I ask because this could cut the cost of the electronics by 60 or 70 percent. It could also allow for simple fabrication of the array of stator segments. I'm thinking bars machined over the length with a bull nosed cutter then welded together in a radial arrangement so the coil field focal point is the end of the half length rotor.


Hi FixedSys,


Yes, the FEMM plots confirm that the stator CSA has to equal or exceed the rotor CSA, if you want most of the stator flux to cross the rotor.
5HP upwards motors, have greater stator thickness in proportion to the teeth depth and offer that larger CSA.


In the thin stator FEMM plot, the stator thickness (both sides combined) is equivalent to about 4 teeth thickness of the 5 teeth shown, so the flux impediment is the stator. The stator enters saturation before the rotor even gets close.

In the thick stator FEMM plot, the stator thickness (both sides combined) is equivalent to about 8 teeth thickness, so the flux impediment is the teeth feeding the rotor, however we can still easily get the flux level through the rotor up to saturation, which for M15 steel is about 1.5T 
 
On the face of it, other geometries could offer improvements however, not something I am considering at this time, as this would be introducing another variable into a mix that we still don't understand.


Regards
L192



listener192

  • Sr. Member
  • ****
  • Posts: 272
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1336 on: October 14, 2018, 08:28:27 PM »
Attached.

1. Output power @15V DC input
2. Unfiltered output voltage showing recovery voltage transitions
3. Unfiltered output voltage driven to maximum. Note: stator in full saturation, as predicated by FEMM simulations.
4. Picture of finished Boost H bridges x 15 driving opposite stator coils in parallel. Note: control/driver hardware behind the bridge boards.
5. Super cap recovery current (yellow) and voltage(about 25V DC for approx 15V DC input) [/size]

7 coils either side are energised (2 pole), 5 slot pitch. Rotor covers 6 poles.

Some further experimentation with software planned however, main push now will be to find new stator from 5HP to 10HP motor.


Highly inefficient at this time, as saturation of stator causes coil inductance to fall rapidly above 10A total input draw. If the stator was thicker and did not saturate, the coil inductance would be higher and recovery voltage/current would increase.


The maximum input on this test was 47A @ 15.5V DC.


The H bridge and recovery circuits remain cool. 4 Ohm resistor (large heatsink), gets warm.


The other two 58A diodes on heatsinks are...


A. Isolation for DC Switch Mode power supply input.
B. In series with 4 Ohm resistor to block initial charging of 27F super cap bank.


The bank charges initially through the H bridge isolation diodes when the HSS switches are turned on.
As the circuit runs straight away the bank charges fast through the recovery diodes, as well.


At the moment the parallel connected coils are 180 degs apart.
There maybe some improvement in rotor coupling by offsetting the coils, so they couple through through the rotor by the shortest path.


L192

gotoluc

  • elite_member
  • Hero Member
  • ******
  • Posts: 3096
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1337 on: October 14, 2018, 09:21:33 PM »
WOW L192, you are the most dedicated true builder I have ever seen so far.

Thanks for sharing
Kind regards
Luc

listener192

  • Sr. Member
  • ****
  • Posts: 272
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1338 on: October 26, 2018, 03:11:24 PM »
Just an update on the analysis of my scheme as it stands...


The move to 30 x  H bridges driving  parallel coils was largely influenced by the need to increase coil current drive, as the inductance of 6 coils in series was limiting current, demanding a higher supply voltage and this device needed to function on 25V DC. As I wanted to use a 2 pole scheme, H bridges seemed the best way to proceed.



Setting aside the my stator saturation issues, identified in FEMM, the main problem is input current not contributing to output.
----------------------------------------------------------------------------------------------------------------------------------------------


The 50/60Hz output is a composite of discrete MMF steps with individual pole reversals only occurring after pole 15 is switched.


The large static current draw is present with or without the rotor, and is due to the low inductance seen for each pair of parallel coils, switched by the individual H bridges. 


The very low inductance is largely a product of  each coil pair being switched on for 6 steps. This maintains the continuity of MMF however, once inductance is overcome (within the first ON step), very little current is required to maintain a static magnetic field, so in this case, the applied power is largely wasted as heat for at least 5 step periods.


----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
The scheme Pierre showed for series parallel operation, was really suited to 4 pole operation, (if you work through it), with 6 coils per pole and three coils in each pole in parallel.


As 2 x 3 coils are energized in series, every time the pole advances, there is a new coil switched into each chain. This make a change to the series inductance, as the new coil(s) charge, while maintaining the pole MMF.

It should be noted that this circuit is effectively a current multiplier* and the voltage increase on the cap bank, over input voltage is constrained by the feed back to the input via the 4 ohm resistor.
Coil energy aided by the boost function of the DC supply rail when the coils are released, the current developed charges the next coil to be activated. The continued current from the supply ensures series coil current is maintained, which in turn maintains MMF.

*I have found an example of a current multiplier that uses multiple pulse transformers that has similarities however, no examples of this this exact configuration.

This direct reuse of coil energy is an important feature of how the device achieves its performance and also explains why the coils have to be kept in a series configuration within an active pole group.
Step charging an inductor is largely a lossless process when the supply is a voltage source. 
The trailing coil(s) in a pole group are being sequentially switched out of the pole group, with the energy being recovered into the cap bank via the LSS body diodes and the recovery diodes. 
This arrangement limits the power wasted as heat.

So, the challenge is to modify my scheme, so the individual coil pairs, see inductance change at every step but without a dropout in the composite MMF i.e. an unbroken sine distribution.


L192


   
« Last Edit: October 27, 2018, 12:36:09 AM by listener192 »

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1339 on: October 27, 2018, 02:28:51 AM »
WOW L192, you are the most dedicated true builder I have ever seen so far.

Thanks for sharing
Kind regards
Luc
It's really too bad that all that dedication, talent, knowledge, effort and money is being spent on chasing a hoax.

AlienGrey

  • Hero Member
  • *****
  • Posts: 3713
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1340 on: October 27, 2018, 09:29:23 AM »
It's really too bad that all that dedication, talent, knowledge, effort and money is being spent on chasing a hoax.
What you mean ? exactly please explain,   Faffing about with inductors can be a mugs game adding bits
and experimenting is best on a one-off prototype arm of the original as practical what was thought to be a good
idea can often turn to a bad idea if not properly tested out first and a waste of resources as we experimenters often
find  ;D ;D ;D good though ain it!

forest

  • Hero Member
  • *****
  • Posts: 4076
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1341 on: October 27, 2018, 09:58:50 AM »
Start small please. Only if you have something make it bigger. Don't waste your time, don't waste money. At anytime mechanical switching is better using rotary device.

listener192

  • Sr. Member
  • ****
  • Posts: 272
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1342 on: October 27, 2018, 11:13:33 AM »
It's really too bad that all that dedication, talent, knowledge, effort and money is being spent on chasing a hoax.
My day job gainfully employs any technical knowledge/ability I may have. My heretical activities are conducted at night under the cover of darkness!
Seriously though, like many on this forum I have spent the last 20 years working on either patent replications, online or my own ideas and so far I have not seen anything close to over unity however, everybody is entitled to a hobby, so think how much money and time I would have spent if that hobby was golf! Some might say that putting a ball around on a green for hours on end is a waste of time and money.
Hobby's are subjective activities, so each to their own! 

L192

listener192

  • Sr. Member
  • ****
  • Posts: 272
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1343 on: October 27, 2018, 11:34:36 AM »
Start small please. Only if you have something make it bigger. Don't waste your time, don't waste money. At anytime mechanical switching is better using rotary device.
Mechanical (rotary) switching, would have the advantage of robustness and not having to worry about protecting semiconductors however, if the success of a device relies on the scale of replication of building blocks, for example a closed loop of circuit building blocks, then rotary switching would become complex and quite a build in its own right.
If you look at my earlier posts, you will see the building block (boost circuit) I am using in this scheme was tested in prototype form before replicating 30 times, which is a sensible path to take.
The discrete electronic route I have taken is a large improvement over previous attempts i.e. relays and two different H bridge boards. The current scheme can generate  voltage over 30Hz to 120Hz  range (limited by delay settings) without reduction in amplitude. In the previous attempts the high series inductance was the limiting factor.

A final point... I realize that many may not have the resources to commit to such an expensive build however, I don't see why that should cause me to limit my expenditure on this project, after all who would bother to comment on this if I were spending the money on golf, or sailing! 

L192

myenergetic

  • Newbie
  • *
  • Posts: 12
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1344 on: October 28, 2018, 11:23:11 AM »
Quote from: listener192

« Reply #1343 on: October 27, 2018, 11:13:33 AM »

"everybody is entitled to a hobby," Hobby's are subjective activities, so each to their own"

Well said and thank you for sharing benevolently
jj

listener192

  • Sr. Member
  • ****
  • Posts: 272
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1345 on: October 29, 2018, 10:21:37 PM »
Attached are some scope shots from the latest scheme.
Still two poles 7 coils each (opposite parallel).

The Arduino digital IO has been extended by treating the analog pins as digital outputs so 60 pins now available.
Individual control of HSS and LSS switches in each of the 15 H bridges.

Links have been placed between the coils and 7+7(parallel) coils are driven in series.

Example:
HSS 1 left = on LSS 7 right = on. 
LSS 8 right = on, LSS 7 right = off.
HSS 2 left = on, HSS 1 left = off.

and so the sequence moves forward.

Unfortunately the use of H bridges doesn't allow the sequence to continue unbroken so 1/16 & 15/30 are not linked.
When the leading coil of the sequence of 7, reaches coil 15/30, the trailing edge coils are sequentially turned off at the same time a reverse sequence starts at coil 1/16.

The feed forward boost into the leading coils in the sequence, terminates prematurely and this is seen in the waveform.

The unfiltered output voltage waveform is starting to look closer to that of Pierre's however.

I ran this at 30V DC input and the no load input current draw was only about 3.5A

Output was about 150W approx, so much better than previous schemes.

I should mention the recovery current was much higher in proportion to the input current averaging about 5A for under 9A input,
 whereas the previous scheme was producing 10A but for an input of 40A.

More testing required before deciding the next step.


 L192

FixedSys

  • Newbie
  • *
  • Posts: 28
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1346 on: October 30, 2018, 05:38:50 AM »
....This make a change to the series inductance, as the new coil(s) charge, while maintaining the pole MMF.

Great work. This is an important discovery that I had missed. This is changing the intensity which in addition to the movement of the field will contribute much to the overall rate of change factor in Faraday's law.
Are we close to being able to put together a timing diagram that describes the interaction of all these factors?

listener192

  • Sr. Member
  • ****
  • Posts: 272
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1347 on: October 30, 2018, 11:10:27 AM »
Attached is a doc that shows part of the sequence. This is only an example as my scheme has 7 coils in circuit at all times.
The failing of this scheme is the requirement for the H bridges to reverse coil current every 180 degrees.
MMF is still generated by each coil in the reverse chain however, the boost function is not realized. This may be seen in the recovery current scope shot in my previous post. Note the absence of recovery current during the negative going part of the output waveform.

The forward progression of boost current only is fully realized while there are 7(+7)  coils active, so 9 out of the 15 steps, each 180 degrees.

I should point out that the changes to produce this were easy and so I tried this first.

Splitting the coils again into a series loop with half bridges and parallel sections as Pierre has suggested, will take much more time to work out. 




L192
« Last Edit: October 30, 2018, 09:30:34 PM by listener192 »

listener192

  • Sr. Member
  • ****
  • Posts: 272
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1348 on: October 31, 2018, 03:38:24 PM »
Great work. This is an important discovery that I had missed. This is changing the intensity which in addition to the movement of the field will contribute much to the overall rate of change factor in Faraday's law.
Are we close to being able to put together a timing diagram that describes the interaction of all these factors?
The stator composite MMF, producing the output waveform needs to be varying in amplitude at a 50/60Hz rate but not dropping to zero apart from the zero cross areas of the sine.

The individual MMF's comprising each step  created by the current through the individual coils, must therefore also not drop to zero however, if they stay static a lot of current is drawn once the inductance of the coil is overcome. By charging  a group of coils in parallel or series parallel and discharging all in series into a leading coil, current is maintained through the charge group  coils until the next step.

I have looked at Pierre's switch scheme drawing out 20 sections or so on paper. I have figured out how to parallel/series charge 6 coils and then to series discharge into a single coil.

What I am having difficulty with, is completing this action in reverse direction to generate the opposite pole. The traveling direction need to be the same as the other pole for the switching sequence but the current direction needs to be in the opposite direction.
Two completely separate switching circuits with separate coils could accomplish this easily, but you have to be able to visualize a scheme that will work with one set of coils.
L192

shylo

  • Hero Member
  • *****
  • Posts: 540
Re: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated
« Reply #1349 on: November 01, 2018, 12:14:27 AM »
Mechanical switching can get very complicated ,I've tried

Electronic is the way to go if you have the knowledge, which sadly I don't.
Using the reverse fields is part of the solution I believe
Like you say I don't golf, fish, hunt, take trips, I enjoy tinkering to each his own.
Thanks for sharing and good luck
artv