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: Sharing ideas on how to make a more efficent motor using Flyback (MODERATED)  (Read 350071 times)

gotoluc

  • elite_member
  • Hero Member
  • ******
  • Posts: 3096
Hi everyone,

thanks for all your suggestions.

@verpies, the encoder sounds like the better way to go but then the problem is me not knowing how to program them becomes the next issue.

As for the ESC's recommendations. I am aware and have some that I tested. The problem with them is they output multiple on times operating at high  frequencies between 8kHz to 20kHz and also reverse phases (AC). How do you capture flyback with such switching conditions?

For now I'll stay with my simple plan of using the 3 optical switches. However, I do have a question for verpies about my TCRT5000 optical sensors.
Can the photo-transistor of the TCRT5000 be good enough to trigger the 4047 chip directly, or would the photo-transistor need to trigger a transistor which will then trigger the 4047?

Thanks for everyone participation.

Luc

verpies

  • Hero Member
  • *****
  • Posts: 3473
@verpies, the encoder sounds like the better way to go but then the problem is me not knowing how to program them becomes the next issue.
Yes, that is a problem if you are using a microcontroller (MCU) ...unless someone sends a programmed one to you.
The alternative to MCU is using several of these CMOS chips and programming them in hardware with jumpers and BCD switches ...without software.  I already know you can do it that way, but I don't know if it is any fun for you.

I do have a question for verpies about my TCRT5000 optical sensors.  Can the photo-transistor of the TCRT5000 be good enough to trigger the 4047 chip directly?
Maybe.
The amplitude of the photo-transistor is strong enough to drive the the 4047 with a 1k pullup resistor, but the rise/fall time of its pulse might not be fast enough.  Doing it "by the book" requires a Schmitt trigger such as the 4106, that "squares up" a slowly changing signal.

gotoluc

  • elite_member
  • Hero Member
  • ******
  • Posts: 3096
Yes, that is a problem if you are using a microcontroller (MCU) ...unless someone sends a programmed one to you.
The alternative to MCU is using several of these CMOS chips and programming them in hardware with jumpers and BCD switches ...without software.  I already know you can do it that way, but I don't know if it is any fun for you.

I don't think that's my kind of fun!... but I appreciate you bringing it up and who knows, maybe for a larger version if this test device proves to have an advantage.

Maybe.
The amplitude of the photo-transistor is strong enough to drive the the 4047 with a 1k pullup resistor, but the rise/fall time of its pulse might not be fast enough.  Doing it "by the book" requires a Schmitt trigger such as the 4106, that "squares up" a slowly changing signal.

Will these work the same? http://www.ebay.com/itm/221769763079
If so, could you please draw me a schematic starting with the TCRT5000 to CD40106 to HCF4047BE with all components needed in between so I can order it all one time.

Thanks for your help

Luc

verpies

  • Hero Member
  • *****
  • Posts: 3473
I don't think that's my kind of fun!
You don't know what you're missing ;(

Will these work the same? http://www.ebay.com/itm/221769763079
Yes, but I'd have to draw a different schematic diagram for the 40106, since 40106 are not the same as 4106 (the former are inverting and the latter are not).
That is of course if you choose to build a circuit that uses one of them.

If so, could you please draw me a schematic...
Below are two ways of conditioning the signals from your optos - one is adjustable and the other is not.
Both circuits are using only one chip (4106 or AD8032).

gotoluc

  • elite_member
  • Hero Member
  • ******
  • Posts: 3096
You don't know what you're missing ;(
Yes, but I'd have to draw a different schematic diagram for the 40106, since 40106 are not the same as 4106 (the former are inverting and the latter are not).
That is of course if you choose to build a circuit that uses one of them.
Below are two ways of conditioning the signals from your optos - one is adjustable and the other is not.
Both circuits are using only one chip.

Thanks verpies for the new circuits.

In the first (non adjustable) circuit I see you have 6 opto's... why do I need that if I'm only switching 3 phases?

The second circuit, unfortunately I don't know what threshold and hysteresis adjustments are for or what they would affect when adjusted?
Care to explain in a simple lawmen way.
EDIT I think I got it, see below post.

Thanks for all your help

Luc

gotoluc

  • elite_member
  • Hero Member
  • ******
  • Posts: 3096
unfortunately I don't know what threshold and hysteresis adjustments are for

I gave it some thought and threshold could be to adjust at what point the opto's rising voltage you want it to switch on and hysteresis could be the amount of time you want the on time to last.

Do I pass?

Luc

verpies

  • Hero Member
  • *****
  • Posts: 3473
In the first (non adjustable) circuit I see you have 6 opto's... why do I need that if I'm only switching 3 phases?
You don't but the the 4106 chip has 6 channels so I had to draw them.
You can use only 3 channels and leave the remaining channel inputs grounded.

verpies

  • Hero Member
  • *****
  • Posts: 3473
I gave it some thought and threshold could be to adjust at what point the opto's rising voltage you want it to switch on
Yes

Do I pass?
1 of 2

... and hysteresis could be the amount of time you want the on time to last.
No, although hysteresis affects the on-time a little. 
On Diag.17, Hysteresis is the vertical distance between Threshold 1 and Threshold 2.

BTW:  If you connect the conditioned signal from an opto to the 4047 chain, then the on-time of the conditioned signal will not matter anyway, because the 4047 chain pays attention only to the rising edge of this signal and stretches its final pulse width to whatever you set it to with the Width1 pot.

gotoluc

  • elite_member
  • Hero Member
  • ******
  • Posts: 3096
Thanks verpies for making everything clear ;)

You're a good teacher!... I've learned more in the past few weeks with your component suggestions and explinations then I have in years.

I appreciate your help

Luc

itsu

  • Hero Member
  • *****
  • Posts: 1845

I finally completed the "pulse sequencing circuit" as designed by verpies here:

http://overunity.com/16167/sharing-ideas-on-how-to-make-a-more-efficent-motor-using-flyback-moderated/msg470273/#msg470273

The resulting diagram is slightly different due to some inferior components (zeners), so we had to add an extra MOSFET (Q5), see below diagram.


The goal was to have a circuit which smoothly controls a pulse motor and to manipulate/calculate the flyback spike from the rotor coil (L1) to be able to put it to good use lateron.

For now we settled for each cycle to:

# short out the capture capacitor C2,
# capture the spike in this emptied capacitor C2
# measure the recovered voltage in this capacitor.

Verpies also designed a spreadsheet with a pie chart which, after plugging in the measured data, visually shows the measured data
and calculate the energy distribution in this pulse motor including the recovered energy in the C2 capacitor.

Be aware that its an European based spreadsheet, so the decimal point is a comma!

I made a video which in 4 pieces shows/explains:

# the diagram and physical setup
# the controls/signals of the 4047 chips
# the signals (voltage/current/timing/capturing) on a running system
# the input power measurements for inputting into the spreadsheet for a running system.

video here: https://www.youtube.com/watch?v=UlEl1V44pPE&feature=youtu.be

Below screenshot 1 is the partial (on time only) input power measurement on a running system
Below screenshot 2 is the partial (on time only) input power measurement on a stopped system

The first picture is a snapshot of the spreadsheet made with the above inputted data and shows the energy distribution in a running system.
Note that the energy which is unaccounted for is probably mostly the kinetic energy stored in the running rotor, but this would be very hard to measure (and thus account for).

The second picture is a snapshot from a stopped system where there is much less unaccounted for energy.

A big thanks to verpies for designing the circuit, the spreadsheet and to assist in ironing out the encountered problems.
Hopefully we are able to further zoom in on certain energy streams and to put the captured/recovered flyback  pulse energy to good use.


Regards Itsu

gotoluc

  • elite_member
  • Hero Member
  • ******
  • Posts: 3096
Thanks Itsu for taking the time to complete the circuit, making a video demo and share your results.

Top notch!

Thanks also goes to verpies for his helping us to get this done.

On my side, just minutes ago I received the last components to complete the opto circuits.

I hope to have something to share next week.

Luc

itsu

  • Hero Member
  • *****
  • Posts: 1845

Thanks Luc,

glad you liked it.

Looking forward to your build/results.

Itsu

gyulasun

  • Hero Member
  • *****
  • Posts: 4117
Hi Itsu,

Great work, thanks for showing the results.  Thanks also goes to Verpies of course.

The spread sheet data clearly show that in coil L1, about 45-50% of the input power is lost. This indicates again the importance of using thick wire whenever there is room for it to minimize copper loss in coils.

Gyula

itsu

  • Hero Member
  • *****
  • Posts: 1845

Thanks Gyula,

so i have to build myself a 38mH coil which instead of 10 Ohm has a much lower resistance, like 1 Ohm.

Any suggestions on how or what type of coil that might be (which core)?

Regards Itsu

Over Goat

  • Newbie
  • *
  • Posts: 20
Thanks Gyula,

so i have to build myself a 38mH coil which instead of 10 Ohm has a much lower resistance, like 1 Ohm.

Any suggestions on how or what type of coil that might be (which core)?

Regards Itsu

http://peakoil.com/forums/new-electric-motor-smaller-cheaper-more-powerful-t68369.html#p1152117