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: Pulse motor build off time.  (Read 115527 times)

tinman

  • Hero Member
  • *****
  • Posts: 5365
Pulse motor build off time.
« on: November 15, 2014, 07:18:57 AM »
Russ from RWG reserch will be hosting the PMBO this year by him self,as i just dont have the time to put into it this year.

But i am going to do a build,and would like you guys to join in and have a go. Lets see who has got building skills to go along with the knowledge.
TK,FarmHand,poynt99-every one,lets see what ya got :D

Here is the start of my build.
The Megga-G

https://www.youtube.com/watch?v=sT7DrGEWyWc

ramset

  • Hero Member
  • *****
  • Posts: 8073
Re: Pulse motor build off time.
« Reply #1 on: November 15, 2014, 10:08:46 AM »
UUM
Has Tinsel ever joined your contest ?


I will thro some dough his way [small budget]...
or donate to some " Koala sore paw  foundation"


If he will run?


is there a prize??
thx
Chet
Ps
Are Bedeni Type motors allowed ,or P.Lindemann /Aarons work ? If so I will invite  ?
and also Rob Mason's [Evolvingape] Pulseometer  fluid pulsed mechanic power system  ??



PPS
gentleman's rules ?
No Smash talk??
or no holds Barred ??
 

tinman

  • Hero Member
  • *****
  • Posts: 5365
Re: Pulse motor build off time.
« Reply #2 on: November 15, 2014, 12:22:38 PM »

MileHigh

  • Hero Member
  • *****
  • Posts: 7600
Re: Pulse motor build off time.
« Reply #3 on: November 15, 2014, 05:06:28 PM »
This popped into my head on another thread so I am sharing it here.

An idea for anybody in the pulse motor build-off.  Coils are always energized with DC, typically from a 12-volt battery.  With a microcontroller you could drive the coil with a custom waveform.  Perhaps driving the coil with an initial  voltage over-shoot spike will give you faster repulsion force in place to meet the spinning rotor magnet, giving you higher RPMs.  The voltage spike gets the current flowing in the coil more quickly.  You can imagine a simple program where the microcontroler is triggered and then it just reads out a look-up table in memory to playback the waveform on an analog output pin.  Then you connect the analog output to a beefy voltage servo-amplifier to drive the coil.  You could do one from scratch using an op-amp connected to a big pair of transistors.  The app note for that circuit is floating around.  Or perhaps you could cheat and use a car audio amplifier.

So instead of powering the coil from a straight 12-volt battery, you are powering it from a high-current voltage servo-amplifier connected to a +/-36 volt power supply (as an example).  You can jolt the coil with a custom waveform and see how high you can push the RPM while monitoring the power consumption.  You could literally start to cook your coil if you were not careful.

You have the microcontroller reading ticks from the rotor so that the software could measure the rotor frequency.  It could then multiply the rotor frequency by 'x' to generate the clock for the outputting of the waveform.  That way the length of the customizable pulse would track and be in step with the rotor RPM.  Do you need to do some kind of averaging when you measure the RPM?

The "clock" is implemented using one of the on-board programmable timers in the micrcontroller.  When the timer counts down to zero, it triggers a software interrupt.  The software interrupt outputs the next step in the waveform, does some conditional testing, and then exits.  So the outputting of the actual waveform is done by interrupt-based code in the "background."

So there you have it:  A background interrupt-driven programmable timer function that outputs the waveform in memory (triggered by a pick-up coil on the motor) and in the foreground code you measure the rotor frequency and then multiply it by a variable 'x' to generate the output waveform clock.  If your Arduino board has an LCD display you could display the RPM also.

Once you have that running, then you can experiment with the voltage waveform that you energize the coil with.  You need a flexible trigger where you can change the angle of the trigger like on the MHOP.  The trigger determines the start of the outputting of the waveform.  Say your waveform is 64 steps in length and you have one byte for each step (0-255).   A value of 0 gives you zero volts output from the servo-amplifier.  A value of 255 gives you say 36 volts output from the servo-amplifier.   With that setup you could really push your coil, get it to smoke if you wanted to.  You may want to make the record length for the waveform table to be 128 or 256 bytes in length.  Of course you could change the  length to whatever you want.

In theory you could find a waveform for pulsing the coil that is an ideal match for the motor and gives you your maximum efficiency.  But please define your efficiency also and then measure the variables and show the calculations.

MileHigh

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Pulse motor build off time.
« Reply #4 on: November 15, 2014, 05:27:47 PM »
How about just using _one_ very Longgggggggggg pulse of HV DC to drive the motor? Would that count? (just kidding, sort of.)

http://www.youtube.com/watch?v=tVfw-TeJ9r4  (presented mostly to show the Arduino tachometer, which also can provide the programmed timing pulses that MH is talking about)

Or an antigravity electrostatic linear or rotary PM?

http://www.youtube.com/watch?v=AX-jrlGC-aA
http://www.youtube.com/watch?v=f-aP7sk48jw


Seriously, folks, I am all out of ideas for this year's buildoff. I may be able to come up with something, but I see that LidMotor is the King of PMs and has already anticipated the design I was thinking of with his great little rotary PerPenduPetulum motor:

http://www.youtube.com/watch?v=Kij6dtkhhbE


TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Pulse motor build off time.
« Reply #5 on: November 15, 2014, 05:47:20 PM »
Chet, I appreciate your kind offer, but seriously, I have already built just about every kind of pulse motor that I have been able to think about. I do have one more that I've been working on "in the background" but it's not really anything new or especially elegant. I also have some interesting designs that I can't build, because I still don't have my machine tooling available, it's all up in Canada and probably lost by now.  The level of creativity and skill that I saw from last year's build-off was so impressive that I am humbled and in awe of the real builders out there. I'm just a dilettante with a couple of oscilloscopes and an Arduino or two, not a PM-Pro! But my brain is working on the problem and maybe I'll be able to have some kind of original idea in time to enter the competition. Right now I think my chances are kind of slim.
 :-[

MileHigh

  • Hero Member
  • *****
  • Posts: 7600
Re: Pulse motor build off time.
« Reply #6 on: November 15, 2014, 06:44:05 PM »
Well I can't resist mentioning the disk rotor with the "C" core form for the drive coil to catch both sides of the disk magnet.  And then, under microcontroller control, you start out in attraction mode, and then when the magnet passes TDC you switch to repulsion mode, and you have tweaked the voltage waveform to optimize the attraction and the repulsion phase of the rotor magnet fly-by.  We have the technology.

And you want more thinking outside of the box?  You do similar setup, but instead of a servo voltage amplifier on the output to drive the drive coils, you have a servo current amplifier.  So your waveform table in your Arduino memory is for the current through the coil, and not the voltage across the coil.  Let the current amplifier generate the voltage spikes all by itself.   Now that would be just uber awesome.

ramset

  • Hero Member
  • *****
  • Posts: 8073
Re: Pulse motor build off time.
« Reply #7 on: November 15, 2014, 08:26:49 PM »
Good stuff fellows..! :)



Thx
Chet

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Pulse motor build off time.
« Reply #8 on: November 15, 2014, 11:21:26 PM »
@MH: now that's a good idea, to have a combined attraction-repulsion drive cycle. With say 4 C-coils and 4 rotor magnets you could get to a point where the coils are essentially AC driven, more or less "on" all the time in one polarity or the other, and this would really torque the rotor around powerfully! It's not too hard to do using something like Arduino. There are several "shields" that just snap onto the main board and incorporate full H-bridge driver chips. For example:

http://www.robotshop.com/en/arduino-compatible-mega-motor-shield-1a-5-28v.html

tinman

  • Hero Member
  • *****
  • Posts: 5365
Re: Pulse motor build off time.
« Reply #9 on: November 16, 2014, 02:03:09 AM »
You would want to start the reverse pulse right after the inductive kickback cycle over shoots the 0 volt line. This then means that the polarity of the inductor has already started to change in our favour-in the right direction ;) I dont think you need all that stuff like the Arduino and programing,as i think a push/pull circuit triggered by opto interuptors would do just nicely.

You also know that it dosnt have to be an electric pulse motor?. It can be gas pulses,fluid pulses,or even bouncing mass pulses-->Aaron's overunity bouncing ball maybe :o

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Pulse motor build off time.
« Reply #10 on: November 16, 2014, 02:17:58 AM »
You would want to start the reverse pulse right after the inductive kickback cycle over shoots the 0 volt line. This then means that the polarity of the inductor has already started to change in our favour-in the right direction ;) I dont think you need all that stuff like the Arduino and programing,as i think a push/pull circuit triggered by opto interuptors would do just nicely.
I agree fully, but using the Arduino (or another full MC system like Stamp or etc) can be really cool too, since you can have it connected to your computer over the USB-serial line and change parameters "live" while running. I like Arduino because of the easy-to-use programming environment ( the "Arduino IDE") and it gives me an excuse to use the "c" language that I learned in school many years ago. It can also provide realtime monitoring of various parameters as I showed in the video above. The Arduino-monitored SNOT even computes the actual kinetic energy of the ball as it travels around the track!
Quote
You also know that it dosnt have to be an electric pulse motor?. It can be gas pulses,fluid pulses,or even bouncing mass pulses-->Aaron's overunity bouncing ball maybe :o
Sure, it's nice to be reminded of that! And that's what I mean when I say I can't currently build some of my designs. I have one that is a little Ringbom-style linear Stirling engine driving a linear alternator, a complete heat-to-electricity "pulse motor" power plant that would probably run on the waste heat coming off a mosfet used to drive a bigger conventional PM! But I can't build it without a precision lathe and milling machine.

qtrhack

  • Newbie
  • *
  • Posts: 21
Re: Pulse motor build off time.
« Reply #11 on: November 16, 2014, 05:19:32 AM »
... But I can't build it without a precision lathe and milling machine.

got one of these around you:  http://techshop.ws/

not sure of you location but could be an option ...

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Pulse motor build off time.
« Reply #12 on: November 16, 2014, 01:29:11 PM »
got one of these around you:  http://techshop.ws/

not sure of you location but could be an option ...

Sure, there is one about 120 miles north of me. Unfortunately they want 150 dollars per month for a membership (or 1395 a year, or 175 for one month).  And it would cost me around 35 dollars in gasoline and over 4 hours driving to make the trip on any given day. If I could afford that kind of outlay, time and effort I'd probably be able to retrieve my own tooling from Canada and work in the comfort (hah!) of home, and not need to drive. "Pray for me, I drive I-35" is more than just a bumper sticker.
Thanks for the suggestion though. I'm sure that they will be making money and having fun up there.

I see the location nearest me is also advertising a job opening for a multitalented person who can do all sorts of things and who has experience.  Wages 11 to 14 dollars an hour, d.o.e.! But the skill set they say they want is worth 40 dollars an hour at least.

But they do provide popcorn and coffee!

However they don't seem to be doing a heck of a lot of business. I glanced briefly at the reservation schedule and it looks like neither of the metal lathes are reserved, at all, for the next week at least.

I'm not knocking the place. It seems like a good idea, a good service. For people with the money to spend, that is.

ramset

  • Hero Member
  • *****
  • Posts: 8073
Re: Pulse motor build off time.
« Reply #13 on: November 16, 2014, 03:30:22 PM »
Perhaps a different perspective ?
call "the man" the runs the lathe [the 12 dollars an hour guy]


and ask him how much to Turn your 15 minute piece ?


PM me the place And I will call them .


I know you still use strings and cans..... ???


Besides we might even get you a sponsorship of similar arrangement for the whole group here
and elsewhere .


thx
Chet

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Pulse motor build off time.
« Reply #14 on: November 16, 2014, 05:39:13 PM »
That's not the way that place works. There is nowhere in the world I know about where you can "rent" a machinist's time and the machine for "12 dollars an hour". If you bother to read that website you will see that one must reserve the machine in four-hour blocks, that you must take their safety course before you are allowed near a machine (of course, they'd be nuts not to require that) and you do the work yourself.   And where did you get the idea that someone could make the parts I need to make in 15 minutes? Have you ever set up a lathe to do precision work? Have you ever machined, say, a piston, cylinder, wrist-pin and con-rod set to 0.0001" tolerances.... in 15 minutes?  That, I'd like to see. They also don't say what the size of the "Jet" brand lathe is. The largest parts I need to make are about the size of a man's thumb, and working down from there. Could I make a watch gear, say, on a 7" swing lathe? Could you?

Come on, Chet, get realistic here. Like I said, I appreciate the offer but it's just not feasible. If you really REALLY want to help me, making some arrangement to get my own tooling crated up and sent down to me from Canada would really be helpful. Other people have from time to time offered to help with this but the arrangement has always fallen through somehow. I think I might even have dodged a rip-off at one point.