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: Has anyone seen Lasersabers new motor runs on 1000uf cap  (Read 294500 times)

conradelektro

  • Hero Member
  • *****
  • Posts: 1842
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #195 on: June 18, 2013, 07:16:00 PM »
I experimented with the "Ossi-Motor 1 Reed 2 transistor circuit" from sMartcreations2010:

It was not very interesting, the motor started to run at 12.5 V and 1 mA. The feedback diodes had no effect whatsoever.

sMartcreations2010 http://www.youtube.com/watch?v=auykiV4Kq68 used different transistors which seem to work better. It is strange that the transistors fire at all.

Greetings, Conrad

gyulasun

  • Hero Member
  • *****
  • Posts: 4117
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #196 on: June 18, 2013, 10:55:37 PM »
Hi Conrad,

Perhaps the very low inner impedance of the power supply 'supresses' the back spikes coming from the coils via the two diodes, however small those spikes might be due to the small ON coil current involved. Just place a diode in series with one of the supply rails, perhaps the spikes will appear and increase the DC voltage level a little across the electrolytic cap  (or the diodes' forward voltage losses 'gulp' the small increase).

Why you say it is strange the transistors fire at all, I wonder.

Thanks for your efforts.

Greetings, Gyula

conradelektro

  • Hero Member
  • *****
  • Posts: 1842
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #197 on: June 18, 2013, 11:59:03 PM »

1) Perhaps the very low inner impedance of the power supply 'supresses' the back spikes coming from the coils via the two diodes, however small those spikes might be due to the small ON coil current involved. Just place a diode in series with one of the supply rails, perhaps the spikes will appear and increase the DC voltage level a little across the electrolytic cap  (or the diodes' forward voltage losses 'gulp' the small increase).

2) Why you say it is strange the transistors fire at all, I wonder.


Ad 1) I will put a diode between the power supply and the positive rail.

Ad 2) In the "Ossi motor circuit" shown in my Reply #200 no current is flowing into the bases of the two transistors (base to base connection via 1 K resistor and Reed switch). Why do the transistors switch? (I checked with the Scope, they do switch, but not completely.)

Tomorrow I will try a circuit with two IRLIZ44NPbF  [R-DS(ON) 0.022 Ohm, V-GS(th) 1 to 2 Volt] , it should run with 2 Volt. See the attached diagram. Comments are appreciated, may be I there is an error in the circuit?

Greetings, Conrad

gyulasun

  • Hero Member
  • *****
  • Posts: 4117
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #198 on: June 19, 2013, 12:47:55 AM »
Hi Conrad,

The MOSFET circuit version looks good to me, it can work, albeit normally a P-channel MOSFET is used at the upper position (where the pnp is connected in your bipolar transistor version).
Perhaps a 10 kOhm resistor would insure a little quicker switch-OFF time instead of the 100 kOhm, it can discharge gate-source capacitors more rapidly (RC time constant is less).

Regarding the pnp-npn transistor circuit why the transistors switch: you can consider the base-emitter junction of the pnp transistor as a diode with its anode at the emitter where the positive supply voltage is connected and its cathode is at the base (the arrow symbol may be seen as a diode symbol pointing towards the base). And this is just the opposite for the npn transistor, its base is the anode of the junction diode and its emitter is the cathode, this latter going to the supply negative rail.

Now the supply voltage appearing across the rails, it "finds" two diodes in series: the positive supply voltage can start to drive current into the pnp transistor's emitter-base diode, biasing it in the forward direction and this current can continue to the base-emitter of the npn transistor via the resistor and the switch, the base current is defined by the 1 kOhm resistor only: base current=[12V-(2*0.65V)]/1 kOhm=10.7 mA whenever the reed switch is ON. (the 0.65V is the base-emitter bias voltages) I edited your schematic to show the base-emitter 'diodes' by the red arrows and the path of the base currents (via the resistor and the switch of course).

Gyula

mariuscivic

  • Full Member
  • ***
  • Posts: 203
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #199 on: June 19, 2013, 09:11:43 AM »
Made some measurments to my toy; it's oscilating from 0.9 to 1.1 mA at full rpm. Also made another rotor more balanced and with no holes to reduce the air drag

conradelektro

  • Hero Member
  • *****
  • Posts: 1842
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #200 on: June 19, 2013, 03:42:25 PM »
Made some measurments to my toy; it's oscilating from 0.9 to 1.1 mA at full rpm. Also made another rotor more balanced and with no holes to reduce the air drag

@mariuscivic: now, with your nice contraption which uses very little power, you can try to recapture electric energy with pancake coils or any other strange coils (wich as some people claim cause no LENZ losses).

@gyulasun: thank you for the explanation of current flow through a base to base connection from PNP to NPN transistor. Still, it is astonishing that I needed 12 Volt to make it work? I tried 300 Ohm and even no resistor at all. Power draw increased but still, it would only work with 12 V or more.

I did some more tests with "disconnecting a coil at both ends":


The circuit with the two IRLIZ44 MOSFETs worked very well. But I could not see any charging of the cap by the diodes.

I set the power supply to 6 Volt, waited till the rotor settled at a constant speed and then switched off the power supply. With my Russian mechanical stop watch I timed how long it took the 1000 µF cap to discharge from 6V to 2V (while the motor slowed down). It always took about 23 seconds, with diodes or without diodes. The same test while the cap discharged from 3V to 1.5V in about 32 seconds, and also no difference.

The circuit works of course even better with only one IRLIZ44. The rotor turns faster with less power and a feedback diode (or LED) increases power draw by about 10% and slows the rotor down.

So much for recharging a cap with the back EMF of a coil! Now, do you believe it will work any better with two Reed switches (in place of the two transistors)? Each coil in my little motor has 90 Ohm DC resistance, in sum the six coils have about 540 Ohm DC resistance.

Greetings, Conrad

P.S.: I bought the Russian stop watch at a fair in Vienna for 20.-- EUR where they sold old mechanical photo cameras (a few years ago). It must be 50 years old or even pre second world war, and is very heavy. It allows to take two times (one digit stops, the other continues). All mechanical, no electronics.

gyulasun

  • Hero Member
  • *****
  • Posts: 4117
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #201 on: June 19, 2013, 08:30:05 PM »


Hi Conrad,

I think the setup would work better with reed switches replacing the two transistors. At least you could get rid of the saturation voltage drops between the collector-emitter of the bipolar transistors and reeds may still have less DC resistance than your MOSFETs at lower gate-source voltages (when your supply voltage is 1-2V only). However, I do not expect a significant improvement.

I have not found an explanation why your bipolar transistor version needs a minimum of 12V supply or higher to operate, sounds strange for sure. Perhaps you may have some other pnp and npn transistor types to test this behaviour? 
It all sounds like one transistor stage (either the pnp or the npn) would limit the maximum current takeable by the other stage and vice versa, here I assume if you omit any one stage and operate the setup with a single transistor, then the single bipolar transistor stage would surely operate from 1-2 Volt supply again as is normal.  But why this limiting behaviour has a voltage window of 12V I do not know yet.
Another notice: the lack of recharging the capacitor even a little may show that too much part of the recovered energy from the collapse is wasted in the coils' resistance (540 Ohm).  Because the source of the spikes are the coils so the generator inner impedance is a drawback from energy recovery point of view.

Greetings, Gyula

conradelektro

  • Hero Member
  • *****
  • Posts: 1842
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #202 on: June 19, 2013, 09:09:55 PM »
@gyulasun: thank you for the reply.

I will try two Reed switches instead of the two transistors. But first I will try to charge a separate cap. My guess is that it will charge to less than the supply Voltage, which would explain why a supply cap never gets charged. Supply cap charging can only be achieved if power is supplied to the rotor (e.g. by blowing air over it) so that it turns faster as it would with the momentary cap Voltage.

One could overcome the lesser Voltage of the BEMF spike by introducing an up-transformer (as Wattsup is suggesting http://www.overunity.com/13523/has-anyone-seen-lasersabers-new-motor-runs-on-1000uf-cap/msg363047/#msg363047), which I will try as well. My suspicion: whatever one does it has to be paid for with supply power.

My other suspicion: the Ossi motor would have performed equally well without the return diodes. Did anyone try that? Lasersaber's 3D-printed motors are in principle Ossi-motors (but with a different and simpler drive circuit). The original Ossi motor used 4 very high impedance coils (for Audio speaker adaptation) which had very high DC resistance. Lasersaber's motors would run on a 1.5 Volt AAA rechargeable battery for many days as did the Ossi motors.

P.S.: I have little luck with the DadHav circuit. It did not work with MOSFETs and not with BC547 / BC549. I could not make it work with any of my pulse motors. By chance I have the same transistors as Lidmotor uses (2N3906 and 2N4401) http://www.youtube.com/watch?v=auNqKNZLbc0 (see at minute 5:26) and I will try them. Like a base to base connection it only seems to work well with specific transistors. Eventually I will order exactly the same transistors as DadHav uses (MPS-A06 / MPS-A56, I do not have the MPS-A56). I like that DadHav's circuit does not need any "trigger" like trigger coil, Reed switch or Hall sensor. The drive coil (or series of coils) is at the same time driver and trigger.

Greetings, Conrad

ALVARO_CS

  • Full Member
  • ***
  • Posts: 178
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #203 on: June 19, 2013, 09:33:27 PM »
hi
I read long time ago a PDF supposedly from Ossie himself in which he recommended as the best position, to put the reed switch INSIDE the coil. (that is at center).
cheers


conradelektro

  • Hero Member
  • *****
  • Posts: 1842
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #204 on: June 19, 2013, 09:46:44 PM »
hi
I read long time ago a PDF supposedly from Ossie himself in which he recommended as the best position, to put the reed switch INSIDE the coil. (that is at center).
cheers

Have to try that. It is difficult with the coils of my present little six coil motor, but in my next design (which is under way) I will leave the hole through the coils open on both sides (through which I can insert a Reed switch). Good idea.

My guess: with the Reed switch inside the coil it's ON-time is very short, which reduces power draw. It is also a good position to trigger the pulse, just as the magnet is dead centre over the coil. The momentum of the rotor assures that it is pushed in the right direction. And the motor would run in both directions equally well (after an initial push).

With my little 6 coil motor I see that the pulse happens as the magnets are almost past the coils. This is good for self starting in the right direction but not optimal.

The power of collaboration, many good ideas are pooled.

Greetings, Conrad

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #205 on: June 19, 2013, 10:48:23 PM »
You should be able to charge an external capacitor all the way up to the peak voltage of the spike, much higher than the supply battery voltage. As you have already figured out, the run cap can't charge up because it is always being shorted out during each cycle. But you can transfer energy from your run battery into an external capacitor very effectively, as you will find out. Then the "trick" will be to switch the external cap back into the circuit at the right time to put its collected energy back into the rotor's rotation. You could do this with a simple comparator circuit and a transistor, probably.
You will need a fast diode, the faster the better, and of course its voltage rating should be high enough to withstand the maximum spike voltage. Something like MUR1560, but the UF4005 - 4007 type rectifiers might be good enough. I've also found suitable fast diodes in old TV circuitboards. RG10A is a fairly common superfast rectifier diode in TVs. RU2 or RU2B are also in there.

gyulasun

  • Hero Member
  • *****
  • Posts: 4117
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #206 on: June 19, 2013, 11:12:06 PM »


Hi Conrad,

Sorry but the air core coils Ossie used have 1.05 Ohm DC resistance, 1.8 mH each. See this link: http://www.overunity.com/8731/the-ossie-motor/msg226572/#msg226572  and Naudin used also 1.8 mH inductances but with 0.55 Ohm DC resistance each: http://jnaudin.free.fr/ossiemotor/indexen.htm   This is one reason why they used series resistances: to limit input current and save reeds from premature burning. But the other reason for the series resistors is that a balance, a matching can be reached between the motor impedance and the battery impedance, this insures the best operation and long running time. On this AC/DC matching, here is what Ossie wrote:
http://www.overunity.com/8731/the-ossie-motor/msg226685/#msg226685

I agree with trying to collect the spikes in a separate capacitor and see the voltage level received.  The up-transformer sounds also good to me and I also think the recovered energy strongly depends on the input energy, minus the losses. (unfortunately, the advantage of high impedance coils for motor mode turns to a disadvantage when capturing the spikes: the high impedance coils will operate as high inner impedance generators when supplying the spikes  so also matching would be needed towards the 'load' that uses up the captured energy (or use the captured energy at a time the motor cannot see that load).
Regarding the DadHav pnp-npn circuit, you sound to be aware of the fact that the rotor magnet should be given a good hand-pulse to start rotation, i.e. induce an intial AC voltage in the coil because otherwise the circuit cannot work. IT may well be transistor type dependent though. The circuit can be considered as an oscillator after the start, syncronized by the rotor magnet inductions.

rgds, Gyula

Magluvin

  • Hero Member
  • *****
  • Posts: 5884
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #207 on: June 20, 2013, 04:25:38 AM »
Ordered 50,000 ft of 42awg wire. Well, 49,000 and change.  ;D Will be here next week. $35 shipped ebay.  ;D

Not a bad price and will have enough to do many things at many turns and many feet. :o ;)

I said this once a long time ago back when Zeropoint132 made his no bearing bedini with 4000 turns bifi, I said, I wonder if it is the fact that the coil is wound with many many turns of fine wire. But I never messed with very fine wire because I didnt think I could get much out really, and I wasnt sure if Z's motor actually worked. But now I have enough of the right wires that I can give it a go. I tried it with much lesser windings and slightly larger wire than Z used, but didnt work. 

One thing I am thinking now is that maybe Z's motor wasnt self running really, maybe it was just like Lasersabers motor. Dunno.

Lasers motor seems to be 'running' on a fine line of balance between the coils driving the rotor from the cap and effective charging if the rotor is accelerating, increasing in rpm from a dead stop or from a standing rpm. Just a slight increase in rpm from an outside influence and the cap voltage goes up seemingly to match the rpm.

Below is what Im going to use as bobbins with 5/16x1/8 neos on the rotor. The washers and black plastic spacers are from ace hardware. I use a superglue that is called Tite Chairs. It doesnt say superglue, but it is very thin liquid and comes with fine precision tips that work well when ya get the hang of it. The plastic washers and spacers are roughed up a bit with 400 grit in the areas where the washers are fit to the spacer. The spacer has a flange on one end so just a lil pressure and a tiny bit of Tite Chairs and then the other end the washer is set flush at the end. It hods up very well for nylon washers and what ever the spacers are made of.

Looking to do 24 coils and try 24 alternating mags and using a bias mag for the reed so it only fires on N poles not S, or vice verse.

Im looking into rectangular coils also as I explain why in this thread below with pics. Just trying some things so far.
http://www.overunity.com/6763/energy-amplification/msg363481/#msg363481

Mags


conradelektro

  • Hero Member
  • *****
  • Posts: 1842
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #208 on: June 20, 2013, 02:11:16 PM »
@Gyulasun and @Tinselkoala: thank you for your comments.

Charging of a 100 µF capacitor with the Back-EMF of the 6 coils in series:

(If you want to see the little six coil motor look at http://www.overunity.com/13523/has-anyone-seen-lasersabers-new-motor-runs-on-1000uf-cap/msg363434/#msg363434 , the six coils have a total DC resistance of 540 Ohm)

I am really astonished, the Reed switch allows a much better charging of an external capacitor than the MOSFET.

With the MOSFET (as a switch) the cap only charged to a little more than the supply Voltage.

With the Reed switch the cap charged to a multiple of the supply Voltage.

See the attached schematics, I am moving on to a second Reed switch in order to disconnect the 6 coils completely.

Greetings, Conrad

gyulasun

  • Hero Member
  • *****
  • Posts: 4117
Re: Has anyone seen Lasersabers new motor runs on 1000uf cap
« Reply #209 on: June 20, 2013, 02:24:40 PM »
Hi Conrad,

Good tests, thanks and a possible explanation for the reed performing much better that it has no any shunt capacitor towards the negative battery rail like the MOSFET drain source capacitance represents (some nF in the low 1-2V  D-S voltage range).
What current consumption is involved from the supply when the reed is used?

rgds,  Gyula

PS Will be away from the internet for a week from this afternoon. Keep up and enjoy experimenting.