can a novel way to double the efficiency of a computer fan be applied to electric car motors

Hi all. in recent years, i have been modifying small brushless motors, computer fans,floppy drive spindle motors and a rewound R.C plane motor with neo magnets. I started with using a mini isolation transformer to sense off the same coils being driven, that switches a transistor on and off, to run the motor. Then i added a run capacitor in series with the driven coils and a second mini iso trnsfrmr to switch a second transistor. so T1 switches the driven coils on and the run cap charges up. Then T2 is switched on and discharges the run cap back through the driven coils, but with the current reversed.As the rotor has advanced 90 degrees,this is complimentary to running the motor.Sort of like using little packets of power twice.See attachments.
Here are some multimeter readings, taken with the computer fan in the pics. The power supply is an unregulated multi-voltage plug pack. 100 hz=50 revs/sec=3000 rpm
4 coils driven pulsed dc 4 coils driven psuedo ac
8.8 v 84 ma 72 hz 9.6 v 34 ma 76 hz
10.2 v 97 ma 80 hz 11.2 v 42 ma 89 hz
13.5 v 122 ma 98 hz 14.8 v 59 ma 116 hz
As it stands it's not self starting. Additional electronics need to be developed.Something like using the standard hall effect sensor/circuit in the computer fan to start, then switching to this "conservation mode". Also pulse width modulation is used to efficiently modulate the motor output, in electric cars. Running in C mode, i guess run caps with different values or individual drive coils would need to be switched in and out,possibly with the aid of zero crossing circuitry, for variable throttle control.
Recently i got a bigger computer fan to run nice using it's own hall sensor, as per the last circuit diagram. it runs even better.

So the big question is.... if i use 70 -200 uf low esr run cap capacity for these little motors that consume less than 1 watt in C mode, is it viable to run say a 25 kw motor like this.Do the necessary caps exist, or could batteries be used, or some sort of hybrid storage device. If it can be done,then this approach may double the range of electric vehicles and battery tools and garden equipment,etc.,with further development.
I would appreciate any comments or advice.Any feedback is welcome.
cheers.
lasermouse.

 

I would say the bigger question is, what makes you think you have improved the efficiency?

 

I don't see any calculation where you used the appropriate units which are watts. In order to show an improvement you need to show that you can run the fan at the same speed using less power, measured in watts. It is not enough to change the current consumed because watts and amperes are two different things.

 

Also, an unloaded fan produces about as much useful work as a car engine running in neutral. The real question is how much air the fan is moving into a defined back pressure versus its power consumption. Since by definition all energy conversion costs energy, the odds of an increase in net system efficiency is low.

ak

 

If you stand the fan on it's side so it can move air what is its stock voltage and amperage input to get a specific RPM out of the blades Vs the same physical conditions after your modification while turning the same RPM's?

And FYI those tiny motors used in those fans have a lot to be desired of for operating efficiency to begin with. A circuit that works for taking a 40% efficient motor to 70% is a good gain but on an already 80% efficent 25 KW elctric motor it might not do anything.

 

I know it's flea power, but the hz figure converts to rpm and the fan motor is loaded by the fan blades. the other motors in the pics are unloaded.if you look closely at the yellow and blue lit areas on the floppy drive motors they drop out in the centres. i assume this is the generator bemf peaking at mid-stroke and it greatly reduces the current momentarily.
this pic is a rudimentary computer fan circuit. there are 2 strands of 2 drive coils. these have to be separated and the strands are wired in parallel, with attention to making sure they are magnetically synced. ie. 1 strand's connections are swapped from standard. running as stock, 1 strand, 2 coils at any time is energised, so i would assume energising 4 coils,( both strands parallel wired), about 50% of the time should give about the same rpm as standard.
try it yourself, as per the diagrams.the iso traffo circuit uses 2 npn's. the hall sensor version uses 2 pnp's and 1 npn.T3. I like mje 340/350's.they are hard to kill. some fans are harder to pull apart than others. if the factory rpm for rated voltage is known, that's a good baseline to compare the modified rpm for same voltage, and the current will be half or less, and the fan will spin faster than stock.
I'm just a hobbyist tryin to make the world a better place, and i understand the importance of careful measurements and comparing apples to apples. But i'm starting to think that the notion that you can use the same bit of power more than once will turn out to be the best kept secret of the 20th century.
cheers.

 

Sorry but using the same power over again belongs to the over-unity and perpetual-motion crowd of ideas.
Once power is used it is gone.

 

Welcome to AAC!
I understand the principle of using the transformers and transistors for coil commutation, but where do the power supply and the motor connect in your schematics?
How do you arrive at the figure of "double the efficiency"?

You aren't using the same bit of power twice. The capacitors and transformer inductances need supplying with energy (first bit of power) to be charged before they can be discharged (second bit of charge). You may be using the power in an unusual way (transferring some energy between capacitors and inductors), but the total power won't be reduced.

 

ok. here are the numbers again.
4 coils driven pulsed dc near 50% duty cycle 4 coils driven pseudo ac near 100% duty cycle
8.8 v 84 ma .7392 W 72 hz 2160 rpm 9.6 v 34 ma .3264 W 76 hz 2280 rpm
10.2 v 97 ma .9894 W 80 hz 2400 rpm 11.2 v 42 ma .4704 W 89 hz 2670 rpm
13.5 v 122 ma 1.647 W 98 hz 2940 rpm 14.8 v 59 ma .8732 W 116 hz 3480 rpm

my apologies for not doin it like this in the first post, but these are the numbers. If you simply cannot accept this, then all i can suggest is try it yourself. It's a bit fiddly. check for a circlip concealed under a sticker on the back, but some might click together. and watch for o-rings, if any, then remember to carefully re-install.Then you have to yank the stator out of the frame. then unsolder the circuit board from the 3 mounting pins. next, the coil wires are fairly thin. carefully with a small pointy thing like a meter probe or a pin ease the 4 wire ends off the 3 mounting pins. solder a short length of thicker enamelled copper wire to the 4 coil wire ends and insulate each join with a little bit of heatshrink. Drill a small hole in the back of the frame under the stator to lead the 4 copper wire extensions out the back. attach 1 thin wire to both of the hall sensor outputs, under each end post. loosely fit the stator back on the frame centre mount. slide the rotor on temporarily. continuity test the 4 coil wire extensions to establish 2 pairs. connect a small voltage, say 1.5-6 v and find the parallel connection of the two 2 coil strands that holds the rotor in position when energised. wrong way round, it's easy to turn the rotor when energised.
to get started use the circuit diagram in this post. put a small piece of tape over the drilled hole to secure the 4 copper wires so the windings can't rub or get caught up by the rotor. if you have the right hall sensor output, the motor should start itself and run on 6 v or more, if the fan is rated at up to 12 v. It may run backwards on the other sensor output, i haven't tried it yet. once that's going add the run cap, 150-200 uf low esr, and T2 andT3. base resistors are 4.7 k. T1 and T2 are pnp as the hall sensor outputs are negative.T3 is npn. give the rotor a good flick to start it and hopefully your in business. do some measurements and let me know how it goes.
cheers.

 

the bigger fan i got running using it's hall sensor says 12 v 160 ma on the back. 1.92 w. Modified,at 14.75 v, it draws 60 ma, .765 W. Today i will measure the hz/rpm in the modified state, then reconfigure it back to standard and report the numbers. at 160 ma, the volts will sag to somewhere around 12 v, on my unregulated power supply.
cheers.

 

correction on that last post. 14.75 v at 60 ma makes .885 watts. my apologies. anyway i will post the numbers for comparison, and hopefully a pic, if i can borrow a camera. It's probably best to use 2 wire fans to try this on.
cheers.