Two Phases from One Pulse Motor ?

[Farmhand]

Hi guy's, I thought I would bring this topic and some interesting shots here for consideration. I think I've found a new way to make pulse motors more efficient.

The charging coil in the diagram is placed at around 45 degrees to the main coil, there is a difference in phase between the main motor coil and the charging coil.

Here's the circuit


This is the currents in the two coils the blue one is inverted and is the motor coil.


This is a comparison I made to see the difference between clamping the coil to the capacitor C2 with a diode and the circuit arrangement above. The top shots are clamped the bottom are as the drawing shows.
The yellow trace is the Motor coil current sensed at resistor R2 the scope grounds are together and the blue trace is from the mosfet drain. Please note the volts per division in the lower shots for the blue trace is changed from 10 to 20 volts.


Here's a video. At about 7:20 to 10:20 in this video I show the difference in the rotor speed and the voltage in the C2 capacitor when the coil is removed and put back. In these video's the circuit was arranged slightly differently than it is now as above. It's a bit better now.
http://www.youtube.com/watch?v=w1_KlgJ09Bs

Efficiency test.
http://www.youtube.com/watch?v=5orxI__rFsU

And with the 12 volt input boosted.  
http://www.youtube.com/watch?v=Yukaj0QQzN0

It's on the bench so I can test stuff.

Cheers

[Farmhand]

Hi again, Here's the voltage and current wave forms. For the scope shots the yellow is the motor coil and Blue is the charging coil. And for the voltage measurements the scope grounds were at the circuit ground the probes at the positive ends of the coils. The thing is there is only 12 volts applied to the charging coil, because of the diode and the the way it works after the charging coil discharges into the cap the positive end of it becomes at the voltage of the charging cap it discharged into. you can see the 12 volts battery voltage applied to it drop slightly under load then rise in volts to the same as the capacitor C2. Because the motor coil has over 20 volts applied to it the current moves quicker and starts just before the charging coil current, the charging coil current ends as is shown in the scope shot 6 mS after the motor coil current ends and looks like the motor coil current starts 12 mS before the charge coil current, (my charge coil has less inductance it should be the same as the motor coil and will be soon). The delay in the peak current seems to be about 12 mS.

We can see the voltage applied is practically in sync to begin with, but the inductor voltage curves down as the motor coil discharges the capacitor C2.

The frequency of the motor was higher when the current wave forms were taken at 74.63 Hz. So the phase delay can be calculated from that being there is two magnets on the rotor.

There is a definite delay in the phase of the currents and it can be manipulated by known means and utilized easily enough.   Even if there was no delay the charging coil could still aid the rotation, it's just a matter of placement. I'll do some calculations and see if I can find the phase angle delay and such things.

To see the divisions better just download the pic and view blown up a bit maybe if anyone wants to be accurate.

The delay in the phase of the currents is related to the pulse width in a way but also due to voltage over inductance and resistance effects as well as the return current from the discharge of the motor coil, in my opinion.

Cheers

P.S. Please tell me if I'm reading the scope incorrectly.

..

[ion]

Not sure if you mentioned the type of motor you were comparing your pulse motor to.

Your comparison should be to a DC brushmotor or electrically commutated DC motor for a fair comparison of power in vs power out.

If you are comparing to a shaded pole AC induction motor, they are notably inefficient, the pole shading being a shorted copper band on one of the poles which dissipates a lot of input power as heat.

Also, pulse motors could be made notably more efficient by considering a properly designed magnetic structure so that both poles of the driving electromagnet  and driven magnets are used.

When efficiency is thus optimized you will wind up with what looks like a standard electrically commutated DC motor.

By comparison, the b=e=d=i=n=i stuff is extremely crude and magnetically inefficient both in power rating and material use.

[TinMan]

Should R1 not be between B1 and C1 ?,this would give you total circuit consumption.Then output would be taken across R3-this is of course if you are wanting p/in to p/out.

[Farmhand]

ION the comparison was just a bit of a lark, a poke at the Keppe mob.  

Tinman I'm not measuring the input or efficiency with that, I'm using them to sense current. I am aware of how to measure the power
and it requires smoothing of the DC. But there is no electrical output now, this is to generate using a generator coil.   Give me time.
It will be an acceleration under load motor, when I get that happening I'll measure the power.

Any real comparison should be with another pulse motor because the point to it is the charging coil is used to help turn the rotor.
I had the charging coil there to begin with just to raise the voltage of the capacitor C2 to dump through the motor coil. But I decided
I might as well use it to help turn the rotor as well. as just an inductor it had not much loss because it has only 1.2 Ohms resistance.
And when I use it to help the rotor the raised voltage it produced only dropped from 26 volts to 23 volts in C2.

I''ve now found a suitable way to return all of the inductive energy for the charging coil to reuse so there is only one battery.

I'm not claiming anything except a new and novel pulse motor principal. It's a resonant charging circuit for a pulse motor and the
charging inductor is used to help drive the rotor as a second phase, and it now returns all the inductive energy from the motor coil to a capacitor
and back through the charging coil. It's not built as best as the idea could be done it's a prototype. The last couple of changes I made have made it much better.
It spins the rotor more efficiently now and with more torque than it did in the video's, so it's improved. A better design would be used for a motor for a
practical purpose one with more coils and more magnets for better efficiency and torque.

I'm not after the "spike" with this, except to enhance current through the coils to drive the rotor.

This is the single battery setup a few people want, so as to do away with the charge battery for experiments.  

As far as I am aware no one else is using these principals in just this way for pulse motors.

I can't believe no one gets it.     Kidding.
I guess I should make a simpler drawing for the inductive energy return on a regular pulse motor circuit though.



Here is a circuit with just the inductive energy return to a capacitor for the motor coil to reuse, without it having to go to the battery.



I can provide scope shots of the energy being returned and reused, if anyone is interested.

Cheers

[Farmhand]

Here's some wave forms. I think I can see the capacitor C2 works to be charged to about 24 volts, the supply is a 12 volt battery.
The spike is caught by the cap C3 then the coil L1 discharges the cap C3 into C2. This keeps the motor currents pretty good, and
reuses the energy from the discharge of the motor coil almost immediately through the charging coil in my setup, or in
the motor coil during the next cycle if no resonant charging circuit is used.

The motor runs much better now even when charging a battery it's ok better when not, the current in the motor coil drops of quicker now.



Cheers

[Farmhand]

I'm about to start generator tests, but here is a drawing showing where to connect a battery for the optional conditioning charge output.
I wonder what would happen if I were to run it as in the drawing but connect the free negative end of the battery to a ground stake outside. 



The generator coil/core setup I am testing should give a sharp rise in voltage (almost flyback type) at double the frequency of the magnet passes,
the coil has not many turns so the voltage will be low but the wire is thickish 0.72 mm.

Anyway cheers

[TinMan]

Hi FarmHand
I for one am looking forward to your result's.
The circuit is most interesting,and for a pulse motor it seem to run that large fan preaty good.
How are you triggering the fet?

[Farmhand]

Hi Tinman, It's optical triggered I use an RPR220 photo reflector which has a sensitivity circuit with logic output by way of a CD4011 chip, the output from the optical sensor module goes to a signal processing board for hardware signal processing into whatever sized pulse width I want with a CD4047 chip, the CD4047 outputs it's pulse to a mosfet driver which drives the mosfet gate. Drawings attached. The optical sensor is mounted on an arm that swings in an arc to adjust the timing. The signal processing board drawing is not finished, some parts are not labeled, I built the circuit first then made the drawing 6 months later so I need to check some part values ect.

The thing with the two phases from one motor is that the phase difference is dependant on the Duty cycle of the "on" time, I need to find a way to keep the mosfet "on" times to 25% duty at all speeds for this particular motor, to keep the phases correct, or just run it at a certain speed at all times using the picaxe chip to control the boosted input voltage to maintain speed under load. There are several ways that can be achieved. One way is to have the setup tuned to have 25% duty at say 2500 rpm with 12 volts input then when a load is added and the speed slows the picaxe senses that drop in speed and applies an ever increasing voltage to maintain speed, another way is to have the motor set to have maximum power at say 2500 rpm and use blanking pulses to limit the speed by the making motor miss pulses then when a load is added the blanking pulses stop and the motor gets full power, there are other ways also, but it requires that I write some more programming code, which is a pain in the neck, I'm learning as I go. At the moment I just use a voltage divider to sense the input voltage after the boost converter and control the input voltage/power with a pot manually over 10 levels of power 9 boost levels and the battery voltage. Yet another improvement would be an automatic timing adjustment based on applied voltage.

The motor circuit has changed a bit, I'll post any major changes. This can be done with any pulse motor if a resonant charging circuit is used, simply scope the currents through the main motor coil and the charging coil to see the phase difference at the desired rpm and pulse width, then place the charging coil to suit.

The end result is the the Two Phases From One Pulse Motor has more torque than a regular single drive coil pulse motor and still with only one switching phase and the charging coil discharges directly to the cap C2 for the motor to use then the motor coil discharges "via the inductive return" back to the charging coil so there is full energy recycling. So far I have not seen a pulse motor circuit that can reuse the inductive energy release from it's motor coil during every next cycle and every cycle. If you know of a similar circuit that returns the motor coils inductive energy release to a higher voltage then back to be reused please let me know.

I'll keep making small improvements.  

I attached the boost converter schematic as well, I can also post the code to run that as well, the coil determines the PWM frequency. The drawing has two coils, I made it for my other boost converter which has two coils for more power. I also designed and built a solar battery conditioner with a similar circuit to the boost converter, it goes into a desulfation mode when the light gets low and pulses the battery from 24 volts or so from a 1000uF capacitor, when the sun is full it keeps the panel voltage at 17 volts MPP while pulsing the battery with 5000 uF, it also goes into float mode when the battery reaches the set voltage level. It's only designed for 20 Watts of panels for conditioning my batteries.   The solar setup can tolerate a reverse charge battery connection and really pings the battery. The code for it is also available but the code is dependent on the components used, for different components different PWM frequency is required ect. It could use some improvements but it is working still after several months through summer.

Cheers

[TinMan]

Quote: If you know of a similar circuit that returns the motor coils inductive energy release to a higher voltage then back to be reused please let me know.

Well i dont know of a circuit quite like this,but i and many others on my forum have been designing single battery ssg's for some time now.
Here is the link to one of the threads- http://iaec.forumco.com/topic.asp?TOPIC_ID=37
I tried not long ago here to start up a pulse motor thread,and see how far we could take the pulse motor-but no interest was shown
I do have one circuit that was very sucsesful,but that never had a chance to be replicated here.It seem's that most here arnt interested in the humble little pulse motor lol-but there the thing i like most.
http://www.youtube.com/watch?v=5HwUuPQn2W8

[Farmhand]

Hi Tinman, If you mean just snubbing the inductive energy release back to the supply through a diode like in the picture attached then that degrades the performance of the motor. And the reason is shown in the motor coil current wave form comparison between coil collapse energy snubbed and coil collapse energy captured to a higher voltage then fed back. By capturing the inductive energy release into a higher voltage the current in the motor coil is prolonged, so a shorter pulse width can be used for the same speed, that was obvious to me in those tests I did. When snubbed the motor has less power and speed but also consumes less power, however that is with the same pulse width. The benefit is there.   I mean it's easy enough to just connect the charge output back to the battery but it slows down the motor. The same thing happened with my SSG.

I've also found that by just connecting the capacitor C3 to the circuit ground a similar effect is had but the motor torque and speed is even better for the same pulse width. Basically I get even better performance for the same pulse width than with a charge battery and with a charge battery is better for the same pulse width than with the diode just connected to the supply, there's no contest. My pulse width can be made constant so I can see the differences immediately and investigate why.  

Also if the capacitor C3 is made a smaller capacity then the motor runs even faster.

I've just also made an improvement to my rotor and now it can do 4450 rpm with a 2 mS pulse width. However it can only do 4000 rpm or so with 2.6 mS pulse width, that is because the duty is dependent on the speed in rpm and the pulse width, and I need to stay below 30 % duty to keep the phases correct, which is why the shorter pulse width for a given speed and torque the better.   It can now run at 2200 rpm for 200 mA from 12.6 volts or 2.52 Watts and is really quiet too.

Acceleration is much improved as well. I'll make another short clip tomorrow I think after a bit of a clean up. I'm waiting on some wire to rust up for new cores, then i'll setup a generator coil to play with.

The rotor improvement also means I don't need to adjust the timing anymore unless I want to tune for torque.

Anyway it's the current wave forms from the motor coil that show the difference in my opinion.

Cheers