Electric Motor Secrets

[Jetijs]

Thanks for the tips.
We have a big store that sells all kind of bolts and screws, they have also non magnetic steel bolts. I will visit them tomorrow but to be sure I will take a neo magnet with me The magnetized bolts in the startor makes the rotor always to end the rotation in alignment to the startor, the attraction force is very weak, but it is still there

[lighty]

@Jetijs

As for the bolts- just ask for A4 steel bolts, washers and nuts and you will get completely non-magnetic stainless steel. Try to avoid Italian produced steel since sometimes it is magnetic regardless of the standard they should adhere to- I prefer German and British stainless steel products because I never found any batch that was magnetic (I guess they have much stricter production methods).

If you cannot find A4 steel in ordinary stores find the nearest marine equipment store (yachting and sailing equipment and parts) and simply try for A4 parts- they usually have highest quality non-magnetic stainless steel products. Also, you could take a strong NdFeB magnet with you and try out the material for yourself before you buy it.

As for the timing, parallel driving of transistors etc.- Peter gave you a good instruction but when you get to the next generation of electronics let me know. As I said reed switches are nothing but unnecessary hassle if you want versatility. Regarding that check your PM.

[Jetijs]

Thanks Peter,
I understand now

Today I changed the bearings to new ones. I assembled the whole thing and turned the rotor with my hand and this time I could not feel or hear anything unusual in the bearings, the rotor moved smoothly and quietly. But when assembling the motor, I noticed that the bolts, that are holding together the startor plates became magnetized. I should probably change them to non magnetic steel bolts. Also the timing adjustment was now harder, because when assembling the motor, I had to remove the timing wheel. When I attached it once again, I could not get the exact timings for all 4 pulses, the timings were good for only three pulses and not for the fourth. That is because my timing wheel is a bit off center. Should I build a new one with weaker magnets so that I can move the reedswitch closed to the wheel? I think that would increase the precision of the timings. Also I could not get anywhere near the results I posted yesterday. With the bulb on the output I could get only about 600 rpm and the current draw was 0.5A instead of 0.4A as it was yesterday. The vibrating sound is still there and it appears in every pulse. It comes from the rotor side, as far as I can hear, and not form the bearings. I checked the gaps between rotor and startor and they are eaqual.
I tried to move the reedswitch around the "sweetspot" when the motor was running, this way I could increase the RPM's but the current draw also went way up to about 1 Amp. Also we lost power for a half an hour just moments ago because of the storm. I had to remove the light bulb form the motor so I have at least some light. Of course I forgot to attach it back to the motor circuit when the power appeared again. This resulted in blown transistor.
It is definatelly not my day

Edit: I figured that in order to get the perfect timing for all the rotor legs, I should use four reedswitches in parallel and only one magnet on the comutator. This way I can adjust the perfect timing for every leg precisely, also the reedswitches would operate at lower frequency.

[Peter Lindemann]

Balancing the Branches

I uploaded a short video on youtube, you can watch it here:
YouTube - Lindemann attraction motor

Peter, I don't understand this balancing of the branches. Can you please explain what you mean?
Thank you

Jetijs,

OK. You need each transistor to operate in it's own little voltage divider (between the two resistors). Also, we know that if we parallel resistors, the effective resistance goes down. And finally, the little reed switch has current limitations that should not be exceeded, or it won't run very long.

So, step one is to take your motor that is running on one strand of wire and start increasing the value of the resistors until the motor starts drawing LESS current. Right now you are using 200 ohm resistors (approximately). See if the motor will run the same way if you put in two 680 ohm resistors in place of the ones you are using now. If it runs the same, try even higher values.

You may also have noticed that running your motor at 30 volts, your little 200 ohm resistors are getting quite hot. (Maybe you didn't notice.) The current in the control circuit is controlled by Ohm's Law (E=IR) So, E/R=I. 30 volts divided by 400 ohms (total of both resistors) equals .075 Amps or 75ma. Since each resistor has half of the voltage drop on it (15 volts), then each resistor is dissipating 15 volts X 75ma = 1.125 WATTS. If you are using 1/4 watt resistors, they should be getting hot.

Also, this energy is unrecoverable, so you want this control current to be minimized. So, you want these resistors to have the largest practical value that doesn't restrict the transistor from operating like a switch (transitioning from a fully ON position to a fully OFF position as quickly as possible).

So, find the highest value for your resistors that does not restrict the operation of the transistor (turning it into a resistor because it can't turn ON fully). Then use three sets of that value of resistor (or slightly less) to make the three voltage dividers to run the three transistors. Each voltage divider will begin by being connected to the ground, which is the negative of the supply and the connection to the emitter of each transistor. The mid-point of each voltage divider will connect to one of the transistor base terminals, and the top of each voltage divider will connect to the bottom of the magnetic reed switch. The top of the magnetic reed switch will connect to the positive supply voltage. This produces three branch currents to operate the three transistors from the single reed switch.

One more suggestion. Your little neon light is coming on because the radiant, longitudinal wave in the inductive collapse sees the long wires of your circuit arrangement as HIGH IMPEDANCE. Clean up your circuit and shorten the circuit paths to bare minimum, so one resistor is right on the transistor from the emitter to the base. The other resistor is right on the base going to the magnetic reed. The neon light is right across the emitter and the collector, and the diode and the bottom of the coil wire are connected directly to the collector. This will eliminate the neon triggering in the first two test set ups.

Keep up the great work!!!

Peter

[Jetijs]

Thanks Steven,
I can't wait to see your results
As for my DC, yes, I use a variac, bridge rectifier and some smoothing capacitors (about 1000uF). I will measure the RPM's with the variac set for 36V and then use three 12v batteries in series for powering to see if a perfect linear DC has any effect on the RPM's

[nali2001]

Hi there Jetijs,
Nice going man great build!

How do you get your DC? I mean a variac is obviously ac and even with a full wave bridge and cap it will not really be clean dc.

On your questions:
Yes the fan will be there for the possible(likely) heat development since every component has losses and also the windings develope heat. And since my system is enclosed it has no real open air cooling like yours so just to be sure I used the fan that forces air through the holes in the end caps.
Laminates are obviously better but since this is a small baby unit and a somewhat experimental approach I play it safe and used the simple solid steel.

Regards,
Steven

[Jetijs]

That noise is coming from the bearings, I am sure about this, because even when I rotate the rotor with my hand, there is that noise. The comutator wheel is a little bit off center, but that is not the cause of that noise, because the noise is there even without the comutator. I will replace these bearings and if that wont work I will consider using those self centering ones
Thanks.

[lighty]

if you check the commutator side you can see that he does have two bearings on there, still I think you may be right adding another bearing, but i would place it on top, where there is none at the moment. Or is that what you actually meant?

Yup, that's exactly what I was suggesting. To put another bearing on the other end of the shaft (on the top).

[Jan H]

@Jetijs

I noticed a strange vibrating sound when your motor is in operation. Something is not balanced or not centered correctly or with low enough tolerance.

Did you consider using another base plate and another bearing so that you can have a 2 point shaft support? You can then use self-centering axial-radial bearing at one side to compensate for eventual lateral movements. I have had experience building powerful motor from scratch and one point support of shaft is simply too unreliable for the forces involved. Especially with pulse motors.

Nevertheless - a GREAT work pal!

if you check the commutator side you can see that he does have two bearings on there, still I think you may be right adding another bearing, but i would place it on top, where there is none at the moment. Or is that what you actually meant?
Also I doubt whether all the noise is coming from the bearings. Tweaking the timing a little bit might also decrease noise.

[lighty]

@Jetijs

I noticed a strange vibrating sound when your motor is in operation. Something is not balanced or not centered correctly or with low enough tolerance.

Did you consider using another base plate and another bearing so that you can have a 2 point shaft support? You can then use self-centering axial-radial bearing at one side to compensate for eventual lateral movements. I have had experience building powerful motor from scratch and one point support of shaft is simply too unreliable for the forces involved. Especially with pulse motors.

Nevertheless - a GREAT work pal!

[Jetijs]

I uploaded a short video on youtube, you can watch it here:
YouTube - Lindemann attraction motor

Peter, I don't understand this balancing of the branches. Can you please explain what you mean?
Thank you

[Peter Lindemann]

You get the GOLD STAR!!!

Ok Peter,
I did some test runs on my motor. The circuit is exactly like in your youtube videos except that I used the 2N3055 transistor and the nominals of the resistors are 200 Ohms (R1) and 196 Ohms (R2). I used only one of the three strands. At first I set the voltage of my variac to 30V and hooked up an analog amp meter. I shorted the output on the first test and got 1.1A current draw and 655 RPM. Next I attached 12v light bulb to the output, now the current draw reduced to exactly 0.4A and the RPM increased to 920. This time I noticed the neon bulb doing some flashes occasionally. It flashed a little bit mode when the motor was starting to gain speed, at the full speed the neon flashed only very few times and dimly. I suppose this is because the bulb can not take all of the charge and what is left goes to the neon. But the light bulb was shining very brightly. For the last test I attached a discharged 1.3Ah lead acid battery to the output. This time I did not see any decrease in current draw, but the RPM went up a little bit to 956. Also the neon bulb was not flashing at all, that means the battery absorbs the recycled energy very well. When I started this last test, the battery voltage increased from 11.60 to 12.24 in some seconds, then it settled down and continued to increase slower. So the results:

Test 1. Voltage 30V Current draw 1.1A RPM 655
Test 2. Voltage 30V Current draw 0.4A RPM 920
Test 3. Voltage 30V Current draw 0.4A RPM 956

And all this with only one strand of gauge 21 wire
Will do some more testing, but I must change the bearings first, because there is some friction in them and that makes the motor noisy and reduce the performance.

Jetijs,

Very, Very Good. Solve your bearing issues and resume your tests. As you add all three strands to the power side (each with their own transistor) your input should triple to about 1.2A. The output to the 12 volt battery will be over 2.0A, so you may want a larger battery to charge than your little 1.2AH one.

Also, you will be able to trigger all three transistors from the single magnetic reed, but you will need to balance the branches to each transistor base.

Keep up the great work!

Peter

[Jan H]

Jetijs,
sounds awesome! can't wait to see a movie of that thing spinning! You really didn't cut any corners. very professional

Bobo,
About the air gap, you are right when you say its hard to get it very small. But the way I imagine it is like having two magnets attract eachother, when they get really close, lets say >1mm the force of them pulling together increases exponentially. I THINK that is what makes the need for a small air gap.

[Jetijs]

Ok Peter,
I did some test runs on my motor. The circuit is exactly like in your youtube videos except that I used the 2N3055 transistor and the nominals of the resistors are 200 Ohms (R1) and 196 Ohms (R2). I used only one of the three strands. At first I set the voltage of my variac to 30V and hooked up an analog amp meter. I shorted the output on the first test and got 1.1A current draw and 655 RPM. Next I attached 12v light bulb to the output, now the current draw reduced to exactly 0.4A and the RPM increased to 920. This time I noticed the neon bulb doing some flashes occasionally. It flashed a little bit mode when the motor was starting to gain speed, at the full speed the neon flashed only very few times and dimly. I suppose this is because the bulb can not take all of the charge and what is left goes to the neon. But the light bulb was shining very brightly. For the last test I attached a discharged 1.3Ah lead acid battery to the output. This time I did not see any decrease in current draw, but the RPM went up a little bit to 956. Also the neon bulb was not flashing at all, that means the battery absorbs the recycled energy very well. When I started this last test, the battery voltage increased from 11.60 to 12.24 in some seconds, then it settled down and continued to increase slower. So the results:

Test 1. Voltage 30V Current draw 1.1A RPM 655
Test 2. Voltage 30V Current draw 0.4A RPM 920
Test 3. Voltage 30V Current draw 0.4A RPM 956

And all this with only one strand of gauge 21 wire
Will do some more testing, but I must change the bearings first, because there is some friction in them and that makes the motor noisy and reduce the performance.

[Jetijs]

Steven,
those are some awesome pictures. Thank you for sharing
I see that you have a fan attached to the shaft, is this for cooling purposes because of the eddy currents? Would it not be better to use thin plates of silicon steel bolted together in one large rotor piece? I know that this stuff is hard to get, but shouldn't it in theory reduce the heat?
Thanks