3 phase tests

[Devo]

First off , I am never sure if I should start a new diary or continue on with the old one.

Flux you asked what the resistance of 3/4 inch original stator was.

When I set my digital meter to 200 on the ohm scale & touch the red & black leads together I get .04 with out any coils

When I go across any 2 leads of the original coil I get 1.4

The New 3 phase 1 layer Stator Tests:

Magnets 1 X 2 X 1/2" thick #35 neo,s 8 mags per rotor dual rotor

6 coils , 20 turns # 15 wire any 2 leads read .5 on the 200 scale (meter reads .4 with no coils)

Rotors set to 3/4 " gap

240. rpm
     
3. volts open
   
   

which is .025 volts per wrap

Rotors set to 1/2" gap

240. rpm
     
4. volts open
   
   

works out to .03333 volts per wrap

Rotors set to 3/8 "gap

240. rpm
     
4. 5 volts open
   
   

works out to .0375  

Rotors set to 1/4" gap

240. rpm
     
5. 9 to 6 volts
   
   

works out to .05 volts

forgot to do one at 5/8" but it seems that from 3/4 " every 1/8 less air gap between

the rotors brought the voltage up .5 volts except the last 1/8" from 3/8 to 1/4 which

brought the voltage up 1.5 volts -3 times the amount of the other increments.

I could not go any closer as I was scuffing the coils.

I will post the 8 coil single phase 20 turn per coil results tomorrow ( the stator is

setting as I type this)

can anyone tell me what kind of increase there would be going from a 35 strength to a 42 strength mag?

Should I post pics of these tests & the machine?

Devo

 

[Flux]

Devo

I managed to find your previous post, it looks as though it was Windstuff Ed who asked for the resistance but for any meaningful tests you need to know it.

Unfortunately your meter is not good enough to measure the low values, you need to do it by a volt drop method. Connect your winding in series with a resistor to a 12v battery and measure the current flowing and the volt drop across the coil, then use Ohm's law to find R.

Choose a resistor about 2 ohms so that the current will stay within range of your meter on the 10A range. Use the meter on the 10A range first to measure the current in the circuit, then remove it( the current change will be negligible so assume it is as before). Now put your meter on volts and connect across the coil to measure the volt drop. To avoid measuring volt drop in the connection to the leads supplying the current, connect your voltmeter directly to 2 bare patches on the coil leads not to the external wires.

Your voltage figures with magnet space look sensible. You now need to find the best trade off between available flux and winding resistance. Small gaps will need thin wire but less turns. Big gaps will have more room for thick wire but you will have more of it. To be conclusive you need to choose the maximum wire size you can get in for each gap with the correct turns for a given cut in, then measure the resistance.

The winding that comes out with the lowest overall resistance will give best results.

I think there is about 10% increase in flux if you go to 42MGO grade from 35.

I have forgotten the significance of the single phase winding, but don't be fooled by the fact that for a given number of turns it will give you a higher open circuit voltage, that is not all that matters.

Thanks for taking the time to share your results

Flux

[benjamindees]

Very interesting data.  Clearly there is an inverse linear relationship between gap and voltage.  If you double the gap, you halve the voltage.  We can see this by the fact that voltage divided by the inverse of gap distance (or voltage times distance) is constant, within experimental error:

 voltage  distance  1/distance  voltage*distance

 3           0.75         1.33          2.25

 4           0.5           2              2

 5           0.375        2.66          1.875

 9-6         0.25         4              1.5-2.25

This is suprising because I, at least, expected the voltage to drop off exponentially with increasing gap.  Clearly it does not.  This must be a result of the dual-rotor setup, having magnets on both rotors.  I'd also like to point out that in all cases you were well within a gap of 1 inch, which is equal to the combined thicknesses of your magnets.  I would still expect this linear relationship to begin to change outside a gap of 1 inch.  

[Flux]

You are right, the point at which it becomes non linear depends on the magnet spacing compared with gap. When leakage becomes considerable it goes non linear.

When leakage is small you are just working along the second quadrant of the magnet curve and with Neo it is linear. If you raise the temperature and add armature reaction from your stator you may fall below the knee and get a permanent loss but it doesn't seem to cause trouble generally.

It is not a peculiarity of dual rotors but the geometry is such that the linearity holds for larger gaps.

Flux

[Devo]

Ok got the single phase 8 coils 20 turns of # 15 done

3/4" air gap = 6.3 volts

.039 volts per wrap

1/2" air gap 7.2 volts

.045 volts per wrap

3/8" air gap = 8.3 volts

.051 volts per wrap

I couldn't get the 1/4 inch air gap as the stator had a coil bent out a bit.......

Both the 3 phase & single phase readings are done after the rectifiers

I am still not sure on resistance Flux , is this correct

I take voltage reading at a specified rpm then using a 2 ohm resistor I take another voltage reading at the same rpm then with these 3 variables I can figure out resistance .

Is this correct? I have a lack of knowledge in this area.....

Also the single phase set up as is will put 3 amps into a 12 volt battery at 360 - 400 rpm(1/2 " air gap) , if I make up a 40 turn per coil single phase will this give me 3 amps at half the rpm ?

I am thinking now with the data I have of making a one layer 3 phase with a 1/2"

air gap & 60 turns to give me a cut in between 200 & 240 as I already have the rectifier loss.

Also of doing a single phase 40 turn 8 coil , would this be better with 2 20 turn coils half a turn out both rectified seperate to reduce vibration & then hooked in series or will the vibration even be noticable on such a small machine?

I also want to make a true overlapping 3 phase , I get 7.2 volts with 20 wraps so I

will do 3 layers in thirds & wire the middle one backwards & set it up in star .

7.2 x 1.7= 12.24 volts at 240 rpm , maybe I should go with 25 wraps to get to a 200 rpm cut in?

when I get this done I will test them one at a time into a 12 volt battery at 300 or 400 rpm & see the output numbers.

on another note I think I could rewrap the single phase at 25 wraps & a little thinner instead of 40 & get down to a 1/4 " air gap with a cut in of around 250 to

280 -pretty high though ?

As always Thanks for the time

Devo

[Flux]

You could measure resistance this way as there is no reactance but it would be easier with a battery and a lot more accurate. Hope this sketch explains it.

Single phase -  If you make a 40 turn coil you will halve the cut in speed. Current into a battery will depend on the resistance of the coil.

I am not sure what you mean by using coils half a turn out to reduce vibration. I think you must be implying making it 2 phase with the legs of one coil over the hole of the other. You can't do this in a single layer, it will have to be 2 individual layers or an overlapped coil arrangement.

If you join the 2 phases in series it will be single phase with a voltage of 1.4 x that of one winding and will vibrate.

If you rectify each phase separately then it will be 2 phase and you will only get the volts of one winding but you should remove the vibration.

I don't think your best option will be single phase with a 1/4" air gap, I wouldn't chase that one.

For these tests to be a true comparison you need to choose wire size to fill the available space for each option.

Flux

[Devo]

that drawing is just what I needed , thankyou! , you probably already posted it ten times but if you did I didn't see it-lol

Yes the 2 phase idea was overlapped & rectified seperate, if the rectifiers are joined in series after won't it double the voltage if each set of coils rectifies

at 7 volts?

if I join them for the 1.4 increase & then rectify do I just join in series or do you join the second phase "backwards" like a 3 phase arrangement?

When you say you wouldn't chase the 1/4" stator is that because of strength issues with the thin stator? just curious as it should be low resistance, only 25 turns x 8 coils = 200 wraps my original was 120 wraps per coil 6 coils 3 phase.

Or would it be from vibration, Just curious.

Thank-you for for all the help/advice - I will try to redo both with resistance readings before I move forward , then I should be able to determine what I need

to best "fill" the stator & recieve the max output.

I wish I would have gone with a 9 or 10 inch disc to make the hub a little bigger & more stable (should have bought hugh's book before starting instead of after)-lol

oh well best way to learn I guess , trial & error

Devo

[Flux]

With the 2 phase even if you rectify each and connect in series you still have a phase displacement between the windings, you will only get double volts with capacitors across each bridge and for real power that is a joke. Also you have twice the diode drop. You will do better to double the number of turns in each phase and put rectifiers in parallel.

If you convert to single phase first, without giving it a lot of thought I don't think it matters if you reverse one phase but if it does you take the case that gives 1.4 x phase voltage.

Your thin stator now will win out because you made it fit all cases. If you increase the gap you can use much thicker wire to fill the larger space. Doubling the csa will give you 1/4 the resistance for the same turns. Allowing for perhaps a 50% increase in turns you will still get a lower resistance for the same volts with a wider gap. If you don't fill all the winding space then obviously the winding with the highest voltage will win, that is why true comparisons are difficult. With thin air gaps you waste a lot of winding area just providing safe mechanical clearance.

If you take the air gap off the linear region where leakage becomes a problem then you also loose. My guess is that you reach the optimum with coils 1/2" to 9/16" thick.

Flux