Confirming the Delayed Lenz Effect

[gyulasun]

Hi Luc,

....  I now understand that a higher voltage across the Shunt Resistor does not necessarily matter if the Phase (cos angle) stays the same. Do I have that correct?

I did not mean it that way, though what you say I think it also matters because an increasing shunt voltage means an increasing input current hence an increasing input power draw while the phase angle may remain the same.  I meant when your LC tank circuit was activated by connecting the 39uF cap, the input current got reduced (voltage drop across the shunt decreased) but the phase angle between the input voltage and input power has changed from the 82° to 34.5°, this caused a significant increase in input power.

....
The First Shot below is the no load centered Reference. We have exactly 4 squares between our rising and dropping voltage phase. Each square should be 22.5 degrees, so then one of the 5 divisions in each square should be 4.5 degrees. The Current is 3 divisions behind the Voltage (3 x 4.5 = 13.5) so 90 - 13.5 = 76.5 degrees.
So would you agree this is an accurate way to get Phase degree?

Yes it is more accurate now than I managed to estimate it but then the 76.5° is even further away from the 90° goal. Maybe you wanted to write 8 squares above instead of 4?

I still don't understand how you came up with the numbers above so I better let you do it. Hopefully in time I will be able to learn the equations

I used conventional AC power calculations whereby the phase angle between the AC current and voltage is considered and you multiply the rms values of current and voltage with the cosine value of the phase angle.
http://hyperphysics.phy-astr.gsu.edu/hbase/electric/powerac.html

I will revise the calculations though I just noticed Thane already gave a hint.

Gyula

[gotoluc]

Hi Luc,

I did not mean it that way, though what you say I think it also matters because an increasing shunt voltage means an increasing input current hence an increasing input power draw while the phase angle may remain the same.  I meant when your LC tank circuit was activated by connecting the 39uF cap, the input current got reduced (voltage drop across the shunt decreased) but the phase angle between the input voltage and input power has changed from the 82° to 34.5°, this caused a significant increase in input power.

Okay I think I get it.

Yes it is more accurate now than I managed to estimate it but then the 76.5° is even further away from the 90° goal. Maybe you wanted to write 8 squares above instead of 4?

I edited it to say:
We have exactly 4 squares between each 90 degrees side of the rising and dropping voltage phase.

I will revise the calculations though I just noticed Thane already gave a hint.
Gyula

Please do the complete calculations. I will go over it on the phone with Thane to get a verbal explanation of how to do the calculations.

Thanks

Luc

[gyulasun]

Hi Luc,

Focusing on your third scope shot now, I assume there is no 39uF connected and CH3 and 4 show the output voltage across the 10 Ohm resistors, respectively.
CH1 is the voltage across the input current shunt and CH2 is the input voltage as before, right?
Now the phase angle between input current and voltage seems to be near 90° for me?  One full wave is 4 msec rounded down (248 Hz is 4.03 msec) and there is 1 msec for the 90° section. The 1 msec section is divided to 5 smaller divisions, this means 18° for any two neighbouring small divisions and I can see 4.5 small such divisions between current and voltage curves, giving a phase angle of 4.5 x 18°=81°

The input power now is P_in=7.16V*0.0177A*cos81°=0.0198W

Output power on one of the 10 Ohm load is P_out1=(0.208*0.208)/10= 0.00432W
Output power on the other 10 Ohm load is P_out2=(0.171*0.171)/10=0,00292W
Summing the two outputs gives 0.00725W

efficiency is (0.00725W/0.0198W)*100=36.6%

Now if you could expand the third scope shot to see more precisely the phase angle (I used an estimated 81°)  and what is more you could attain a phase shift much nearer to 90° degree between input current and voltage, then the real input power would be much less hence efficiency should improve.  The exact 90° phase shift would mean a fully reactive input power and no real input power (cos90°=0)  this is what Thane aims at.

Gyula

[gotoluc]

Hi Luc,

Focusing on your third scope shot now, I assume there is no 39uF connected and CH3 and 4 show the output voltage across the 10 Ohm resistors, respectively.
CH1 is the voltage across the input current shunt and CH2 is the input voltage as before, right?
Now the phase angle between input current and voltage seems to be near 90° for me?  One full wave is 4 msec rounded down (248 Hz is 4.03 msec) and there is 1 msec for the 90° section. The 1 msec section is divided to 5 smaller divisions, this means 18° for any two neighbouring small divisions and I can see 4.5 small such divisions between current and voltage curves, giving a phase angle of 4.5 x 18°=81°

The input power now is P_in=7.16V*0.0177A*cos81°=0.0198W

Output power on one of the 10 Ohm load is P_out1=(0.208*0.208)/10= 0.00432W
Output power on the other 10 Ohm load is P_out2=(0.171*0.171)/10=0,00292W
Summing the two outputs gives 0.00725W

efficiency is (0.00725W/0.0198W)*100=36.6%

Now if you could expand the third scope shot to see more precisely the phase angle (I used an estimated 81°)  and what is more you could attain a phase shift much nearer to 90° degree between input current and voltage, then the real input power would be much less hence efficiency should improve.  The exact 90° phase shift would mean a fully reactive input power and no real input power (cos90°=0)  this is what Thane aims at.

Gyula

HI Gyula,

yes, that is correct!... and no Capacitor used.

Third Scope shot is the same as the Second Scope Shot. Nothing has changed other then the Scopes Voltage and Time Divisions so you can see the complete picture.

So your 81 degrees estimate is not correct.

I have checked all the Transformers I have, even a high end Audio Power Amp Toroid. None of them are 90 degrees out of phase idle (no load). The Toroid is at 40.5 degrees with no load.
So I don't know why we have to be at 90 degrees to get a winner. If we start with 40.5 and we connect a load and it stays at 40.5 then is the Power on the load not coming from Reactive power?

Luc

[gyulasun]

Luc,  sorry for the misunderstanding.... 

P_in=7.16*0,0177*cos76.5=0.02958W

efficiency (0.00725W/0.02958W)*100=24.5% 

You wrote: I found that it's only @248Hz that there is Zero Phase Shift when under Load. See Second Shot below. Frequencies above 248Hz the Phase slowly goes up and below 248Hz Phase slowly drops.

My question is where do you mean exactly the zero phase shift happens at the 248Hz frequency?

According to Thane, a 90° phase shift is the goal between the input current and voltage. This is a situation when input power is fully reactive and in case this happens at the 60 Hz mains frequency than utility WattHour meters in most homes would not measure the load's consumption.  Of course reactive current would still load the mains and this reactive current should still be supplied by the utility providers.
Considering your question on the 40.5° phase angle: it does not represent a fully reactive power like it would when the phase angle were at or very near to 90° angle.

Phase angle, Phi = arctan(XL/R)   so if you have R=120 Ohm and a coil's inductive reactance X_L=150 Ohm at a given AC frequency, than the phase angle between their current and voltage is arctan(150/120)=51.34°  arctan function is available in Windows built-in scientific calculator and click on Inv icon inside it when wish to take arctan. (the 120 Ohm is the coil's DC resistance)  So in this example we have both a reactive and a real (heat) dissipation for this coil. If you had a coil which would have say X_L=800 Ohm at an AC frequency and it would have only 1 Ohm DC copper resistance, its phase angle would be arctan(800/1)=89.92° so pretty close to 90° an almost ideal coil.

This means that to get a phase angle very near to 90° the L/R for a coil has to be a high value, meaning a low resistive part. By manipulating the frequency to find this situation may help but I guess it brings in capacitive part of the coil windings (like for a MOT secondary) and an unpredictable nonlinear core behaviour at the higher than manufactured frequencies, so all in all here is where you have to ask Thane on any further such questions. I really mean this, without any bias.

Gyula