– Clip 1 Introduction – Clip 2 Course Description – Clip 3 Test Parameters – Clip 4 Test Equipment – Clip 5 Overview – Clip 6 Module setup – Clip 7 Configuration and prerequisites – Clip 8 First run – Clip 9 1 KHz – Clip 10 AM MOdulation – Clip 11 Low frequency triggered – Clip 12 Pulse triggered – Clip 13 Summary – Clip 14 End of part 3
Clip 1
Clip 1
In this video I will work on one deficiency we discovered in the previous video. The resistance, caused by the back EMF, avoiding the flow of high current at the level of 13 Volt. WE investigate how the coils take the frequencies and how they behave.
Clip 2
Clip2
Part 3
Solid State Battery charging
2 X 4 Modules via 2 frequencies– Analyse the performance of the isolation transformer via 2 frequencies
– Change ratios of both frequencies to various values and investigate the coil feedback
Clip 3
Clip 3
Test Parameters
1. Voltage source is a Bench Top Power Supply
2. Charging one battery with 12 Volt and 8 AH
3. Two four Amplifier Modules with IRFP450 MOSFET
4, 10 coil/windings Isolation transformer
5. Two Signal Generators
Clip 4
Clip 4
Test Equipment Part 3
Tools
– TTI Signal Generator TG5011A
– Keithley 2110 Bench Digital Multimeter 5.5
– Agilent InfiniVision 3000 Series oscilloscope DSO3014A
– Uni-T DMM UT61A, and VC99
– Agilent U1242B
– HP Laptop
– Bench Top Power Supply 20 Volt 5 A
Clip 5
Clip 5
We are moving now to the third part of the solid state driven battery charging system. We used in the previous videos in Bedini style only one frequency on all 8 modules. Here I will investigate the influence on the coils by using multiple frequencies. All other components are unchanged. The isolation transformer with its 8 coils. In this video I will use as example only two frequencies. The first signal will come from the TTI high precision 50 MHz pulse and wave signal generator. The second signal comes from the build in signal generator from the Agilent oscilloscope. We will both of them and will analyse how that looks like on the oscilloscope.
Clip 6
Clip 6
Here on the table you see the same components we used before. The TTI will drive one segment or the last 4 modules and the first segment is driven by the signal generator of the oscilloscope.
Clip 7
Clip 7
The frequencies I use here in my first example does not have any significance. It did choose them randomly and stopped for a phenomenon I like to show. I use at the TTI 31 KHz and on the oscilloscope around 9.5 KHz. We remember that in the last video we had for 8 modules a current consumption of 810 mA. I will see if we can break that benchmark and drive more current through the coil. The duty cycle on the square wave is for both 41%. We have the battery charged as left over from previous test at 12.37 Volt and leave it at that for a moment. We will look later at charging performance when we find a signal frequency pair which is promising as contender to charge the battery.
Clip 8
Clip 8
I have now a load connected to the battery to keep the voltage low. That is to avoid that the battery is charging too fast. The Agilent DMM 1242B will be used sporadic to show the reactive current from the positive polarity of the power supply to the coils of the isolation transformer. I will start now the test at 13 Volt from the power supply and at 12.1 Volt at the battery. current draw is now 940 mA and YES the load is connected and YES the battery is charging. I will not go into that details but will address that at later stage. On the oscilloscope you see interesting interference patterns. The white line is the power measurement in RMS as product of the yellow line from the pulse to the battery and the green line as the current of the pulse to the battery. We measure 75 Volt Peak-Peak and have 650mA current. That is less what we had before but our RMS value is higher now. The reason for that is that the distribution of power is not stretched over the cycle of the frequency and not only as ?EMF pulse at a very small fraction of the wave. Prove for that is the battery is charging fast with the load connected to the battery. I will now later the frequency on the TTI and like you to observe how the waves behave. What you can witness is that one frequency is dominating and the second frequency is in the background, filling the gap. The do both now mingle and create a new resonant frequency. They stay both separate and influence each other. You also notice that the square wave is visible. The wave form is not distorted. That is the strength of the isolation transformer. Voltage level does not change and there is not much to gain from the current scenario except I can drive more current into the coil at 13 Volt power supply. Another aspect to consider is that we use a soft steel core material which does only keep the magnetic flux at frequencies below 1 KHz. We have at the TTI now 68 KHz and I will now adapt the frequency to the same level and see how that looks like. At 70 KHz we see a pulsating in a slow rhythm. When go now down to 68 KHz we see a very large up and down jumping of the power. Both signal generators operate individually and are not synchronised. That means they will be out of phase and in phase per cycle. In phase is what we have seen. Very little current draw. Out of phase is maximum current draw. This condition does allow me now to create AM modulation by masking the frequency of one signal and let only the other frequency through at the base frequency. Wounding the coil on a different core material and not twisted, we would not see this kind of pattern on the coils. We would be much more flexible in the way we can drive the coils. I will show in the next videos this kind of pattern when we use the Arduino Mega 2560. Instead of using only 2 frequencies we can drive all coils direct. that will be interesting. We can see here potential but it is difficult to align the correctly. It also does not seem to benefit the battery. It does move around the same value. WE move up now the frequency of the oscilloscope and see at twice the frequency again this pulsating of the wave. Now only with a smaller level of current draw. The higher we go the less current we can drive through the coils. I stop the update on the scope and show that we have not much interference of the signals but we still have a large section within the cycle which is without event. Going into MHz we see now wave drains. The are small and voltage is not in the correct degree to the current which has a negative effect on power or the RMS value. Battery is not charging anymore. Load is still connected to the battery and you will notice that the voltage level has dropped to 12.1 Volt.
Clip 9
Clip 9
If we go now in the frequency rang of the effectiveness of the soft steel core, we can get more benefits from the magnetic coupling of the coils through the core. I show here a 1:18 ratio of the frequency. One is set to 1 KHz and the other to 18KHz. You can see the pulsating within the wave. I can adjust that and it will run faster. That is still the sign of the braided wires on the core, or the isolation transformer. When I reduce the value of the 18KHz down to the same value of 1 KHz, see what will happen. I have a higher current draw. I see the interference pattern and can hear a noise which is standing out of the 1 KHz signal. I will put on the core a piece of iron that you hear it better.
Clip 10
Clip 10
I used a little rentch that you can hear the frequency. You will hear a slow staccato in a 1 KHz level. That is 12 Hz different to the first signal. Within that signal is a slow pulse of the power going up and down. I have now the opportunity to increase this by increasing the secondary frequency above the 1 KHz. By doing that I can adjust it to the mains frequency of 50/60 Hz. it would allow us to extract this energy and harness it for further use. At this level we adjust the back EMF and can increase the frequency of it but only in the same encapsulated cycle of the main frequency. But what it allows me is to consume more current and rive the battery charge higher, You see the battery has already reached 13 Volt. I was not able to do that before. We draw between 1.3 and 1.4 A from the power supply and the reactive current is around 1.7 A
Clip 11
Clip 11
If we go now far below the 1 KHz and see the lear back EMF pattern. i raise the frequency and see on the white line how that does impact the RMS power.There is not really much to gain from here. Then higher the frequency the less power draw and less power available to charge the battery. If I go above 1 KHz on the signal of the TTI and above 1 KHZ on the signal of the Oscilloscope things start to get interesting.
Clip 12
Clip 12
The whole picture changes when we use pulses instead of a square wave. The current draw is very high. The pulse width is very low and we in fact create mini back EMF needle son the scope. I am with my pulse width around 10-20 us and the frequency is around 70 KHz and on the scope 10 KHz and pulse width of 70 us. i will reduce that as well down to 10 or better 20 us. WE are now far below what we did consume t our previous test. We have only 300 mA draw and reactive around 500 mA. I increased the pulse width slightly and measure 5 Watt RMS on the scope. The battery is charged above 13 Volt and we have a reactive current of 2.5 A and power consumption around 25 Watt. All individual pulse look like that with small ripple down. We have here extended the back EMF to longer period and can harness more energy in the battery by using the same voltage we used before. However the influence of the isolation transformer can not be broken and resonance is something the isolation transformer does not allow. That is the reason why communication wires are twisted. In order to change that we have to use a new design and use a core material which can operate in very high frequencies. Coupling can be high based on the material we use from low like ferrite to higher like Nanoperm to very high like Metglas.
Clip 13
Clip 13
Summary
We applied one optimisation function by using 2 frequencies. This allows us
now to break the back EMF Resistance and drive more current through the
coils. The general function of the isolation transformer cannot be broken and
we see a close relationship between both frequencies. No resonance can be
achieved of the coils. Below 1 KHz the back EMF dominates. Above that
frequency the coil relations do. One frequency is behaving as a carrier signal
and the second frequency as an amplitude modulation. We can generate this
way in the 1 KHz cycle a 50 Hz signal with a 1 KHz sound or tone you could hear.
Using pulse signal allows us to replicate the back EMF to a far greater deal by
drawing more current and increasing the power to the battery through the down
ringing of the coil. We will look in the next video how individual port signals
via the Arduino mega 2560 will look like.
Clip 14
Clip 14
End of Part 3
Solid State driven Battery charger
via 2 frequency signals