– Clip 1 Introduction – Clip 2 Course Description – Clip 3 Test Parameters – Clip 4 Test Equipment – Clip 5 Overview Test setup – Clip 6 Driver power consumption – Clip 7 First run – Clip 8 Oscilloscope details – Clip 9 Driver current and voltage 1 – Clip 10 Full performance 4 modules – Clip 11 Chart 4 modules – Clip 12 Calculation 1 – Clip 13 Calculation 2 – Clip 14 Overview setup 8 modules – Clip 15 Driver current 8 modules – Clip 16 8 modules run – Clip 17 Second update 8 modules – Clip 18 Final update 8 modules – Clip 19 Chart 8 modules – Clip 20 Calculation 3 – Clip 21 Calculation 4 – Clip 22 Summary – Clip 23 End of Part 2

Clip 1

Clip 1

 In the second part of solid state battery charging I focus on four and eigth modules charging. I will investigate how the power consumption to power output is and the charging performance.

Clip 2

Clip2

Part 2
Solid State Battery charging
Module 4+8

– Analyze the performance of four and eight modules of the Isolation transformer
– Compare power input to power output
– Measure the charging rate and compare with the 60 Watt charger

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. Four and eight Amplifier Modules with IRFP450 MOSFET
4, 10 coil/windings Isolation transformer

Clip 4

Clip 4

Test Equipment Part 2

Tools

– Keithley 2110 Bench Digital Multimeter 5.5
– Agilent InfiniVision 3000 Series oscilloscope DSO3014A
– Tenma DMM 7432,
– Uni-T DMM UT61A, and VC99
– Agilent DMM U1231A, U1232A and U1242B
– Agilent IR-Bluetooth adapter U1177A
– HP Laptop
– Bench Top Power Supply 20 Volt 5 A
– AC/DC current probe to DMM from Agilent and Tektronix

Clip 5

Clip 5

 I will test in this video the 4 and 8 module connection to the isolation transformer. I have a variety of new measurement instrumentation installed to measure various values. I will walk you through what I will measure. Here on the tablet on the right you see the real-time update of all Agilent DMM’s. Later I will use the laptop to record the values and store into a spreadsheet. On the U1242B we measure the battery power in parallel to the Keitley 2110. it is a precision of 0.025% and matches the values displayed from the Keithley 2110 DMM. WE can measure also harmonics which are important to identify for motor applications where harmonics destroy the windings. WE also the option to conduct a temperature differential measurement. That becomes important in a later stage when we measure the temperature of the battery during the charging time. On the U1231a and U1232A I measure voltage and current coming from the power supply and fluctuating to the power supply as reactive power. The power supply does only indicate the energy leaving the device but not what is circulating in the circuit where the power supply is connected to. The Tenma does again capture the pulse current to the battery via the Agilent current probe.

Clip 6

Clip 6

Here you can see the setup on the isolation transformer. Connected is the 4 module part connecting to the first 4 ports. I have also installed to DMM’s to measure now the current and voltage on the drivers to add later on on the board. Every module is getting the same frequency signal

Clip 7

Clip 7

 I have the system now running for a couple of minutes and I like to show you some details before we start logging the values into a chart. The Agilent is measuring with more precision the voltage output with 13.01 Volt. We notice that from the 2  module test that only an increase of 40 mA can be registered. The AC+DC current has increased and I measure on the current side on the Agilent 1.273A. This value is the reactive current of the system. In order to reduce current at this point a choke would be used but we can instead use a transformer and capture this energy and divert it back into the power supply. Her we have the opportunity to store the power for the power supply, creating a self-running system. I have at the moment a load connected to keep the battery voltage at the same level. I show you now the screen of the Agilent logger software where all Agilent DMM’s are connected. I disconnect the load now and watch the update on the screen. You see the voltage of the battery increases rapidly and the current in the circuit drops slightly. The power voltage does not change at all. Connecting the load again reverses the process.

Clip 8

Clip 8

 Here on the Oscilloscope I like to provide with some updates. We measure on the voltage spike 150 Volt Peak to Peak. That is slightly lower than before. Current is only 2.7A. It is much less than before. With the peak voltage and current we would calculate around 250-300 Watt. However if we look at the RMS per cycle we are left with not more than 2 Watt. this has of course only theoretical value because with 2 Watt a battery charge cannot be performed. It also indicates that we have a lot of losses in our setup. We have the reactive power not used. We do not use the magnetic flux of the coil. The coil design and using the same frequency is creating a high resistance based on the back EMF and is restricting current to flow.

Clip 9

Clip 9

 We take a measurement now of the power requirement of the drivers and read 6.28 Volt and 54 mA

Clip 10

Clip 10

 I will start now with my first test. The system is currently connected to a load and kept in balance. I start the logging and remove the load. As you can see in a couple of second we charged 200 mV. It is very fast charging and I do not require to take a break and come back for more updates. It is nice to see that the U1242B and the Keithley 2110 showing identical values. We have charged in one minute about 300 mV. The Keithley logs in 5 second intervals and you see the 2 digit jump of charge whereas on the U1242B you see the increase gradual. I let it run to 500 mV and make break than. I will continue than with the 8 module test.

su_divider]

Clip 11

Clip 11

 On the chart you see the very step charge at the beginning and slowing down. I have selected a section in the middle and measure for a 100 mV charge 1 Minute and 25 Seconds

Clip 12

Clip 12

 We integrate our measurements into the formula

V = 250 Volt Peak
I = 2.7 A Peak
RMS = 1.93 Watt

Reactive Power
V = 50 Volt I = 1.263 A, 63.15 Watt RMS
Input power = 13 Volt * 0.79 A = 10.27 Watt

COP = 6.14 or 614 % of performance
(Potential not utilised)

Clip 13

Clip 13

  We integrate our measurements into the formula

V = 13 Volt Power Supply                                    Driver V = 6.28 Volt
I = 0.79 A                                                                        I  = 0.054 A
P= V * I = 13 Volt * 0.79 A = 10.27 Watt            6.28 Volt * 0.054 A = 339 mW

Frequency 420-440 Hz, Pulse width 40-41%
100 mV charge in 1 Minutes 25 Seconds * 15 = 21.25 Minutes or

Power requirement = 10.27 Watt * 0.2125 h = 0.0037 KWh

From a 60 Watt battery charger, charged to 13.5 Volt in 39 Minutes and 12 Seconds
Power consumption is 39.12 Watt or 0.03912 KWh

The charge power performance is 1057 % compared to the standard
charger or COP of 10.57 and from a charging time of view 184% increasing
the overall COP to around COP =19.45 or 1945 %

Clip 14

Clip 14

I like to give you an update of my 8 module connection scenario. We use 8 individual connections to 8 ports of the coil. Both modules are connected in parallel for the driver power and the impulse power to the battery. All driver input signals are the same, getting one frequency provided.

Clip 15

Clip 15

 We have a look now on the driver power consumption and read 80 mA for 6.28 Volt. WE continue with the 8 module test.

Clip 16

Clip 16

 We have at the moment a load connected to keep the voltage below the threshold of 12 Volt. I will start logging and will disconnect the load. The first thing we notice is that we consume only 20 mA more compared to 4 Modules. This indicates that the resistance in now very high for the current to flow. The first drawback of manny to see. I measure on the driver side 103 mA for 6.26 Volt. We have an increase of current on the pulse current to the battery, as well as the current increase on the fluctuating current. The oscilloscope does still read 2.7 A. Peak – Peak. Voltage dropped dramatically to 185 Volt. The voltage for the reactive power has dropped as well. The math on the oscilloscope measured now 2.5 Watt RMS. That is only slightly higher than with 4 modules. We have charged in 1 Minute and 10 Seconds 400 mV.

Clip 17

Clip 17

 We are 1 – 2 Minutes in the test and have reached 600 mV. The current between the power supply and the power mosfet is dropping. That indicates the increase in resistance on the battery. Even so when not direct in relation to it. We harness only the pulse not the direct current . There is feedback effect to the reactive power reducing the current. The power supply is stable delivering the same power.

Clip 18

Clip 18

 We reach now 12.7 Volt and the charging period for 100 mV takes now longer. We are 13 minutes in the test and have charge about 700 mV. Because we have a lot of losses now I consider in future tests to increase the voltage and the pulse width to see if I can compensate for that. Looking at 20 mA more current requirement but only marginal increase of performance makes it almost pointless to use 8 modules. In order to optimize that we have to make a couple of changes. These changes are design based and not just adding more power.

Clip 19

Clip 19

 Even when almost no value is added the average time fame looks better with 1 minute and 5 seconds. Please note that this is only a theoretical value and does not provide the full details. We use it as a calculation example between the modules.

Clip 20

Clip 20

 We integrate our measurements into the formula

V = 185 Volt Peak
I = 2.7 A
RMS = 2.53 W

Reactive Power
V = 45 Volt, I = 1.338 A = 60.21 W RMS
Input Power 13 Volt I = 0.81A = 10.53 Watt

COP = 5.72 or 572 % performance
(Potential not utilised and less than compared to 4 Modules)

Clip 22

Clip 22

Summary

It becomes apparent that adding modules is not equal to more performance.
The problem we have here is that the EMF is causing a large resistance in the
coils, avoiding current from flowing. We see only a marginal current pickup and
that is even negligent more with 8 modules. This does call for a design change.
This is the scenario for the Bedini SG charger of the wheel and the solid state charger.
The deficiencies are
1. Very small time frame for the back EMF within a cycle to provide power.
2. High resistance of the coils themselves based on the back EMF
3. High reactive power not used
4. Magnetic flux leakage not harnessed and therefore wasted energy
5. Battery resistance increases with charge or voltage level and making it not
possible to fully charge the battery. A small sealed lead battery charger with a
power consumption of 10 Watt does charge the battery to over 14 Volt in 30 Hours.
A total of 0.3 KW of power consumption. Almost 10 times more than the fast 60 Watt
battery charger. And 16 times longer.

Clip 23

Clip 23

End of Part 2
Solid State driven
Battery charger

Clip 29