– Clip 1 Introduction – Clip 2 Course Description – Clip 3 Test Parameters – Clip 4 Test Equipment – Clip 5 Discharge full battery explained – Clip 6 Measure resistance – Clip 7 Discharge Start – Clip 8 Discharge 2 – Clip 9 Discharge chart 1 explained – Clip 10 Discharge chart 2 explained – Clip 11 Discharge chart 3 last section – Clip 12 Bounce back voltage chart – Clip 13 Bounce back voltage – Clip 14 Discharge chart overview – Clip 15 Power chart – Clip 16 First Summary – Clip 17 Recharge Arduino – Clip 18 Update recharge – Clip 19 Update 2 recharge – Clip 20 Update recharge increased voltage – Clip 21 Update recharge increased voltage 2 – Clip 22 Update recharge increased voltage 3 – Clip 23 Update recharge increased voltage 4 – Clip 24 DC module introduction – Clip 25 Recharge DC module 1 – Clip 26 Update DC module recharge – Clip 27 DC module increased voltage – Clip 28 DC module at 13.4 Volt – Clip 29 DC module charge complete – Clip 30 Arduino charge chart – Clip 31 DC module charge chart – Clip 32 Calculations 1 – Clip 33 Calculations 2 – Clip 34 Prepare for discharge – Clip 35 Start voltage for discharge – Clip 36 Discharge start – Clip 37 Voltage bounce back – Clip 38 Discharge chart explained – Clip 39 End of discharge – Clip 40 Discharge chart – Clip 41 Calculations 3 – Clip 42 Second Summary – Clip 43 End of Part 5

Clip 1

Clip 1

In this series I will look at charging power and power delivery of a 8 AH battery. This small battery is recommended not to discharge more than 24 Watt. I therefore decided to use a reference load of 13-15 Ohm to keep the discharge in a safe margin.

Clip 2

Clip2

Part 5
Performance Analysis of a Back EMF charged battery

– Use a reference load to drain the battery to a threshold of 11.5 Volt
– Recharge the battery to 13.5 volt
– Apply the reference load again to drain the battery to 11.5 Volt
– Compare power output to power requirement to charge the battery

 

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. Arduino Mega 2560
6. LM2596S DC-DC step down voltage regulator, 3 A
7. DC charging module with Back EMF spike

Clip 4

Clip 4

Test Equipment Part 5
Tools

– EA, PS 2000 B, 84 V 5 A, Power supply
– Keithley 2110 Bench Digital Multimeter 5.5
– Agilent InfiniVision 3000 Series oscilloscope DSO3014A
– Agilent U1242B, U1231A, U1332A
– Agilent U1117A IR-Bluetooth adapter
– HP Laptop

Clip 5

Clip 5

We charged the batteries with the Back EMF very fast and as logical conclusion we now need to test if the so called surface charge is going to be a deep charge.  Before this demonstration I conducted tests and can confirm that we have to use a longer charging time frame with lower power levels. The batteries don’t take well the voltage spikes and become unstable.

I will use a higher frequency for the spikes and a lower voltage and current delivery. I will also introduce a DC module I developed as a first prototype which I will dedicate more time on in future videos.

Clip 6

Clip 6

Our first test is to measure the resistance. We add all the cable as well and get around 15 Ohm. You can apply Ohm’s law and divide it from the voltage of the battery available to see how much current it can draw.

Clip 7

Clip 7

I will start the logging now until I get the first reading and will then apply the current draw. We see a current draw of 867 mA and the voltage drops to 12.647 Volt. YOu see around 30-50 mVolt drop for each interval of 5 Seconds. It produces 10.62 Watt and not 12-13 Watt as thought.

Clip 8

Clip 8

 Now we see the bounce back voltage of the battery. It is a standard behaviour here at 12.389 volt and it stays at this level. You will notice as well the current on the bottom is increasing at the same voltage level. Voltage has now increased to 12.390 volt. I let that run and get to you in a couple of minutes.

Clip 9

Clip 9

Here you see the voltage and current chart. It is nice to see a straight line after the voltage settles to the load. It will take longer until the curve is dropping.

Clip 10

Clip 10

 On this chart further in the test you see the slow but steady decrease of voltage. The current drops only slightly. This has to do with the ratio of values and is always related to each other as power after Ohms law and Ampere’s law.

Clip 11

Clip 11

This chart show the last section of the test isolated. Voltage and current drop in parallel.

Clip 12

Clip 12

The load is now disconnected after we reach the 11.5 Volt and we measure now the voltage bounce back. We let it settle until it does not move any higher and will then start with the discharge.

Clip 13

Clip 13

The battery has reached now 11.833 volt and only increases slowly. I will add all details now on the board as first overview of the discharge performance.

Clip 14

Clip 14

We discharged the battery in 5 Hours and 44 Seconds. It is much higher compared to the 60 Watt load we used before.

Clip 15

Clip 15

The average power consumption over time was 9.47 Watt.

Clip 16

Clip 16

 First Summary

The power delivery did prove to be better than expected. This battery has
been charged to 15 volt with a combination of Back EMF spikes and pulsed
DC. The voltage was settled for two weeks without charging for this test.
This test need to be repeated many times in order to define a rule for charge
and discharge. Calculators are only guidelines and vary wide. For a commercial
power system with often swapped batteries the tolerance need to be tide to
deliver the power required.

Clip 17

Clip 17

We measure on the battery now 11.933 volt and will apply a new charging pattern I designed and use 765 Hz as frequency for the Arduino board and go down in the voltage to 11 Volt. The current is regulated on the power supply and the initial value is the threshold I did configure which settles after some seconds to the actual current draw which is 330 mA. The battery charges steady but slowly. I realised that this is more safe and provides more reliable results from a charging point of view. I will calibrate the oscilloscope now in order for you to see the values.

Go to top

Clip 18

Clip 18

I set the voltage now correctly and see a current draw of 330 mA at 11 Volt. This represents only a 5th of the original power consumption in previous tests. On the scop you notices the same spike for each coil but also a much longer current curve. This means we can harness more energy in the battery for longer.

Clip 19

Clip 19

 We are now 10 minutes in the test and have 170 mV charged. In the interval of 5 seconds we charge 1mV. However we charge only with 3.4 Watt power consumption.

Clip 20

Clip 20

I have now after 300mV charged the voltage on the power source increased to 12 Volt. This does draw 430 mA from the power source. The current based on the increased resistance did drop so I had to increase the voltage to compensate. The charging level is the same now as before. 1mV per 5 second interval.

clip 21

Clip 21

 I have at 12.490 Volt the voltage increased again to 12.6 Volt. We draw 530 mA now. It is good to see that the current rises as well we have available to supply to charging battery.

Clip 22

Clip 22

 I have increased the voltage now to 13 Volt. We have charged 12.6 Volt on the battery. We draw 600 mA. You see on the cope that we bypassed a threshold where the activity in the coil becomes more turbulent. This is a difficult point because the battery to be charged does not like it when charged high in that fashion.

Clip 23

Clip 23

 We have reached now our second milestone of 13 Volt charged and used the last part charged with 14 Volt. That was 15 minutes ago. Current draw is 710 mA. I will stop that now her. Let the voltage come down and will restart charging with new pattern to 13.5 Volt.

Clip 24

Clip 24

 I here you see a circuitry I build. It is simple a combination of DC charging and Back EMF. On the top right you see a capacitor bank. It is 50 volt and 10000mF. I reduced with a TVS the voltage to 33 Volt. I use then 3 step down voltage regulator. One is used to drive with 5 Volt the driver of the mosfet. The other one is used for charging the battery with 14.5 Volt. It will not apply 14.5 Volt direct because voltage drop based on current draw will apply. I will not go to much into details. I will dedicate a larger part with Toroids development in future videos.

Clip 25

Clip 25

 The voltage on the battery has settled down and we read 12.541 Volt. That is our starting point to charge the battery with my new device. I apply 11 Volt and start up. The caps require a small tab from the battery in order to provide power to the driver. The system is then self regulating afterwards. We measure at 11 Volt a current draw of 110mA and we charge 3 mV every 5 seconds. You see on the cope nicely the pulses from voltage and current. We measure 80 volt peak to peak. The power is around 2.6 Watt peak to peak. The power we draw from the power source is very small. Only 1.1 Watt but the charging rate is much better as before. the efficiency of this device will increase when I optimised the core winding techniques.

Clip 26

Clip 26

 I have disconnected the Back EMF from the battery and we have achieved a charge of 12.953 Volt. Now we charge pure DC. The resistance of the battery has increased and as a result the current drops down to 80 mA. Connecting the Back EMF has a negative impact based on wrong design and false impressions. I will dedicate a video about that to highlight the issue and the wrong concept of pure Back EMF charging.

Clip 27

Clip 27

 We have reached 13.119 Volt and I have increased the voltage. The current did drop and we could not increase the charging level if the battery. I have also reduced the frequency to be able to drive more current into the system. The voltage into the battery doe not change. It is voltage regulated. I read the voltage on the capacitor and the voltage on the regulator. Botht do not change. Technically it should not have any impact on the battery to charge but it has. The charging frequency has increased. I will continue to charge to 13.5 volt and wait for the voltage drop to settle before I start with the discharge procedure.

Clip 28

Clip 28

 We are now 35 minutes in the test and we have charged 13.395 Volt. The charging level per interval of 5 seconds is about 2-3 mV. The voltage on the voltage regulator is now increasing based on the charge level of the battery. This is good way to charge a battery always with the right level depending on the charged level. it allows us to keep charging the battery even when no more current is supplied or the voltage increased. The voltage regulator shows 14.18 volt and the capacitor indicated 14.9 Volt. The current on the power source has decreased. We reach soon the 13.5 volt and will give it 30 minutes before I start to continue with my test.

Clip 29

Clip 29

We have reached now 13.5 Volt charging level on the battery and I will disconnect the power source. I watch the voltage drop and decide, when no much movement on the battery is visible to start to drain with my reference load.

Clip 30

Clip 30

The chart displays the performance with each voltage level applied. The charge time is much longer as it was before. The total time to 13.5 Volt is 2 Hours 19 Minutes and 43 Seconds. On the voltage drop we see first the steep drop and then the decrease until it would be parallel. I have not waited so long and will start at an earlier level.

Clip 31

Clip 31

On the DC module chart you see 4 areas which indicate the voltage increased or the use of the Back EMF. The overall charging time was much faster as was the reduction of current requirement from the source.

Clip 32

Clip 32

We integrate our measurements into the formula

Arduino Back EMF pulse charge, 765 Hz
Time: 2 Hours 19 Minutes and 43 Seconds

Power: 20 Seconds 3.63 Watt
1 Minute and 26 Seconds 5.16 Watt
5 Minutes and 55 Seconds 6.68 Watt
61 Minutes and 36 Seconds 7.80 Watt
10 Minutes and 22 Seconds 9.94 Watt

DC Module Charge
Time: 33 Minutes and 1 Second

Power:
5 Minutes and 55 Seconds 1.1 Watt
4 Minutes and 55 Seconds 0.88 Watt
22 Minutes and 11 Seconds 1.68 Watt

Clip 33

Clip 33

We integrate our measurements into the formula

Total Charge time to 13.5 Volt
2 Hours 52 Minutes and 44 Seconds

Total power consumption

DC Module average 1.51 Watt for 33 Minutes and 1 Second
Arduino average 8.8 Watt for 2 Hours 19 Minutes and 43 Seconds

Charge time and pattern need improvement and depends on the
resistance and capacity of the battery. A dynamic automatic
charge pattern need to be programmed.

Clip 33

Clip 34

 We start now with the discharge test. The voltage drop is now steady with 1 mV every 5 seconds. We have passed the 20 minute mark and I will go down to 13.3 Volt.

Clip 33

Clip 35

 We reach now the value of 13.3 Volt and I have prepared everything to record the discharge time. yes this value is much higher than the charge of the battery before. It will tell us if the voltage level is an indicator for the charging level of the battery as stated.

Clip 33

Clip 36

 We are now a couple of seconds in the test and notice the same current draw but the voltage drop was much higher than before. The voltage is dropping rapidly and it does not look good. At this rate we will not get the same power out of this battery.

Clip 33

Clip 37

 At 11.8 Volt we see that the battery is bouncing back. This is interesting because it is not related to the increased resistance of the resistor due to heating. The temperature is about 60 degree Celsius. What we see is an increase in voltage and an increased in current draw. It seems that the reserves in the battery is tapped and is fighting the current draw from the load.

Clip 33

Clip 38

 On the chart you can both curves for Voltage and current. We see the deeper voltage drop and the increas later. It continues from here steady with a trend upwards.

Clip 33

Clip 39

WE have reached 11.506 Volt and the chart does show the overall discharge curve. Discharge this time around was much faster than before. I will add all information now on the board and summarize afterwards.

Clip 33

Clip 40

The discharge chart does show all details at a glance and gives for the whole time frame 2 Hours  24 Minutes and 22 Seconds. That is roughly 50% of the previous charge. The battery in the first test was charged to 15 Volt.

 

Clip 33

Clip 41

We integrate our measurements into the formula

Power charge to power discharge comparison

Charge
DC Module average 1.51 Watt for 33 Minutes and 1 Second
Arduino average 8.8 Watt for 2 Hours 19 Minutes and 43 Seconds

Discharge
Average 9.94 Watt for 2 Hours and 24 Minutes and 22 Seconds

In Joules for charge we have 76762J
For discharge we have 86100J
This is a COP of 1.12 or 112% of performance

Clip 33

Clip 42

Second Summary

It is difficult to understand the charging pattern and the therefore resulting
performance. It looks like we are on the right path but I noticed many issues
which need to be addressed. A production system need to be able to adapt
to any battery capacity. That means we require components and configuration
for each battery type and capacity. The DC module did perform much better
which is surprising.
This battery type Is under normal condition charged with a small charger
which requires 13 Watt. We might use with Back EMF too much power,
because the configuration is incorrect. To get to the 5 Hours discharge time
we have to charge to 15 Volt as well.

Clip 33

Clip 43

End of Part 5
Solid State driven Battery charger
Performance Analysis of a Back EMF charged battery

Clip 33