Development


After researching how induction works we began to develop our own induction charging system.

System Design - Dock
System Design - Portable

First we did a general system design. For the dock we decided we needed USB power, a switcher/oscillator/wave generator, Micro-controller and a coil. For the portable we needed a coil, diode rectifier, voltage regulator, battery charging chip, Micro-controller, LCD, and a Lithium Polymer (LiPo) battery.

The diode rectifier is a simple 4 diode configuration that is mainly used to turn the negative part of a sine wave into an all positive signal.

rectifier diagram

The voltage regulator is a device that holds the output voltage to a high level of accuracy. We picked one that can Step-Up/Step-Down voltage and output a steady current.

converter diagram - chip converter diagram - graphs

The battery charging chip feeds the battery charge, but makes sure the current is optimal for charging the battery at its present state of charge. The chip also will slowly back off the applied charge current as the battery begins to approach full capacity.

charge chip diagram

The micro-controllers are each used for different purposes. The Dock micro-controller is used to generate a 100kHz PWM square wave, and this is for controlling the H-Bridge (full configuration). The portable micro-controller is used to fuel gauge the LiPo battery based on current samples and coulomb counting.

micro-controllers

Next we developed a state diagram that outlines the different situations we would have to account for in the programming and building.


System State Diagram

Next we ordered development boards that would do the outlined functions of the system above, and made a more advanced Dock state diagram for our system. This state diagram was our first take at what we thought would incorporate everything our system needs to do. We incorporated proximity sensors that will enable a LCD to display the important battery data. This can be done simply by programming an interrupt into the PIC firmware that activates based on a compare capture, or the setting of a bit. This was just a first draft and a completed system diagram is shown later.

Shown below is a concept drawing to create a sine wave through Pulse Width Modulation, this digital approach was not as favored as using a full H-Bridge for power reasons.

Sine wave Creation

This is our first set up for making a sine wave for transmitting across coils. Eventually we changed everything on this set up to make our lives less stressful. We did this by first selecting a USB enumeration chip that allows you to program the demanded current up to 500mA in on-board EEPROM. After pulling power from the USB we start a PWM output using a PIC16 instead of the PIC18 shown in the diagram. This PWM goes to a H-Bridge we favored over a Class D amplifier because of the exact load needed by the amplifier requires unrealistic impedance matching conditions on the receiver.

Our first attempt to create a working dock had some problems. First we realized that programming a PIC to enumerate the USB port on your PC for power requires a ton of code. So we decided to use a chip that would do all the USB enumeration for power on its own. When the enumeration has finished the chip toggles a pin from high to low. This is convenient because now we are able to have a PIC micro-controller wait for USB enumeration to have completed before it starts outputting the PWM signal.

Class D Amp

This is the Class D amplifier we tried to use.

We ran into another problem after the PWM part of our system. Our Class D amplifier wanted to see a constant 4 ohm load otherwise it wouldn't work. Our reasoning for choosing a Class D amplifier in the first place was we thought it could easily convert the PWM signal to a sine wave. After realizing that the Class D amplifier wasn't going to work in our system we picked a H-Bridge to replace it.

dock test circuit

This is our dock test circuit with the H-Bridge wired into the circuit above.

Our reasoning for implementing the H-Bridge is we wanted to send 2 PWM signals one to each side of the H-Bridge, but the signals would be 180 degrees out of phase with a minimal amount of dead time to prevent shoot through. This gives double the power, because of the double in peak to peak voltage swing versus a half bridge configuration. This in a way amplified the transmission signal. So we set the H-Bridge up in Full-Bridge mode. This gave us a clean output signal double the peak to peak voltage of a Half-Bridge. Another problem we encountered with the H-Bridge chip is we first picked one designed for motor driving. This was a bad idea because after the PWM went higher than 41kHz the signal began to distort. So we found a new H-Bridge designed for laser diode control and thermal electric cooling rated for 1MHz switching and up to 3 amps. We set up the new H-Bridge in Full-Bridge mode like we had before and began sending the output to our coil. The signal going into the coil from the Full-Bridge looks like a square wave with arched top and bottom rails.

receiving test circuit

After we had our dock transmitting we set up a quick receiving circuit. This was a coil, 4 diode rectifier, switching regulator, and LED. When we placed the receiving coil over the transmitting coil the LED lights up.

Once it was demonstrated that we had power transfer, the quantitative analysis began. Since the Portable was solid by this point, the fuel gauging was reliable and interfaced with the Dock. We measured 5 volts and 130mA into the battery where 10mA of this power was to the fuel gauging and receiver circuitry, and since the Dock was pulling 325mA from the USB we were left with 40% efficiency, 2X our requirement!!.

updated system diagram updated system diagram

This is our updated system diagram


Project Planning


The semester started with a plan to completion, given by the project manager, in the form of a Gantt chart below. The milestone performance is listed on the Home page. We are happy to report that development and prototyping occurred on schedule to meet the project completion deadline.

Click here to view our Gantt Chart in PDF format