In Part 1, I have highlighted the need to reduce the number of private vehicles to mitigate climate change. In Ahmedabad, the usage of AMTS, BRTS and the Metro had dipped 22% in the past 13 years while the population of cars and 2 wheelers had risen by 84% in the same period. It is not that the government is not putting efforts. Recently, the Valsad District Collector tried to implement the ‘Bicycle to Office’ programme but with little success. Mitigation measures are not easy, but as an engineer, I believe that technology can offer elegant options to the mobility sector just as any other.
A Battery Electric Car (BEV) cannot be an answer. Lithium-ion battery manufacturing facilities consume approximately 60 kWh of electricity per kWh of battery. Other steps in the supply chain such as mining and processing of lithium consume an additional 50 kWh of electricity per kWh of battery. A typical BEV battery of 60 kWh capacity would consume 110 x 60 = 6,600 kWh of energy. Manufacturing and assembling the entire small e-car weighing about 1200 kgs consumes approximately 20,000 kWh of energy. In the end, if the consumer charges his/her BEV at night (when power plants are burning fossil fuels), particularly by offsetting his/her roof-top solar energy exported during the day, the entire objective of climate change mitigation is lost.
In terms of GreenHouse Gases (GHG), the harm done by an electric car does not end with its manufacturing and use; even its scrapping produces more GHG than conventional (ICE) and hybrid engine cars according to a report by IIT Kanpur’s Engine Research Lab that calculated the Life Cycle Analysis (LCA) and Total Cost of Ownership (TCO) of vehicles. While BEVs emit 15 to 50% more GHG in comparison to Hybrid and ICE vehicles, the cost of buying, using and maintaining BEVs, too, is 15 to 60% higher per km as compared to the others.
Choose the Netherland Model
We address the climate change mitigation issue by being thrifty. Accordingly, our message to the mobility sector is: ‘Be Thrifty’. Riding a bicycle (the Netherland model) is considered to be sparing, frugal, thrifty, economical and prudent. However, the Indian consumers are observed to be attracted and ‘addicted’ to the US model (cars). Majority are accustomed to driving or riding powered vehicles and it, therefore, becomes difficult to convince them to manually pedal a bicycle. In India, cars are viewed as a symbol of ‘status’, while bicycles as a poor man’s vehicle.
Though the perception of Indians can be changed back only slowly, considering the affinity towards powered vehicles, we decided to take a step in the direction as suggested by United Nations’ COP27 to mitigate climate change. We have, therefore, designed a revolutionary new product – a solar bicycle, a light (< 27 kgs), low cost (< ₹25,000), powered personal vehicle for day to day short commutes. Studies have shown that 80% of the trips travelled are of distances 8 kms or less and best suited for a solar bicycle. Solar bicycles were launched last year, under Project Savitré, representing the worldwide first solar bicycle rollout – a disruptive innovation in electric mobility. Project Savitré addresses 5 main issues faced by the current generation: (a) CO2 mitigation, (b) urban congestion, (c) fossil fuel depletion, (d) oil imports, and (e) deteriorating health. All the 5 issues addressed by solar bicycles are of national importance.
Unfortunately, the current government policies are not conducive for the growth and promotion of solar cycles. The central (Fame II and PLI) subsidies, state (₹ 10,000/kWh) subsidies to EVs, unfair Net-Metering Solar Tariff and Gujarat Solar RoofTop subsidy, collectively, are hurting solar bicycles. While we are engaging in talks with the governments, we are also busy educating the public by publishing articles. This is one such article under the Project Savitré series to help correct the unfair policies and bring in awareness about solar bicycles – the only true net zero vehicles. We would encourage the readers to take the lead and motivate your local community to participate in the ‘Bicycle to Work’ programmes. Climate change mitigation is everyone’s responsibility.
Why Solar MPPT Charge Controller?
To make Project Savitré a success, the solar bicycle needs to be designed and engineered well. To make a solar bicycle safe to ride and look elegant, it became necessary to reduce the size and weight of the solar panel mounted on it. This was one of the challenges to be overcome. To design a suspension to allow the delicate solar panel to absorb shocks on rough terrain was yet another. The former challenge was overcome by designing a solar MPPT charge controller that is efficient and capable of extracting maximum power from a ‘size and weight constrained’ solar panel.
Calculations showed that with an MPPT algorithm, a 40 Watt panel would suffice to meet the charging needs of a 36 Volt, 5 Amp-hour (180 Wh) Li-ion battery in 5 to 6 hours. This enables one to ride a solar bicycle for 15 to 30 kms on a daily basis.
We also designed mounts for 40 Watt solar panels over the rear carrier with suspension. Panels having different dimensions such as (a) 515 x 460 x 50 mm and (b) 638 x 355 x 20 mm, to suit different types of bicycles were chosen. The width of the panels, namely 460 mm and 355 mm, respectively, are less than the front steering handle of a bicycle, thus the panels pose no obstruction while riding.
Block Diagram of an MPPT Charge Controller
Unlike a PWM charge controller that is merely a switch connecting the solar panel to the battery (see Part 1 of the article [1]), the MPPT charge controller is a full-fledged feedback control system.
In the feed-forward path lies a DC-DC converter that boosts the voltage from the solar input voltage level to the higher battery voltage level (see Figure 1). The most critical path is that of the feedback control.
The feedback control block takes information from both the output as well as the input to generate an error signal that is able to control the DC to DC converter.
The objective of the feedback control is to enable the system to operate at the maximum power point which is shown in terms of voltage, current and power as Vm, Im and Pm respectively for a 100 Watt solar panel in Figure 2. The operating point on the characteristic curve shows Vm= 17.7 Volts, Im = 5.65 Amps and Pm = 100 Watts as the maximum power extracted from the solar panel. Assuming switching loss of 6 Watts, the Power output is Po = 94 Watts (conversion efficiency = 94%), and the battery voltage and battery current are Vbat= 39 Volts, Ibat = 2.41 Amps respectively (see Figure 2).
When we use a 40 Watt solar panel instead of 100 Watt (due to the size and weight constraints), for the same battery voltage Vbat= 39 Volts, the battery current would be proportionately lower at Ibat = 0.96 Amps.
DC-DC Boost Converter
A number of DC-DC boost converters are available in the market. A popular one is the 120W, 10-32 Volt input to 35-60 Volt output, 6 Amps DC-DC boost converter module that is based on the Texas Instrument’s (TI) or onsemi’s UC3843 – a fixed frequency, current mode PWM controller chip designed for DC-DC converter applications (see Figure 3). The boost converter is designed with a huge toroidal inductor and two large heatsinks so that it can handle and boost 100W of solar power.
This module is equipped with a multi-turn potentiometer that allows one to adjust the output voltage. One may get tempted to use one such converter with an MPPT charge control. But the problem that arises is due to the nature of the load, namely the Li-ion battery. In Part 1, we have shown how the current-voltage characteristics of a battery differs from that of a solar panel [1]. A battery allows one to draw less or more current at a fairly constant voltage, i.e., the current-voltage characteristic is a straight vertical line. Even with a multi-turn potentiometer, it is extremely difficult to control the operating point on this vertical line due to an extremely sensitive output voltage feedback, and absolutely no input voltage feedback.
Even if one manages to control the operating point with the multi-turn potentiometer, one needs to bear in mind that the battery voltage, though stable at a time instant, is actually rising slowly with time and after a few seconds of charge reflects as a change in the operating point and the multi-turn potentiometer needs readjustment again.
What we need here is a DC-DC converter that takes the feedback from the input (solar) voltage rather than the output voltage. Of course with minor modifications, the 120W converter available in the market can be made to operate with input voltage feedback.
Digital Control
Another problem that engineers commonly face is with the design of digital control using embedded software. Even with the use of a powerful microcontroller such as the ARM Cortex (see Figure 4), and a high speed ADC and DAC, for conversion of samples from analog to digital world and then back to the analog world after digital processing respectively, the feedback action may not be fast enough to ensure stable operation at the MPPT point. This is because a number of samples need to be processed before a sound decision can be taken, namely, whether the solar panel needs to be loaded more or loaded less.
A single wrong decision could make the digital feedback system highly unstable and one could observe an oscillatory behaviour.
Advantage Analog
Unlike a digital feedback system that needs to wait for ADC conversion to process thereafter, an analog feedback system acts almost instantaneously making the system operate in a highly stable state. Unfortunately, an analog system lacks the ability to perform arithmetic and logical operations, compare old results with new ones, or make decisions under different error conditions. All of this is important particularly when one needs to operate at the MPPT point, and not just at any other point on the curve, in a stable manner.
A Novel Hybrid MPPT Charge Controller
A hybrid (analog + digital control) approach is able to capture the advantages of each of the two systems, namely analog and digital, while eliminating the disadvantages of either. Figure 5 shows the feedback control block that controls the ON time of the switching device (MOSFET) in the ‘Power Block’ or the DC to DC converter block. The error amplifier in the ‘Feedback Control Block’ takes in two inputs, namely, (a) the analog (solar) input voltage Vin, and (b) a signal generated by the microcontroller’s DAC. The microcontroller’s signal is generated after the embedded software performs computations based on the solar (input) voltage, Vin, battery or load (output) voltage, Vbat or VL, and battery or load (output) current Ibat or IL. Even if the microcontroller’s output signal gets delayed due to the computation intensive nature of the task assigned to it, the direct analog connection from the solar input (Vin) ensures that the operating point is stable and does not deviate away significantly from the MPPT point.
Use of battery charging (load) current Ibat or IL instead of input current Iin in the feedback loop has multiple advantages. The battery current being lower than the input current, the losses in the current sense resistor is significantly less. Since the battery voltage at any instant, say during a one millisecond computation loop, is practically a constant, the battery current Ibat can be used as a measure of battery power Pbat. Hence the position on the curve where Ibat is maximum can be considered to be the MPPT point.
The microcontroller comes in handy to perform house-keeping operations. Table 1 shows the various states of the charger during different phases of the battery charge operation.
Precaution and care needs to be particularly taken when BMS (Battery Management System) gets activated at a charging voltage level that is lower than the ‘Full Charge Voltage Level’. During such a situation, the microcontroller should terminate the charging process immediately; else the boost converter will continue boosting the voltage and it may reach exorbitantly high values damaging components such as electrolytic capacitors.
Acknowledgment
I thank Chintan Bamania, Hitarth S, and Tarang Patel, BE Electrical undergraduate project students working on the above project under my guidance for permitting me the use of images from their project report for this article.
For Project Students
With an intention to popularise solar bicycles, we have offered details of a novel hybrid solar MPPT charge controller design that has made production of solar bicycles, as shown in Figure 6, feasible. I (author) welcome query or feedback from academia as well as the industry.
References
[1] Vithal Kamat, Solar MPPT Charge Controller or E-Bicycles & EVs – Part 1, Electrical India, ISSN 0972-3277, Vol 63, No. 5, May 2023, pp 44-47.
Vithal Kamat has a Doctorate in Artificial Intelligence from the University of New Brunswick, Canada as a Commonwealth Scholar in 1996. He completed Masters in Control and Instrumentation from IIT Bombay. His current role – reviving a sick industry as a Managing Director of Baroda Electric Meters Ltd. Current interest lies in exploring ways to replace the Human centric Judiciary with an AI Judiciary, to replace the 24-hour clock with Ghati clock, and to replace ICE vehicles with solar vehicles.
