By Jim Ringold

Replica of the first transistor. Public domain.

The Road to Solid-State Electric Vehicles – Part 1

The first step is to tell the obligatory joke about computer controlled cars. How to restart a broken electric vehicle (EV) using Microsoft software? You close and open the windows of course!

But this joke makes a point. The computer of the EV (which controls the entire vehicle in the case of Tesla), the central processing unit (CPU) and associated software / firmware that tells the semiconductor devices of an EV how to operate, must operate the VE without fail. The computer just can’t freeze at 60 miles per hour! To help explain how it works, this is a multi-segment discussion of “solid state” devices, including central control processor (s) and support. Firmware (instructions embedded in the CPU) and Software (instructions written on the computer that can be changed at will) that make an EV possible.

All of this central control is required and works through commands and feedback from all of the other semiconductor devices to perform the operational functions of an EV. Some vital operations are a) keeping the battery in good condition, b) starting, steering and stopping, and c) operating lights, wipers, turn signals, horn, etc. You get the drill. Then there are the secondary functions that are not vital to the safety of people – power windows, door locks, environmental controls and displays showing vehicle status, navigation, entertainment, etc. (Note: Displays have their own processor to display images on the screen.)

It is good to keep in mind during this discussion that when Mr. Musk presents designs he always tries to use “first principles thinking.” Tesla’s design goes back to basics at the start, starting from scratch. It ignores “this is how we’ve always done it” to form new ideas and make clear design / setup decisions.

Also, keep in mind that the thought of the first principle is extremely unfair to existing car manufacturers, as they do not immediately have that luxury by thinking of the first principle. These manufacturers have a huge investment in existing tooling and designs. They must train the required new computer engineering staff, establish new suppliers and, worse yet, continue to build internal combustion engine (ICE) cars in the meantime. All of this is seemingly unfair and poses an even greater challenge for a vehicle company leadership team that does not have the new “solid state” electric vehicle assessment skills needed to chart the best course. the future of electric vehicles for their business.

It took a lot to prepare the ground to make the current harvest of electric vehicles possible. Batteries must have evolved far beyond the long-standing lead-acid batteries that have been used to start fossil fuel cars from the invention of the electric starter, to the series of batteries based on gasoline. lithium which made the laptop practical, but even a step beyond them. Solid-state switches (relays) had to be perfected to control the high power supplied by the new lithium-ion batteries.

“The Regency TR-1 which used Texas Instruments NPN transistors was the world’s first commercially produced transistor radio. Image courtesy of Cmglee (CC BY-SA 3.0 license).

Solid State – Discrete Transistors – 101

Before the term “solid state” was coined, these switches were called transistors, invented in 1947 by Bell Labs in Murray Hill, New Jersey. A bipolar junction transistor (BJT) entered the market in 1950. A large number of handheld “transistor radios” in 1955 entered the market (note that $ 49.95 then equals $ 350.00 today. hui). The low power consumption of these small, discrete three-legged transistors allowed the use of AA series batteries for power. Until then, red and shiny glass vacuum tubes were used as switching devices for radios and early primitive computers.

A quick word on logic switching in the world of Boolean algebra: First developed in 1874, this is one of the earliest human developments leading to the creation of EV. Boolean deals with two states, true and false, denoted 1 and 0. In the 1930s, it was discovered that Boolean algebra could be applied to switching circuits. This matured into two types of transistors. The first transistor switching uses the gate “and” – as in “this and this “allows electricity to continue to flow. The second is a door “or” – as in “this Where this “allows electricity to flow. There is also a “no” gate that we will ignore here. To date, all computers and control devices use these rules, but in a more sophisticated and faster way.

The first solid-state computers (early 1950s) used these small “discrete” three-legged transistors to perform functions. They still took up space, but not as much as the vacuum tubes which required an entire room. The vacuum tubes created a lot of heat, so a lot of air conditioning was needed. IBM announced its first discrete transistor computer in 1955.

The first transistors used germanium as a switching medium. Bell Labs accidentally discovered that silicon dioxide works better. We have also learned to superimpose materials to make transistors of better voltage and higher voltage. Then comes the MOS, then in 1959, the MOFSET transistors are developed. MOS transistors became the first personal computers in the 1970s.

While good for processing 1s and 0s, the early transistors could only switch up to about 22 volts DC. You cannot turn a 110V bulb on and off with a transistor. You had to use a mechanical switching relay with the low voltage of the 22VDC transistor to operate the magnetic coil and the relay contacts to drive the 110VAC to light the bulb.

The higher heat and voltage that a MOFSET transistor could handle began to pervade the power supply industry. This brings us to the 1970s. There is a whole alphabet soup of transistors developed in the following years. They became faster, cheaper, more reliable and could withstand higher voltages. But still far from what it would take to switch a high voltage, high amperage battery on and off in a variable manner to control an electric motor like an electric vehicle.

The next step needed for an electric vehicle were smaller devices where thousands of transistors would be needed for logic 1 and 0. The integrated circuit that could “fit” hundreds, thousands and more of transistors into integrated circuits would be needed. . But that’s the story of part two, the technological advancements needed to build usable electric vehicles.

Appreciate the originality of CleanTechnica? Consider becoming a CleanTechnica Member, Supporter, Technician or Ambassador – or Patreon Patron.


Got a tip for CleanTechnica, want to advertise or suggest a guest for our CleanTech Talk podcast? Contact us here.

Source link