Anup started his engineering when he was in eighth grade, when he tinkered around and started repairing valve radios. He started earning his pocket money from this tinkering. Today radios can be fitted into a 2 mm by 2 mm chip, and being use and throw devices, are unfortunately not repairable. There goes any ideas for pocket money for Reva, Anup’s daughter. Anup’s love for electronics though has continued through his 25 years at KPIT.
Anup started by giving an overview of technology adoption. He gave a very interesting example from the book ‘Crossing the Chasm’, by Geoffrey Moore. The main point that Moore makes is that there is a chasm between early adopters and early majority. The early adopters take emotional decisions, whereas the early majority take pragmatic decisions. This is a big difference in behaviour. This change does not get understood by startups, and hence most of them fall into this chasm and perish. In order to survive, what is required is to a pivoting of the model. Successful startups realise that their success depends not just on breakthrough technologies, as much on the presence of a supporting ecosystem.
In VC terminology, there is a 86% mortality by the time series B is reached. Series B funding is raised by those companies who have established a viable product. By the time you get to Series C funding, you have proven an ability to scale up, By this time another 13% of startups are gone. So Series D onwards funding is raised only by products that have been proven commercially.
I digress here to share some good news that Anup has for budding entrepreneurs (I hope that includes Anup himself) of the COEP batch of 1990. The sweet spot for startup success is highest when the entrepreneur is at the age of 50. Apna time aa gaya hai :-) Here is the reference
- Connected vehicles
- Shared vehicles
- Autonomous vehicles
- Electric vehicles
If you look at the adoption curve, we see that conversion from IC engine to electric is happening at the same pace as the conversion from horse carts to cars happened 100 years ago. In engineering terms, the most challenging part of vehicle design is the powertrain. And for this reason, most auto majors have kept this in-house. Electric vehicles challenge this thinking. Powertrain tech is being democratised by the motor-battery combo. As this technology is simpler, it is available to everybody.
Across the world, there are 200 serious Electric Vehicles start-ups at this time. Anup sees history repeating itself in the EV space. A hundred years ago, there were a similar number of ICE automobile manufacturers. Consolidation happened in a relatively short period of 10 years. He feels that in the next 10 years, there will be probably 10 global EV survivors. My question to Anup was: Is it possible that we will have 8 to 10 component manufacturers globally, rather than automobile companies? Case in point being Shimano in the cycle industry. Anup’s consolidation rationale is based on the hundreds of millions of dollars of investment in launching a new car. And that too with dismal success rates. You need deep pockets to be able to withstand the shocks of product failures. This can only happen with big companies.
From the 1950s to 1980s, auto tech saw stagnation. Electronics made its entry into the automobile with Bosch and their fuel injectors. The eighties saw a movement from passive to active devices. One of the first such devices was the Bosch designed ‘active’ fuel injector, which went on to replace the ‘passive’ carburetor.
The hallmark of good technology is that it is ‘Repeatable, Always Available.’ The first breakthrough for technology was when man could identify a tool and process for creating fire. A software product does not have the frictions associated with the hardware products. There is no inventory, there are no distribution issues. But in order to survive as a software product, you have to be the best in the world, because proximity is no longer a barrier in this market.
The world of computers work with the binary digits of 0 and 1. But in real life, there is no 0 and no 1. 0 and 1 are voltages in the real world. The convention being that anywhere between 0 to 0.8 volts is equivalent to the binary digit 0 and anywhere between 2 V to 5 V is identified as the binary digit 1. The critical issue in a real life system then, is what does the system do when the voltage is shown between 0.8 V and 2 V. In other words, how does a system handle errors? A lot of work in auto software is happening in this space of error management.
60 million lines of code went into the writing the Facebook program. Controlling today’s automobiles requires a 100 million lines of code. And for autonomous vehicles, it is expected to be 300 million lines. Software will soon be deciding the USPs of the car - and as with any service, it can be location agnostic. We have already seen that happening with Tesla’s - Firmware Over The Air updates.
A typical iPhone has 6 core processors. A typical car has got more than a hundred processors, And 8 to 10 networks. There is 2.2 kilometres of cables that run in the car, with 1350 segments. The weight of the wiring harness itself comes to about 42 kg in a typical car.
GPU’s - Graphic Processing Units - that dominate computing in the automobile today - are power hungry devices. A typical GPU consumes 100 Watts of power. In an EV, this is a sacrilege. Alternate technologies emerging are Neuromorphic hardware and FPGA, which promise to reduce this consumption to 5 watts.
Conversion of automobiles from a product to automobiles as a service is bound to happen. We might end up with a Tata Sky kind of subscription model. Interestingly, we have already started seeing this with electric motorcycles like Revolt. Companies might end up making some moolah by selling location based services to their automobile users.
Another interesting hypothesis that he has is that Fuel Cells will dominate long distance transport. He is pessimistic about the use of supercapacitors in automobiles. Another of his hypothesis that I liked was that you will see a rise in the popularity of 10 to 12 seater Ola / Uber taxis. This is the optimal size for an on-demand service.
Battery heat management remains a concern area for Anup. And this is the reason he feels that the adoption will be the best in buses, followed by cars and only then in two wheelers. India's environment conditions, will see a lot of failure rates in summers, as battery performance above 45 degree Celsius is questionable. You will need chilling and air conditioning for the battery to survive such summers. A silver lining on the horizon is solid state batteries.
The organiser of the program was Sangeeta Kale. She is the HOD Physics at DIAT, Khadakwasla, Pune. A doctorate in the area of superconductivity, she is currently doing a lot of work in nano materials, a subset of material science. The Nanotechnologists approach is bottom-up, starting from the atomic level. We chatted up about applications of nanomaterials: sensors, drug delivery, filters and of course batteries. We must invite her to one of our COEP get togethers to get her to talk about Nanotech.