Last month, Ajith “AJ” Kumar, an engineer who works for a company founded by George Westinghouse, got the better of the legendary inventor. Mr. Kumar secured his 362nd patent, one more than old George.

His employer, North Shore-based rail technology firm Wabtec Corp — which acquired GE’s transportation business in 2019 and Mr. Kumar along with it — threw him a party.

GE had done the same seven years and 100 patents ago.

Mr. Kumar, vice president of technology innovation at Wabtec, shows no signs of stopping. He’s got 40 to 50 patents pending. “Not all of them will be issued,” he warned.

Advertise with us

Known as the “father of the AC Locomotive” — his inventions and persistence helped to put electronics at the center of a locomotive’s power mechanism — Mr. Kumar has had a 45-year career during which he never directly managed anyone. Has he found the key to longevity?

This interview was edited for space and clarity.

PG: Tell me the story of how you got your first job.

A: I went to school in India, then came to graduate school at Stanford, Calif. Then I was working in the summer at Green Bank, W.Va., where they have a radio telescope. I think they were looking for intelligence from outer space or something. At that time, 200 people lived in that whole place. You cannot drive a car, you cannot have a TV, because of all the electromagnetic interference. It was very isolated. So I thought, OK, that's not the kind of job I want to be in.

PG: So you applied to GE?

Advertise with us

A: I saw an ad for maybe a (GE) television or a refrigerator or something like that. I wrote to that address, saying that I had this background. I want to work on power and control, simultaneously. (GE) sent a telegram to me saying to call them collect. I called them and they said come to Erie.

(When I got home), there was a telegram waiting in my room. It was an offer. I said, okay, this company must be really good to make decisions that quickly. I should take it.

PG: Tell me about the flywheel bus that I've read about? That sounds fun.

A: In the late ’70s, we were trying to get off oil and end the pollution in New York City. So, the concept was we can make a bus with no engine — no propulsion equipment, no batteries, no nothing. Just a small flywheel. And when you come to a bus stop, (the bus connects to electricity) and spins the flywheel up to high speed for 15 seconds. And that’s enough energy to go 5 miles. In New York City, there’s a stop every quarter of a mile.

PG: Why didn’t that pan out?

Advertise with us

A: I don't know. Maybe GE wanted to go away from that business. They wanted to concentrate more on the locomotives. And I can understand that there may not be too much profit in this niche market. So, how many people would buy that kind of bus and how many cities would do that today? And put infrastructure at every bus stop to charge it ...

PG: You've talked about how it is important to have failures and that you've had failures.

A: Many.

PG: Can you many can you tell me some?

A: Flywheel buses (for) one.

Advertise with us

My last patent — 362. It didn’t become commercially successful either. But it’s a neat concept.

PG: OK, tell me about it.

A: The concept (is) to detect a broken rail with a sensor at the bottom of the locomotive. So, if a rail is broken, the first train which goes by, (a derailment) may not happen. But eventually when the rails are misaligned, it then degrades. So I was trying to say, ‘OK, if I can detect the first broken rail, maybe if the first train doesn't derail, we can prevent the second one.’

PG: Why didn't that take off? That sounds like a great idea.

A: It's a great idea. Great idea have problems, too.

Advertise with us

So, what do you do if you detect a broken rail that’s not (actually) broken? Do you have to stop the train and inspect it? There's a huge cost associated with that. Then there are rails which are intentionally broken, because when you have signals, you don't want the signals to cross through. We need to have a database to say that and it takes a huge amount of effort.

PG: It sounds like this will be commercialized at some point.

A: It will be. Eventually the track database will be available, maybe mandated by government or some other force. To remove the risk of false positives, you need a lot more experiments to make sure that the failure modes are all known.

PG: If there was a technology that you could foresee when you started your career to work on, what would it be?

A: Robotics. Autonomous operation. So you don't need a huge train to be run, you can send every car to drive by itself to wherever it needs to go. And you don't need to transfer or shift tracks.

Advertise with us

PG: What do you tell the young engineers that you mentor? What's the most common question that you ask them when they bring you their ideas?

A: The first question is, ‘Is it going to be practical’ — cost effective? Is it going to take 10 to 15 years to mature? That's the first question they need to think about. Because, you can also spend time on things which maybe take 15 years, but there will be ideas, side projects which come out of it. So I won't discourage any of them.