He measures the soot your engine tries to hide - and he has been at it longer than most startups stay alive.
Inside a small building in Ladera Ranch, California, a five-person company is trying to do something the giants of the auto-parts world have spent years circling: build a sensor that can watch tailpipe soot in real time, cheaply, for the entire life of a vehicle. The company is EmiSense Technologies. The sensor is called PMTrac. The person who has staked the better part of a career on it is Patrick Thompson.
The pitch is deceptively plain. Modern engines pass their soot through particulate filters, and regulators increasingly want proof that those filters keep working long after the warranty lapses. Most existing particulate sensors are resistive - they collect soot until a threshold trips, then burn off and start over, leaving engineers to model the gaps in between. EmiSense's sensor is electrostatic. It produces a continuous signal, responds faster, and reads finer particles. That is the whole bet, and it is not a small one.
In March 2025, the bet got its biggest validation yet: EmiSense announced an agreement with a global Tier-1 automotive supplier - the kind of company that manufactures parts by the millions for the world's automakers - to commercialize the technology. Samples started shipping in 2025. Serial production is targeted for 2028. For a sensor that has lived in the lab and on test benches for years, that timeline is the difference between a clever idea and a product on the road.
Then, in March 2026, Thompson widened the aperture. EmiSense launched a low-cost version of the same electrostatic sensing for large diesel generators - the backup power that keeps data centers and hospitals running. The framing was pure operator: the monitoring system, the company said, costs less than 1% of a generator's value. Cheap enough that nobody has to think too hard about installing it.
"Now that we have initial adoption in automotive, and a strong Tier-1 supplier partner, we are eager to support evaluation and adoption in other segments," Thompson said. It is the sound of a founder who has finally cleared the hardest gate and is looking at the rest of the map.
ePM technology has been selected as the most promising for meeting future OBD requirements.
Look across Thompson's resume and a habit emerges. He keeps finding signals that are hard to capture, then building the instrument that captures them. Out of UC Berkeley in 1990, he landed as a founding employee and VP of Operations at New Media Corporation, a maker of laptop peripherals that Tyco International eventually absorbed. The product was hardware; the lesson was operations - how to actually ship physical things at volume.
Next came InterWorks Computer Products, which he co-founded to supply modular digital audio subsystems - VOIP building blocks for the networking-equipment vendors wiring up the early internet. Sanmina Corporation, a contract-manufacturing heavyweight, acquired it. Different signal, same instinct: package a hard technical capability into a module someone bigger wants to buy.
Then the turn toward engines. Thompson founded Innovate Motorsports, built around exhaust-gas sensors, and there he made something enthusiasts still talk about: the LM-1, a wideband lambda meter that reads air-to-fuel ratio. For a generation of tuners and racers, the LM-1 was the gauge you trusted when you were chasing the perfect mixture. AutoMeter Products acquired the line. The garage legend became a product on a shelf.
EmiSense is the fourth act, and the most ambitious. The signal this time is soot - ultrafine particulate matter measured in real time - and the stakes are regulatory, environmental, and global. It is also the one he has not sold. After three exits, Thompson is still in the CEO chair, which tells you something about how much this one matters to him.
The old way of sensing soot waits for buildup, then resets. EmiSense's electrostatic approach reads a continuous signal - which, in a world of tightening Euro-7 and Tier-4 rules, is the difference between guessing and knowing. A loose, illustrative read on what the company claims its sensor improves:
Bars are illustrative of EmiSense's stated advantages (faster response, superior sensitivity, continuous output), not measured benchmark values.
Diesel, gasoline, and hybrid engines - on-board diagnostics that confirm a particulate filter is doing its job for the life of the car.
Large diesel generators in data centers and hospitals, where running beyond strict "emergency" conditions triggers tougher Tier-4 rules.
Industrial and multi-fuel setups - natural gas to off-road - where continuous soot monitoring was previously too costly to bother with.
Low-cost continuous monitoring makes a ton of sense - whether for pre-certification testing, evaluating retrofit components, or ensuring energy availability in advance of any crisis.
This agreement marks the next chapter for ePM technology.
Customers are still demanding innovation and advancement in clean and efficient combustion.
PM filters are proven solutions, and next-generation PM sensors are required to ensure filters are operating properly for the life of the vehicle.
The challenge is that operation beyond strict 'emergency' conditions requires compliance with more rigorous Tier-4 emissions standards.
Two ventures were swallowed by industry heavyweights: InterWorks by Sanmina, and the LM-1 line by AutoMeter.
The LM-1 air/fuel ratio meter became a staple in tuner garages - the kind of tool enthusiasts still swap stories about online.
The electrostatic design works across diesel, gasoline, hybrid, natural gas, and multi-fuel engines.
EmiSense built research ties with UC Riverside and Southwest Research Institute to validate the work.
After three companies, EmiSense is the one he has kept - still in the CEO chair as production nears.
The whole run traces back to a UC Berkeley degree in 1990 and a career spent shipping hardware.
Make real-time, low-cost emissions sensing standard - across vehicles, generators, and industrial combustion - and put the PMTrac sensor into serial automotive production by 2028. EmiSense frames it as the founders' desire to positively impact air quality. The instrument is small. The intended footprint is not.