When the U.S. Navy’s fleet of F/A-18F Super Hornet fighter jets is headed for the scrap heap because an essential chip is unavailable, what do they do?

They turn to Ryan Hatcher, the CEO of Phoenix Semiconductor. Hatcher repackages off-the-shelf semiconductors into devices that are virtually identical to the phased-out chips.

“You’re not going to park a US $100 million aircraft, whether it’s a 737 or an F-35, for a $1,000 chip, even if that chip originally cost five bucks,” Hatcher says. “We’re hooking up the supply of components you can get off the shelf today at volume, at low cost, with the demand for parts that are no longer available, but for which there’s tremendous marginal need.”

In the Navy’s case, they turned to Phoenix specifically for a replacement part needed to maintain the jet fighter’s bleed-air control unit that regulates cockpit air pressure and temperature.

Phoenix, founded in 2023 and based in Austin, Texas, connects chips on an interposer (the electrical bridge that connects chips within an electronic package) that the company designs into a package with a pin-out (the metal leads carrying signals from a chip to a printed circuit board) similar to the original component.

“When you drop that part into the socket, it looks identical, indistinguishable from the original,” Hatcher said. “There are no board updates. There are no software updates. There are no firmware updates. It just operates like the original.”

Solving Legacy Chip Obsolescence

Phoenix is aiming for the high-mix, low-volume demand that large chip manufacturers don’t want to touch. As more systems used in aerospace, healthcare, and industry face obsolescence because the legacy chips that run them are no longer available, Ryan sees a “long tail” of demand exceeding his ability to supply.

“There is demand for billions of dollars of legacy chips every year, and much of that demand literally doesn’t have a product in existence anymore,” says Jonathan Bronson, a managing partner at venture capital firm J2 Ventures. “In many of these cases, Phoenix is the only game in town.” Bronson became a board member for Phoenix in June 2025 after earlier investing in the company.

Leading chipmakers like Intel or Taiwan Semiconductor Manufacturing Corp (TSMC) count on large production runs to maintain high capacity utilization, a key measure of profitability. A production run for one chip in a large fab can last up to three years. Retooling a production line for a new chip results in downtime and lower yields that hurt profit.

“No one’s going to pick up the phone for a couple hundred chips,” Hatcher says of big semiconductor foundries like Samsung, where he worked from 2013 to 2019. “The fundamental mismatch is that semiconductor companies go out of business if they have high mix, low volume. Defense companies, more generally, need a high mix of parts, but very low volumes.”

Phoenix and its 15-member team aim to bridge that mismatch, assembling prototypes in their Austin, Texas lab that they later outsource to companies like Mikros, QP Technologies, or TTM for commercial production. In June, the company received an ISO 9001 certification, an internationally recognized standard for quality management systems published by the International Organization for Standardization.

To start, Hatcher likes to find low-power chips originally made for the Internet of Things (IoT) market in chip-scale packages that are nearly the same dimensions as the silicon inside.

“We take a collection of these and then put them onto an interposer, an MCM (multi-chip module),” Hatcher said. “The bottom, the pin-out, is like the original chip. One related challenge that we have faced is to source or fabricate legacy packaging like lead frames or CERDIP (ceramic dual in-line package) housings and adapt those packaging solutions for MCMs instead of single-chip die.”

Defense and Industrial Chip Demand

Defense applications make up the largest share of Phoenix’s business, which also includes medtech, industrial applications, commercial aerospace, and oil and gas extraction. One customer makes audio-processing chips for stadium sound systems worth hundreds of thousands of dollars.

“All of these systems are not high volume, but super-high dollar value,” Hatcher says. “Those are our customers. Customers that are largely ignored by traditional [original equipment manufacturers]. For the OEMs like Texas Instruments or NXP, this obsolescence business is a constant pain in their butt because they’ve got these customers that have these long tails of demand. They’re good customers, but they’re annoying at the end because they (the OEMs) need to flip over to the next generation. We are working with the OEMs to essentially take over production of certain lines of products.”

Hatcher aims to invest in automation tools and facilities to expand production.

“At some point, we’re going to look to develop our own highly optimized manufacturing flows for high-production, low volume, because even your Mikros and TTMs, they’re not semiconductor fabs, but they still have pretty substantial switching costs to go from one product to the next, one board to the next.”