Product1 distinct publisher2 min readPublished
Obsolescence comes in two kinds: the part is gone, or the knowledge to use it is gone. The second kind can be reversed by studying hardware you already own; a shortage report treats both kinds the same way.
The Product Desk · Product desk
Compiled by The Product DeskSomething wrong?How this is made
Stock records answer whether the part is on hand. The aircraft asks whether it is configured, and this case shows what happens when those two answers diverge [3][18]. There was no standard programming procedure to fall back on, so the work began by probing the components and characterizing their functions until the programming requirements could be reconstructed [5][6]. Phoenix then settled on an approach, produced devices and verified them before shipping [10]. The under-30-days figure is the company's own account [4], though the report has NAVAIR receiving programmed parts and supporting documentation inside a month [9].
The account is silent on the cost of either path and on what NAVAIR's fallback would have been [15]. That matters for how the case gets used. Read as proof that reverse engineering beats requalification, it is being pushed past its evidence. What the evidence supports is narrower: a device that sits inside the pilot breathing path [17] was recovered from inventory rather than replaced from a catalogue [1].
What landed was a test candidate. Phoenix's scope closed with programmed devices, documentation and its own verification, and NAVAIR's service testing sits after that line [9][10][11][16]. Anyone budgeting from the 30 days should book it as the characterization leg, not the clearance.
This is for the program office that has been told a part is obsolete and has not asked which kind. Put the physical part on one axis and the configuration knowledge on the other. When both the part and the knowledge are available, that's an ordinary buy. When both are gone, you're in redesign territory, the case every sustainment budget already knows how to fund. If the part is gone but the knowledge survives, it becomes a sourcing exercise: a second source, aftermarket stock, or a fab willing to run an old process. The F/A-18 quadrant is different: the part is available but the knowledge is gone [3], and it is the quadrant that gets misfiled, because a shortage report shows the same red cell either way.
The tradeoff sits in what you are trusting. A reconstructed programming path is inference from observed behavior rather than a supplier specification [5][6], which is where aging fleets keep ending up as original suppliers leave the market and the documentation goes with them [12][13]. Deciding that inference is good enough for a bleed air control unit is a test decision, and per the account that decision was still open when the boxes arrived [11][16].
Ranked by verification strength, evidence, and original report placement.
Phoenix Semiconductor engineers helped NAVAIR recover a supply of unusable legacy chips for a safety-critical system on the F/A-18 Super Hornet.
The devices support the aircraft's Bleed Air Leak Detection Control Unit, or BALDCU, which controls bleed air delivered to pilots through their oxygen masks.
NAVAIR had replacement chips available but could not put them into service because the devices lacked the programming needed for the aircraft system, leaving the Navy with inventory it could not readily use.
Phoenix said it solved the problem in less than 30 days.
The team did not have a standard programming procedure for the devices, so engineers treated the chips as a black box and studied their behavior.
Probing the components and characterizing their functions produced results that helped engineers reconstruct the programming requirements needed for the BALDCU.
Distinct publishers with included, body-backed reporting in this cluster.
Follow any of these and your For You feed starts watching them — no settings page required.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
One desk, one supplier's account
Trace any fact here back and it lands in the same place: Phoenix's description of its own save, written up by Interesting Engineering. The single outside voice is a deputy assistant program manager at PMA-265 who is never named and is quoted in two fragments. There is no part number, no device count, no contract reference, no NAVAIR statement, and no document from the delivery package that anyone outside the transaction has seen. The technical narrative is coherent, which is not the same as being checked.
One delivery, clearance pending
Something real happened and it is narrow: reprogrammed devices plus documentation reached one program office. Nothing indicates how many units, whether any went into an aircraft, or whether the recovered programming survived NAVAIR's evaluation. By the story's own last sentence, the deliverable is a process for the Navy to assess — a first customer at the front door, not a fielded fix.
The clock stops before the finish line
"Revive" and "30 days" describe delivery; qualification for a system that feeds a pilot's oxygen mask is the part still outstanding, and the story admits as much in its closing line. Add a headline that reaches for the jet's 1,200-mph top speed — a number with no bearing on programming a control-unit chip — and the packaging runs ahead of what was demonstrated. The engineering claim itself is modest and plausible; the overstatement is in how much of the problem the month is said to have closed.
The subject supplied the story
This news exists because a supplier wants credit for a difficult save, and it is a co-founder carrying the title chief growth officer who delivers the closing quote. The government quote costs the program office nothing and reflects well on it too. Neither party has any reason to volunteer the unit count, the price, or how much of the qualification burden remains — and none of the three appears.
Thin sourcing, honest gaps
We are reading one account, so certainty is limited by construction. But it is internally consistent, it attributes its speed claim to the company rather than asserting it, and it says outright that testing lies ahead — the blanks are visible rather than papered over. That makes our read of what is and is not established steadier than the source count alone would suggest.