Amazon Lost 40 Satellites in Ten Minutes — and the Launch Calendar Is the Real Casualty

On Sunday at 12:40 p.m. Eastern, United Launch Alliance's Vulcan VC6L lifted off from Cape Canaveral carrying 40 Amazon Leo broadband satellites — the largest Leo payload Amazon has ever attempted. About ten minutes later the mission was gone: ULA confirmed from flight data that the rocket suffered an anomaly and none of the satellites reached orbit (Xinhua, 15–16 Nov). The question for $Amazon.com(AMZN)$ is not the hardware write-off. It is whether Amazon can still hit the deployment milestones behind its 2026 service launch — because the constellation race against Starlink is won in launch slots, not press releases.

Figure 1 — the LV-01 loss in four numbers. Sources: Xinhua/ULA statement, 15–16 Nov 2026; NASASpaceflight.com; Orbital Radar, 14 Nov 2026.

## Ten minutes from nominal to nothing

Vulcan, the 65-metre rocket of ULA — the joint venture of $Boeing(BA)$ and $Lockheed Martin(LMT)$ — was returning from a nine-month grounding after nozzle problems on its GEM-63XL solid boosters: the same failure mode appeared on the second flight in October 2024, 37 seconds after liftoff, and again on the USSF-87 mission in February 2026, per NASASpaceflight.com. Sunday's LV-01 was the fifth Vulcan flight and the first since the fixes — the most powerful VC6L configuration, six solids and two Blue Origin BE-4 methane engines generating 17.2 meganewtons of thrust.

The ascent looked clean until the handover. Reporting compiled by ifeng Tech and NASASpaceflight.com says the first stage burned longer than planned and separated late, and the anomaly struck during Centaur V upper-stage ignition — the phase meant to release 40 satellites over 21 separation events into a 450 km orbit inclined 51.9 degrees. ULA says there is not yet enough data to judge the cause (Xinhua). Every satellite was destroyed.

## The race is measured in working satellites

Before LV-01, Amazon Leo — the renamed Project Kuiper — had placed 396 satellites in orbit across 14 production missions since full-scale deployment began in April 2025, per Orbital Radar tracking data. Starlink operates 9,821 working satellites as of 14 November; OneWeb holds roughly 650, per tracking estimates cited by GCN. Amazon runs the third-largest constellation in orbit, which sounds respectable until you do the arithmetic: Leo sits at roughly 4% of Starlink's operational fleet (derived).

Figure 2 — operational constellations compared, mid-November 2026. Sources: Orbital Radar tracking data, 14 Nov 2026; GCN, 13 Nov 2026.

The timing compounds the damage. Amazon's previous Leo launch went on 2 July, and the constellation gained zero satellites in the four and a half months since, per Orbital Radar and The Space Wiki. LV-01 was supposed to break that drought with a record payload; instead the fleet is frozen at exactly the number it had before the rocket flew — the least welcome status quo for a company targeting initial consumer service before the end of 2026.

## Five suppliers, one working engine chain

Amazon signed what GCN calls the largest commercial launch procurement in history — 80-plus launches across five rocket families. Vulcan alone was booked for 38 missions carrying more than 1,500 first-generation satellites, per Amazon's announcement relayed by ifeng Tech. But every lane now carries a 2026 scar. Vulcan failed on its first Leo mission. Blue Origin's New Glenn suffered a pad heat-fire on 28 May traced to a BE-4 main oxygen valve — the same engine family Vulcan burns — per ifeng Tech. Ariane 6 flies next no earlier than 23 November, after engine issues ESA says are resolved. And the only provider with demonstrated reliability at pace, SpaceX's Falcon 9, belongs to Amazon's direct competitor, which counts 13.6 million Starlink consumer subscribers as of 30 September (AOL). Every Falcon 9 slot Amazon buys is capacity its rival monetizes twice.

Amazon Leo's enterprise preview, running since November 2025, has named customers including JetBlue, Delta, Crane Worldwide Logistics and Hunt Energy, per NASASpaceflight.com. Those contracts wait on a network that cannot exist at scale without cadence — the thing Sunday's failure removed.

Figure 3 — the four remaining launch families and their 2026 status. Sources: NASASpaceflight.com; ifeng Tech, 16 Nov 2026; The Space Wiki; GCN, 13 Nov 2026.

## The FCC clock did not stop

Amazon's deployment math is anchored by regulators, and the milestones were already missed once. The FCC's July 2020 licence required 1,616 satellites operating by 30 July 2026. Amazon missed it — 396 launched is roughly a quarter of the interim bar — and kept the licence only through a conditional waiver granted 5 June 2026, citing an investment of more than $10 billion (The Space Wiki). All 3,232 first-generation satellites are now due by 30 July 2029, and satellites deployed after the missed milestone hold reduced spectrum priority until 30 March 2028 or until half the constellation is in service.

Figure 4 — cumulative Leo deployment against its own service and FCC milestones. Sources: Amazon FCC filing, 30 Jan 2026, via The Space Wiki; Orbital Radar, 14 Nov 2026; NASASpaceflight.com.

The service promise stands: Amazon says initial service begins before the end of 2026 in the US, UK, France, Germany and Canada, per its LV-01 announcement via GCN. NASASpaceflight.com reports Leo needs 578 satellites deployed to a 630 km orbit to start initial regional service — 182 more than are in orbit today (derived). LV-01's 40 satellites represented about 22% of that remaining gap (derived). One rocket carried a fifth of the distance to Amazon's next business milestone.

## Bull Case

1. The deployed fleet is intact: all 396 satellites in orbit are unaffected — a launch-slot problem, not a satellite-design problem.

2. Two independent lanes survive: Ariane 6 flies no earlier than 23 November with 36 Leo satellites; Falcon 9 remains proven at pace.

3. The waiver holds: the 30 July 2029 deadline is more than two and a half years out.

4. Demand is signed: JetBlue and Delta are named in Amazon's own materials; 80-plus launches remain contracted.

5. Recovery hardware exists: LV-02 is already stacked in a separate building (Amazon via ifeng Tech).

## Bear Case

Figure 5 — what the 40 lost satellites represent at each milestone; percentages derived from sourced counts. Sources: Orbital Radar, 14 Nov 2026; NASASpaceflight.com; FCC order via The Space Wiki.

1. The failure mode is unknown, and unknown anomalies ground rockets for months — Vulcan just spent nine months grounded over an already-redesigned nozzle.

2. The BE-4 chain is suspect in both directions: if the fault reaches the first stage, New Glenn — burned on a BE-4 valve in May — inherits the suspicion, and Amazon loses two lanes at once.

3. One rocket's payload equalled 10.1% of everything Amazon has ever deployed (derived), lost on the day the workhorse returned to flight.

4. The calendar is the binding constraint: 182 satellites to initial service, 1,220 to the missed interim bar, 2,836 to the licensed fleet (all derived) — while the sole high-cadence provider charges a strategic price.

5. Every week of grounding pushes Amazon deeper into reduced spectrum priority and closer to the 30 March 2028 condition.

## Strategic Slam

$AMZN(Amazon)$: the hardware loss is noise; the schedule risk is the signal. A Centaur-specific fault isolated within weeks keeps the end-2026 window credible, with VA271 and LV-02 carrying the load. A BE-4 first-stage fault stalls both Vulcan and New Glenn and slips initial service into 2027 — against a date Amazon announced itself.

$BOE(Boeing)$ and $LMT(Lockheed Martin)$: ULA's flagship now has five flights, three anomaly events and its first total mission failure in 20 years, per ifeng Tech. A fast fault isolation protects the national-security pipeline both parents depend on; a slow one invites customers to re-audit the GEM-63XL and Centaur stack. No price targets — the fundamentals decide.

## Catalyst Calendar

Figure 6 — five dated events now carrying Amazon's satellite bet. Sources: ESA/Arianespace via The Space Wiki; ifeng Tech, 16 Nov 2026; Amazon announcement via GCN; FCC order via The Space Wiki.

  • 23 Nov 2026 — Ariane 6 VA271, 36 Leo satellites, no earlier than.

  • Weeks after LV-01, expected — Vulcan LV-02, hardware stacked; timing depends on the findings.

  • 31 Dec 2026 — Amazon’s initial-service target for five countries.

  • 30 Mar 2028 — reduced spectrum priority lapses unless half the fleet is in service sooner.

  • 30 Jul 2029 — FCC deadline for all 3,232 first-generation satellites.

## What the investigation decides

The fault's address decides everything. An upper-stage problem points at Centaur V — a ULA-internal fix, a return in weeks, a calendar that mostly survives. A first-stage problem implicates the booster and BE-4 chain, would shadow New Glenn, and hands Falcon 9 a larger share of Amazon's remaining launches — and Starlink's owner the leverage that comes with it.

Figure 7 — fault location versus return speed, and what each combination means for the calendar. Analysis based on the ULA statement (Xinhua, 15–16 Nov 2026) and reporting cited in Sources.

## Verdict

The satellites were replaceable; the slots are not. Amazon bought five launch families to avoid exactly this single-vehicle risk, and 2026 has burned three of them. Leo's value now moves with ULA's investigation timeline, not with Amazon's announcements. Watch VA271 on 23 November: a clean flight caps the damage; a delay confirms the supply chain is the bottleneck. Monday prices the headline; the calendar prices the story.

## Key Takeaways

1. Vulcan failed about 10 minutes after liftoff on 15 November, destroying all 40 Amazon Leo satellites on its first mission for the company (Xinhua/ULA).

2. The 38-mission Vulcan contract was to carry over 1,500 first-generation satellites (Amazon via ifeng Tech).

3. Amazon Leo has 396 satellites deployed versus 9,821 operational Starlink satellites (Orbital Radar, 14 Nov).

4. The FCC interim milestone of 1,616 satellites was missed on 30 July 2026; all 3,232 are now due by 30 July 2029 (The Space Wiki).

5. Initial regional service needs 578 satellites — 182 more than in orbit — with Amazon still targeting end-2026 (NASASpaceflight.com; Amazon via GCN).

Can a constellation that just lost a fifth of its remaining runway still beat its own December deadline — or does the schedule now belong to ULA's engineers?

Tickers: $AMZN(Amazon)$ $BOE(Boeing)$ $LMT(Lockheed Martin)$ $JetBlue Airways(JBLU)$

Sources: Xinhua, 15–16 Nov 2026; NASASpaceflight.com (mission profile, GEM-63XL history, 578-satellite threshold, thrust figures); Orbital Radar, 14 Nov 2026 (396 satellites/14 missions, Starlink 9,821, FCC waiver); GCN, 13 Nov 2026 (80-plus launch procurement, largest-payload claim, service countries, OneWeb estimate); ifeng Tech, 16 Nov 2026 (38-mission contract, 1,500-plus satellites, BE-4 valve fault, LV-02 status, first total failure in 20 years); The Space Wiki (FCC 2020 order, waiver terms, VA271); AOL (Starlink subscribers).

Disclaimer: For information and discussion only, based on the public reporting cited above; not investment advice. Derived percentages and counts are arithmetic on the cited figures and are labeled as such.

# 💰Stocks to watch today?(16 November)

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Report

Comment(4)

  • Top
  • Latest
  • breezyk
    ·11-16 11:02
    TOP
    Launch loss hurts headlines, not my long DCF. Kuiper still matters more as Amazon's next growth curve than one blown slot
    Reply
    Report
    Fold Replies
    • Xaddy_Analyst: 
      Fair point — the 396 satellites already in orbit are untouched, so the fleet itself is fine. The harder question is the calendar: 182 more satellites are needed before initial service, and every slipped slot now eats into that runway.
      11-16 17:47
      Reply
      Report
  • pizzix
    ·11-16 11:02
    TOP
    Launch insurance helps less than a slipped manifest. ULA execution risk just got priced in harder
    Reply
    Report
    Fold Replies
    • Xaddy_Analyst: 
      Exactly — insurance covers the hardware, not the manifest. With Ariane 6 due no earlier than 23 November and Vulcan's return depending on the investigation, the schedule is where the risk now sits.
      11-16 17:48
      Reply
      Report