Why could Ariane 4 compete commercially but not Atlas V?

[Pipcard]

Both rockets used hydrogen in the upper stage, different engines and fuel types, and had multiple configurations ("dial-a-rocket") with varying amounts of solid boosters (and for Ariane, liquid boosters as well). Arianespace (European "georeturn") and Lockheed Martin/ULA both used multiple subcontractors. These factors are said to increase costs.

https://forum.nasaspaceflight.com/index.php?topic=38883.msg1449732#msg1449732

In fact, the empirical evidence is opposite your claims.   Rockets with hydrogen upper stages are known for being expensive (Atlas, Delta, H-II, Ariane).  The low cost rockets (Falcon, Soyuz, Proton) do not use hydrogen in the upper stages.

https://forum.nasaspaceflight.com/index.php?topic=38883.msg1450251#msg1450251

Not sure why you think having multiple engine designs is a good thing.  Sure it may wring out the last percentage of "efficiency", but the #1 goal should be cost, which is the efficiency of the entire system.  And multiple engine designs, while maybe individually more efficient, are a drag on overall costs compared to a single engine type system like Falcon Heavy.

2 main things:

1. ULA manufactures very little in-house. They have many contractors (Aerojet, RUAG, Energia, OrbATK, etc) producing major components, and all of those contractors themselves have other subcontractors. Each level of contracting increases costs because they all add their own profit margins, and eventually what should have been a 5 million dollar engine grows to 10 or 15 million after 3 or 4 layers. Theres also the costs of integrating everything,  ULA and all the contractors need to spend money communicating to coordinate any design changes or purchase orders. And each piece has to be separately transported from wherever it was built.

2. Lots of configurations. Atlas V can have 0-5 SRBs, 1 or 2 RL-10s, a 4 or 5 meter diameter fairing (both of which come in several lengths), plus all sorts of minor customizations. This means they have to keep more production lines open (even if a given part will only occasionally fly), its harder to swap rockets between payloads (less schedule flexibility), and any design changes take much more work because they have to be checked against [some huge number] of variants. Delta IV is even worse for this.

SpaceX is the obvious comparison here. They make nearly everything themselves (and what they don't make internally is mostly off the shelf parts), they have only a small number of facilities to transport parts between and coordinate with, and they have only 1 (soon 2, with FH, and even that is >90% common production) rocket configuration  in service

So how could Ariane 4 capture a lot of the launch market (over 50 percent) in the 1990s but Atlas V couldn't compete?

[Eric Hedman]

Ariane 4 was able to reduce costs.  Their dual payload SPELDA (Structure Porteuse Externe de Lancement Double Ariane) fairing allowed them to offer payloads to GTO at significantly less than competitors.  This was one of the biggest reasons.  It was a smart idea.

[libra]

Oh geez... Ariane 6 -bashing is not enough, now they try to apply the "SpaceX paradigm" to old Ariane 4... ("same fuel ! same engines ! mass production !) 

Some reasons

- SPELDA, as mentionned above

- Viking engines were the same on stage 1 and stage 2 (and rugged and cheap enough ISRO still use them)

- extreme modularity: Ariane 40 / 42 / 44 , "L" and "P" pods with either Vikings or solid-fuel, 2 or 4 of them
(40, 42P, 42L, 44LP, 44L) With SPELDA it was possible to fine-tune a booster to the client needs, and Ariane commercials become pretty gifted at that.

- EELVs entire production line was designed around high flights rates that never happened (repeating Titan III ITL of 1963, and Shuttle mistakes made in 1972  - in 1996 - third time is not the charm here).

- Ariane launch infrastructure in Kourou and production line in Europe was carefully optimized for two things
a) 12 flight per year, average
b) the GTO / GEO comsat market

- Ariane 4 - 135 missions - was also a derivative of the 1-2-3 earlier models. Can't remember how many flights they got together, probably in the 180 ballpark. HM-7 didn't changed much and still hasn't on Ariane 5.

- and so Ariane 4 still benefited from "the shuttle mistake = big opportunity" of 1986

- Ariane never was an ICBM and there was no US military on the way (USAF and others can be PITAs at time for commercial companies and NASA)

- Lockheed and Boeing are, well... Lockheed and Boeing. Bright for building aircraft, not-so-bright for commercial space

Note that, because of the Hermes spaceplane, Ariane 5 beginnings were very, very bad. It was an oversized, bloated, inflexible monster : the solid boosters had grown exponentially during design, to try and lift an obese Hermes (they started as P170 in 1985 and ended as P240 a decade later: those numbers correspond to the mass of solid prop in them).

Atlas V was to compete with Ariane V and, truth be told, between 1996 and 2003 Ariane 5 put itself in serious trouble. Ariane 4 hold the line, however, during that time.