This news item quotes Elon Musk as saying

it would make sense for the government to rely on ULA's Delta 4 and the SpaceX Falcon 9, phasing out the venerable Atlas 5.

"You don't need both of those rocket families," he said. "I think it would make sense for the long-term security interests of the country to probably phase out the Atlas 5, which depends on the Russian engine, and have ULA operate the Delta family,

I understand that since then ULA have followed quite a different approach, that of keeping the Atlas 5 and the Delta 4 Heavy and just retiring the Delta 4 medium, which happens to use solid boosters not used on the heavy version.

Its very easy to get bogged down in corporate mud-slinging regarding the strategies of both ULA and SpaceX to engage with government clients. That said, it does seem sensible from an assurance point of view to keep the Delta 4 medium instead of the Atlas 5, given the latter's dependence on the RD-180 engine.

Does anyone know of any technical reasons, or technical and manufacturing/economic drivers that explain ULA's approach. E.g. the solid booster topic may be relevant? Please try to keep it to rationale that is

  • referenced

  • avoid dipping into speculation about their business strategy, tax payer value for money etc

  • $\begingroup$ Bear in mind that the longer term strategy isn't D4H + Atlas 5, but D4H + Vulcan. $\endgroup$ Jan 31, 2016 at 21:16
  • $\begingroup$ Right, but I find it curious that, on purely technical grounds, one would want two completely different technical roots, unless that was part of a diversification initiative, they are a private company not the government customer. Similarly, the Arianespace A5/Soyuz/Vega situation seems expensive to maintain. $\endgroup$
    – Puffin
    Feb 1, 2016 at 13:36
  • $\begingroup$ @RussellBorogove: Isn't the Vulcan supposed to replace both the Atlas V and the Δ4 Heavy? $\endgroup$
    – Vikki
    Jun 20, 2019 at 0:05
  • $\begingroup$ @Sean I don't actually know. $\endgroup$ Jun 20, 2019 at 2:13

1 Answer 1


The design of the Delta 4 is such that each CCM (Common Core Module) has fittings to support thrust through the strap on mount points. It has to support 2 other CCMs attached for the Delta Heavy model, and then lower mount points for the various combinations of SRBs.

For some reason I do not fully understand the Delta 4 pays a heavier price in terms of both manufacture (more mount points required) and payload penalty for these mount points.

The Atlas 5 of course has mount points for up to 5 SRBs and each core has support for 0-5 without the same penalty that the Delta 4 seems to pay.

Ignoring even that, with Delta 4 being a LOH/LOX vehicle, it is more expensive to build than a Ker/Lox vehicle. Hydrogen is much colder than LOX, causes metal to become brittle, and it leaks much more (being a significantly smaller molecule) and whatnot. These are all surmountable problems (witness the fact the D-4/D-4H work, the Shuttle worked, all other LH vehicles in history have worked) but they add cost.

Thus the Delta 4 from a pure cost perspective is more expensive than the Atlas 5. (Now I get into the speculation you wished to avoid) It seems like ULA actually made a decision based on cost for once, in the EELV program. As unlikely as that seems based on past history. They could not dump the Delta 4 Heavy since some occasional NRO payloads require that size, so we end up with this odd decision. Cancel the D4 Medium but keep all the costs associated with making cores for it.

  • $\begingroup$ Thanks, I think my take-away is the LH seems a plausible cost driver. Having thought a bit more about your first suggestion re. mount points, is the idea that the future core design would be changed so as to eliminate the smaller mount points? $\endgroup$
    – Puffin
    Feb 1, 2016 at 13:39
  • $\begingroup$ Problem is, payload requirements are such that for a single stick you really often need a SRB. But i doubt they are going to spend much money on redesign just for the rare D-4H launches at this point. $\endgroup$
    – geoffc
    Feb 1, 2016 at 14:15
  • $\begingroup$ That seems to suggest that the mount points are not themselves such an economic driver. $\endgroup$
    – Puffin
    Feb 1, 2016 at 15:01
  • 1
    $\begingroup$ I think they are just not going to put any money into changing the CBCs. For a non SRB launch, the mount points gives a payload penalty. for the D4H launches a payload penalty, but who cares? So the cheapest D4 (no SRBs) has lower payload so pretty much always needs an SRB which raises costs of launch. $\endgroup$
    – geoffc
    Feb 1, 2016 at 15:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.