The Hydrogen Fallacy


SA 2010 Soccermaniac
Addon Developer
Jun 22, 2008
Reaction score
(For the TL:DR version, scroll down to the bottom of the post.)

It is the position, or opinion of some (such as the writer of this blog post) that liquid hydrogen/oxygen ('hydrolox') rocket engines are a poor choice for first stages of launch vehicles, because they are 'thrust limited' and/or have a lower engine T/W.

However, there is no intrinsic correlation between thrust and propellant combination. If you had the motivation, resources and time, you could theoretically construct a 20 or 30 meganewton hydrolox engine.

The complaint that hydrolox engines have a lower T/W is, however, true. It should be noted that T/W also differs for reasons other than propellant combination (and the resulting physical requirements). It can change due to engine cycle, area ratio, and construction- factors that (generally) do not intrinsically correlate to propellant combination.

T/W however is more of a rocket equation (higher engine mass = higher dry mass) issue than an overall vehicle thrust/weight issue, since the engine makes up only a small part of the overall wet mass- in fact, the dry mass as a whole is quite a small part of the takeoff mass.

Using the RS-68 engine mass figures from PWR (see here) and the Delta IV CBC mass figures from the Space Launch Report website (DIV data sheet here), we see that the RS-68 for less than three percent of the mass of the wet CBC; if its mass were to change drastically, it would certainly not make much of an effect on the overall T/W of the stage.

Most of the wet mass is propellant. This is where differing propellant combinations make the largest effect. Since hydrolox engines have higher ISP than kerosene/oxygen ('kerolox') engines, a hydrolox stage will, by the nature of the rocket equation, require less propellant than a kerolox stage with the same payload and velocity- it will have a lower wet mass.

For a set stage T/W, it will also have a lower thrust engine- simply because a lower thrust level is require than with a kerolox stage.

A comparison can be made between the Delta IV CBC and the Atlas CCB; the sea level thrust of the latter is roughly 30% higher than the former, but so is the total stage mass. In fact, the CBC, at a stage T/W ratio of 1.295, has a higher T/W than the CCB at a T/W of 1.28. In other words, the T/W of the hydrolox stage is (slightly) higher.

Comparing hydrolox upper stages to kerolox lower stages in this regard is unfair, since the upper stages perform a different purpose and thus have a different set of requirements. Upper stages will have a lower T/W than first stages regardless of propellant combination.

Furthermore many of the claims made by the author of the article simply do not make sense. For example, the core stage of an SDLV is not a first stage- it is effectively a ground-lit second stage. Even on the largest version of the Ares 5 (with 6 RS-68 engines and 5.5 segment boosters) the core produced less than 40% of the takeoff thrust.

The Delta IV (as far as I can read) does not perform any throttle-down to avoid overstressing the structure from dynamic pressure. One would surely imagine that Boeing would have the foresight to factor extra gravity losses from throttling into their calculations, if it were necessary to do so.

The figures posted are pretty misleading in that they are to a 500x500km, 64.5 degree orbit rather than the more usually discussed lower altitude, 28.5 degree orbits. Using John Schilling's Launch Vehicle Performance Calculator the performance of the DIVH to such an orbit is roughly verified (with the same calculator, payload to a 200x200km, 28 degree orbit, is given as roughly 24 tons- no mystery throttling required to account for simple physics) however the stated performance the 5,4 and 5,2 variant is under-represented. The Atlas figures are over-represented, seeming more like the payload figures to lower inclination, lower altitude orbits. One wonders what methods (and motives) went on behind these calculations.

So there you have it: hydrolox engines aren't intrinsically limited in thrust, engine T/W does not drastically effect stage T/W, SDLV cores aren't true first stages, and the Delta IV performs no mystery throttling.

In short: there is no intrinsic physical disadvantage to a hydrolox first stage. Potential economic and/or technical disadvantages are very relevant, but an entirely different issue.


T.Neo :tiphat:


Jan 8, 2010
Reaction score
Nice to read, I was searching info about it on google and found your entrance T.Neo :)