Hydralox with Oxygen Thrust-Augmented Nozzle?

[acksed]

We mostly know that the problem with hydrogen is the low density, requiring large tanks. Liquid oxygen is denser by far. It's even denser than kerosene or liquid methane, to the point that Starship's proposed propellant transfer is stated to be mostly LOX.

Compromises between tank size, nozzle size and thrust, among others, requires a more oxidiser-rich mixture ratio - I think around 5.5 to 6 O:F - when the greatest specific impulse of hydralox is found when it's more fuel-rich and with a vacuum-optimised nozzle.

Here's the proposal: a thrust-augmented nozzle can be over-expanded at sea level as long as fuel is injected downstream of the combustion chamber and the resultant pressure is equal or higher than atmospheric.

Injecting oxygen into a more fuel-rich hydrolox combustion stream, say 4 or even 3.5, to create a normal or even leaner O:F ratio of 6, 7, even 8 inside the nozzle would increase the combustion temperature, the thrust and maybe even the thrust-to-weight ratio of the whole rocket in the boost phase, as losing more oxygen at the start would reduce its mass more quickly.

I'm envisioning a sort of 1.5 stage to orbit, except with smaller drop tanks of oxygen.

A higher nozzle temperature would require more cooling, but also supply more energy to run a H2 expander cycle. (Yes, I know about the size limitations.) Oxygen cooling might be necessary on top of that, and I'm not sure you could run the pumps for the LOX off expander cycle, but Launcher's E-2 engine uses O2 for cooling at least.

Nozzle design would be tricky, no lie. It'd have to be large enough to provide good vacuum performance, small enough to fit into the rocket, resistant to extra hot oxygen... maybe mix oxygen and hydrogen cooling... and have injectors for the hypersonic nozzle flow.

Could it be simpler to design an engine to accept a greater range of mixture ratios? Yes. Could an extendable skirt do the same job? Yes... but they generally don't come back. This is a way to have a vacuum nozzle from the ground up, so to speak.

So has this ever been tried? Am I off-base entirely or is there something there?

[Robotbeat]

Yes. I like this idea.

[soyuzu]

Variable mixture ratio Hydrolox was a pretty common SSTO concept, I think.

Always liked the idea of dual mixture ratio hydrolox for any one crazy enough to try reusable SSTO. Start off running LOX rich, switch to fuel rich at the equivalent of staging. Bulk density similar to methlox, but with full flow staged combustion ISP of 360 at sea level LOX rich going to 460 fuel rich ISP in the vacuum. LOX rich tan mentioned by JonGoff on the other thread is an interesting variation, lower pressure injection of the extra TAN LOX reducing the delta in pumping requirements between fuel rich and LOX rich modes.  Would still pose some interesting pump design problems.

Actually you can make it very oxygen rich at the beginning, even to 12:1, to reduce chamber temperature. Basically a tri-propellant engine with two propellant.

[Robotbeat]

12:1?
“You are like a little baby!”
I’d go 26:1 for O:F ratio for a first stage for hydrolox. You want super high bulk density and low Isp in the beginning of flight, in this case 287s Isp. 26:1 O:F would give you bulk density nearly that of methalox.

[Bob Shaw]

What sort of engine materials could withstand white-hot O2 going through and past them? Would ablative nozzles be the answer?

[Robotbeat]

For an RLV, there’s another great argument for being oxygen-rich:

liquid oxygen is like 2 orders of magnitude cheaper than liquid hydrogen.

1 ton of LOx is about $100. 1 ton of LH2 is about $10k, and an absolute pain in the rear to transport and store. So the headache of using hydrolox can be reduced somewhat by being extremely LOx-rich in the first stage (or earlier part of the flight if you’re doing SSTO or whatever like in this thread).

With a 5.5:1 O:F ratio, 1000 tons of propellant costs $1.6M. For a 26:1 ratio, 1000 tons of propellant costs $470,000, a factor of 3.5 less. Tank and engine mass scale typically more closely to propellant volume than to mass, so if we assume the dry stage mass is 0.05kg per liter of bulk propellant, then for a first stage with a sea level Isp of 2.8km/s for a 26:1 O:F and 3.6km/s for a 5.5:1 O:F, and a required stage delta-v of 3.5km/s, then although the lower Isp option requires 39% more propellant mass, it requires about 35% LESS dry mass (which translates to lower fabrication costs… also total volume is lower just due to how much denser oxygen is than hydrogen) and its propellant costs are a factor of 2.5x lower.

So there’s pretty good argument in favor of LOx-rich lower altitude or first stage flight.

[acksed]

12:1?
“You are like a little baby!”
I’d go 26:1 for O:F ratio for a first stage for hydrolox. You want super high bulk density and low Isp in the beginning of flight, in this case 287s Isp. 26:1 O:F would give you bulk density nearly that of methalox.

You can do that? It might be like an oxygen-rich flamethrower at the start.

What sort of engine materials could withstand white-hot O2 going through and past them? Would ablative nozzles be the answer?

I think ablative nozzles are a single-use thing. You might need to line the nozzle with whatever SpaceX uses in their Raptor turbopumps, though I think the lower density of the exhaust and maybe some hydrogen curtain cooling could compensate.

[Robotbeat]

If you add *enough* oxygen, it actually reduces the temperature enough that material requirements aren’t as insane. Moderately oxygen-rich might be the worst place to be in terms of material requirements as the temperatures are very high but you’re also very oxygen-rich.

[Jim]

though I think the lower density of the exhaust and maybe some hydrogen curtain cooling could compensate.

Might as well then use the standard 6:1 ratio

[Jim]

12:1?

I’d go 26:1 for O:F ratio for a first stage for hydrolox.

That doesn't burn.  Need higher than 4% H2

[Robotbeat]

12:1?

I’d go 26:1 for O:F ratio for a first stage for hydrolox.

That doesn't burn.  Need higher than 4% H2

Correct… EXCEPT: That’s molar concentration (roughly, volume). 4 moles of molecular hydrogen is 8 grams. 96 moles of oxygen is about 3072 grams. That would be a 384:1 O:F ratio *BY MASS*, so we’re still well within the combustibility limits (you can double check this by adding up the chemical energy of the hydrogen plus the thermal heat needed to turn liquid hydrogen and liquid oxygen into gaseous hydrogen oxygen and hydrogen, and then high temperature gaseous water vapor (steam) and gaseous oxygen… it checks out and there’s plenty of energy left over).

From: http://conference.ing.unipi.it/ichs2005/Papers/120001.pdf

(Which also shows that at high pressure, the lower limit is actually ~6% molar ratio of hydrogen in oxygen… but again the 26 O:F mass ratio implies about a 38% molar percentage of hydrogen to total propellant… stoichiometric would be an 8:1 O:F, or a 67% molar ratio of hydrogen to total propellant, and a standard 6:1 implies 72% molar ratio of hydrogen to total propellant.)

[acksed]

So it can work.

What kinds of chamber and nozzle pressures in this hypothetical engine are we talking here? Thy olde RL10 has a modest 42-44 bar (no idea about nozzle pressure), while Aerojet's tests just went up to 34.47 bar for the chamber and an estimated 13.79 bar for the TAN.

On the other end of the scale is the RD-701: up to 294 bar in boost mode and a still-high 124 bar in sustain. Nozzle pressures unknown.

Lower pressure and high thrust is a good deal, but high chamber pressures seem to help wring out the last drop of performance.

Where's the sweet spot? What can we get away with if we want a 1.5 stage rocket with hydrolox engines packing oxygen-boosted TAN?

Side note: I found this article detailing lightweight liquid hydrogen tanks for fuel-cell aircraft. 67kg tank carrying 150kg H2 sounds good, but the devil is in the details. Composite tanks need a jig; large composite tanks need equally large jigs, and a defect means you have to scrap the whole thing. It might not stand up to rocket flight conditions either. Still, I'm interested.

The company itself says its tanks are also suitable for methane or oxygen. We would need lightweight drop tanks...

[Jim]

So it can work.

No, it can burn.  It doesn't mean the idea is viable.  Afterburning in a nozzle doesn't really help.

[Proponent]

At very high O/F ratios, what would be the power source for the pumps?

[redneck]

Using Kero/LOX for a thrust augmented nozzle alleviates some of these concerns. There are some techniques to get a very low L* in the augmented attachment. High thrust at lift off with the augmentation reducing or eliminating the Pa/Pc losses. High expansion ratio later when the augmentation gets out of the way for a high expansion ratio nozzle.

For first stage use launching in atmosphere. In vacuum, augmentation is a net loss.

[acksed]

At very high O/F ratios, what would be the power source for the pumps?

I don't have good answers for that, all this is just speculation, but it depends on how large I want the engine to be.

If I wanted a thrust-augmented equivalent to RL10, the expander cycle would probably still suffice for the hydrogen side, but for the oxygen side an expander cycle might need need to flow oxygen down the bottom of the nozzle - not that oxygen's a particularly good working fluid, we're taking advantage of the sheer mass flow - and it might additionally take a burner with heat exchanger to augment the heat gathered.

[Robotbeat]

At very high O/F ratios, what would be the power source for the pumps?

Staged combustion would still allow high power, although perhaps not as high chamber pressure as, say, Raptor.

[laszlo]

Watching Jim and Robotbeat debate makes my inner nerd tingle. This is why I paid to join this website. Thanks guys.

[sevenperforce]

At very high O/F ratios, what would be the power source for the pumps?

Staged combustion would still allow high power, although perhaps not as high chamber pressure as, say, Raptor.

At launch, you could use an expander cycle for the hydrogen side and ORSC for the oxygen side, since the amount of LOX you'll be pushing is far higher than the amount of LH2. You could also supplement the ORSC with a LOX-expander boost pump. As the ascent continued, you could throttle down the oxygen-rich preburner gradually, until the LOX-expander boost pump ended up doing the majority of the work.

One neat idea I saw from an old Aerojet patent was to use a preburner without any turbine, simply as a way to extract more heat to operate the expander cycle via a heat exchanger inside the preburner (attached). In their concept it was a fuel-rich preburner operating a single-shaft LH2 expander cycle, but you could imagine a similar arrangement with an oxygen-rich preburner, a turbine, a heat exchanger, and a split LOX-expander. Initially you'd be routing most of your oxygen-rich gas through a turbine, but you'd use a valve to gradually put less through the turbine and more through the heat exchanger, so that the expander turbine power increases while the preburner turbine power decreases.

[Robotbeat]

A 26:1 O:F ratio would imply roughly 3 times as much pumping power for a given chamber pressure (with similar amount of fuel flow). So reduce the chamber pressure to a third, and you'd have sufficient power. Raptor has ~300 bar, Merlin has ~100 bar. So you'd be down near Merlin chamber pressure, maybe BE-4 chamber pressure. Still not terrible at all. But further optimization is likely possible.