Shielding engines from aerobraking

[Fraktal]

Just yesterday, I launched my first mission to Duna which arrived safely, but ended up using quite a bit more fuel (more than half) than I anticipated. This is due to me having decided to experiment with an engine design I never tried before.

This is the engine assembly of the orbiter: four Terriers radially attached with a roughly 26° rotation so that their thrust vectors narrowly avoid the edge of the heatshield while still being behind said heatshield and thus fully protected during aerobraking. Yesterday I successfully tested this design out with an aerocapture at Duna, which used up slightly more than 1/8 of the heatshield's ablator to insert the craft into a 61 km orbit, with nothing on the craft ever showing overheat bars. I also wish to future-proof this design eventually to the point where it could survive aerocapture at Eve at 120% reentry heating.

But anyway. Due to the offset thrust of the engines, I found out the hard way that KER's option of calculating delta-V with vectored values does not apply to the KER readouts for the duration and timing of maneuver burns: my trans-Duna injection took quite a bit more time than indicated, missed the planet by more than 300 million kilometers and used up more than half of the orbiter's fuel. If it weren't for the orbiter having been designed for aerocapture, I'd have been stuck at Duna due to using up too much fuel for propulsive deceleration.

I thought that the few hundred m/s lost from angling the engines is acceptable, but this is clearly not the case. So my question is, which of these two options is better for increasing efficiency by way of lowering the thrust offset angle?

1. Offset the Terriers outwards, partially exposing them to the airstream but allowing them to operate at a lower offset angle.
2. Add more girders between the engines and the heatshield to increase their distance from each other, increasing weight by 125kg per girder (for reference, the whole orbiter-plus-lander assembly is exactly 61 tons when fully fueled) but allowing me to lower engine offset while still maintaining full heat protection. The heatshield is already autostrutted, so it shouldn't have stability problems. One extra girder already allows me to lower the offset to 14-15° without blasting the back of the heatshield.

Edited October 27, 2020 by Fraktal

[bewing]

A 26 degree angle loses you 11% of your efficiency. If that used up more than half your fuel, then you wasted 40% of your fuel on maneuver mismanagement. It had nothing to do with your engine or design or offsets or angles. Doing a few hand calculations to "fix" the burn start time and managing burn duration by eye will entirely fix your problem.

Also, if you used up only 1/8th of your heatshield ablative, then you either did not need ablative at all, or you installed 85% too much, and you can reduce your payload mass by trimming your ablative by a lot. Which will get you even more fuel remaining at Duna orbit insertion.

 

[Fraktal]

2 minutes ago, bewing said:

If that used up more than half your fuel, then you wasted 40% of your fuel on maneuver mismanagement.

Upon launch, the orbiter had about 2516 m/s. After aerocapture and circularization into a 61 km equatorial orbit, it has 1089 m/s left with the weight of a fully fueled lander. After separation, the lander has 2087 m/s; a round trip to Duna with parachute descent is going to use up nearly all of that, so the orbiter's delta-V is going to rise from the mass decrease and I'm planning to transfer the lander's remaining fuel over to the orbiter for trans-Kerbin injection (the lander has two Thuds, so the orbiter will likely have higher Isp). If I'm lucky, I'll even have enough fuel for circularization around Kerbin.

12 minutes ago, bewing said:

Also, if you used up only 1/8th of your heatshield ablative, then you either did not need ablative at all, or you installed 85% too much, and you can reduce your payload mass by trimming your ablative by a lot. Which will get you even more fuel remaining at Duna orbit insertion.

I do trim my ablator down for single-use crafts, but this orbiter is designed to be 100% reusable, with the heatshield attached via a docking port so that it can be jettisoned and replaced at Kerbin once it's used up. That and I have no idea how much ablator it's going to take to aerocapture around Kerbin during the return trip; if it's going to be similar to what it took to aerobrake around Duna, I should be able to squeeze out one more round trip before it needs replacement.

[king of nowhere]

1 hour ago, Fraktal said:

 

I do trim my ablator down for single-use crafts, but this orbiter is designed to be 100% reusable, with the heatshield attached via a docking port so that it can be jettisoned and replaced at Kerbin once it's used up. That and I have no idea how much ablator it's going to take to aerocapture around Kerbin during the return trip; if it's going to be similar to what it took to aerobrake around Duna, I should be able to squeeze out one more round trip before it needs replacement.

i was always able to do a full aerocapture on duna without any kind of ablator, without ever coming even close to taking damage. including an intercept from dres with a relatively high speed. duna is surprisingly forgiving in that regard.

kerbin, not so much. you'll definitely need the heat shield for return aerocapture.

[Lt_Duckweed]

8 hours ago, Fraktal said:

Upon launch, the orbiter had about 2516 m/s. After aerocapture and circularization into a 61 km equatorial orbit, it has 1089 m/s left with the weight of a fully fueled lander. After separation, the lander has 2087 m/s; a round trip to Duna with parachute descent is going to use up nearly all of that, so the orbiter's delta-V is going to rise from the mass decrease and I'm planning to transfer the lander's remaining fuel over to the orbiter for trans-Kerbin injection (the lander has two Thuds, so the orbiter will likely have higher Isp). If I'm lucky, I'll even have enough fuel for circularization around Kerbin.

I do trim my ablator down for single-use crafts, but this orbiter is designed to be 100% reusable, with the heatshield attached via a docking port so that it can be jettisoned and replaced at Kerbin once it's used up. That and I have no idea how much ablator it's going to take to aerocapture around Kerbin during the return trip; if it's going to be similar to what it took to aerobrake around Duna, I should be able to squeeze out one more round trip before it needs replacement.

You can aerocapture everywhere (including Eve) without using any ablator, as heatshields are already insanely heat resistant on their own.

[Fierce Wolf]

Using robot parts to hide the engines under the shield?

[Fraktal]

I don't have Breaking Ground.

[Fulgora]

I'd recommend to ditch the Heatshield approach completely in favor of wings, because

• No Ablator that can deplete
• Greater surface area (except when comparing against the 10m heatshield)
• Control surfaces allow for much greater control when breaking (point prograde or up depending on how much you want to brake)
• Add some wheels and you can land on the runway as well (with some practice )
• Its lighter and requires no engine offset (see last picture)

 

Spoiler

The engine bay shields very well even when pointing almost straight up.

Duna return aero capture at Kerbin with a 50km Periapsis is easily doable (everything attached to the engine bay is shielded from the heat).

Mass of the craft is 45t after Duna return.

Edited November 1, 2020 by Fulgora
Added VAB screenshot

[Fraktal]

I don't have the proper parts for that yet and I'm deliberately trying to use tech 6 and lower parts.

[Mikenike]

34 minutes ago, Fraktal said:

I don't have the proper parts for that yet and I'm deliberately trying to use tech 6 and lower parts.

Bite the bullet, attach the extra girder, and lower the angle, more weight but more efficient overall.

[Fraktal]

Actually, I'm currently experimenting with a slightly lighter design where I drop one of the Terriers and replace the other three's offseted girders with a stack tricoupler. Minor improvement, yes, but still gave me a couple dozen m/s more dV. I also offseted the engines outwards to reduce the rotation angle, but still need to find the exact angle at which the thrust vector just barely avoids the heatshield. The heatshield still uses the same girdered configuration, just radially attached onto the bottom of the tricoupler and offseted into the exact center.

Edited November 5, 2020 by Fraktal