Noob needs rocket science lessons.

[ArmchairPhysicist]

So I have a problem. My rockets are too big and over engineered. My munshot rocket is so expensive and complicated that Germany would be proud. It easily runs mun science landings, but it's expensive and ugly. 

 

Heres the rundown on it. Typical command pod and science suite built for kerbin re entry. This is mounted on a mk1 liquid fuel tank that powers my beloved atomic rocket. Under this is a 2.5 meter tank and a poodle engine. Pretty standard lander and return package right? 

Heres where it gets ugly.

under the poodle are two large tanks and a mainsail.

Next we have my Radially mounted vegetable boosters (asparagus style liquid boosters) 

These boosters are four freaking twin boars. I hate using this many but it works.

 

This rockets staging is as follows.

stage one fires all four twin boars and the mainsail. 

Stage two drops the first two twin boars when they run out of fuel. 

Stage three drops the second set of boars and I cut power to my mailsail.

It this point my apoapsis is at about 400 kilometers.

once at max altitude I fire my mailsail and use its fuel to put myself in kerbin orbit and start the mun burn. 

Stage four drops the mainsail and stage five fires my poodle. 

My poodle has enough fuel to finish the mun burn, get me to munar orbit, land and boost me back into a munar ejection trajectory.

After I pull off a perfect landing I do my science, fire my poodle and drop it mid flight to further pollute the mun.

Stage five fires my atomic rocket which is perfect to put me into a return trajectory and have enough fuel to have a semi powered deceleration. I drop my uranium power engine in the upper atmosphere and proceed to land. Mission accomplished.

 

My trouble is this rocket cost 122k money. I need advice for a cheaper, smaller rocket that can put a science lander capable of kerbin return on the mun.

 

[Starman4308]

It would help to have a craft file, because then we can see how much fuel each stage has, etc. I would suggest installing either KER or MechJeb, and checking the delta-V. My initial suspicion is too much thrust, not enough fuel/burn time per stage.

EDIT: And it may help to refer you to KerbalX, a craft file sharing website.

Edited May 14, 2017 by Starman4308

[Foxster]

Could you post a piccie?

[ArmchairPhysicist]

Sure, I can. It will be just a bit. I'm testing my first shuttle style craft

[Human Person]

You have many stages there! I usually build a 2 stage rocket for Mun landings with a delta v of at least 3000 m/s each.

a good configuration to start with is a Terrier upper Stage and a Skipper first stage (add boosters here, if TWR is too low)

You don't need nuclear engines.

Edited May 14, 2017 by Physics Student

[Capt. Hunt]

The NERV is a very heavy rocket engine, the LV-909 has similar vacuum performance but is much cheaper and lighter.  Also, what you've described is called a "Direct Ascent" lunar mission.  One way to make do with cheaper, smaller rockets is to do a more apollo style lunar orbit rendezvous.  It's slightly more complicated, but you won't have to land your entire rocket on the mun.

[bewing]

You're not getting enough use out of your beloved atomic rocket, to start with (I love me some nukes, too). You can use one nuke to get all the way from LKO to the surface of the Mun, and back to KSC again. Obviously, you would need to add more fuel to it for that -- but that gets rid of 3 of your rocket stages. Assume 1200 to 1600 fuel for your nuke, instead of 400. If you're going to have an extremely high efficiency engine somewhere on your ship, you need to try to use it from the moment you are in space, until the moment you get back into atmosphere again.

Your Kerbin reentry stage is most likely over-engineered and much heavier than it needs to be -- and that extra mass cascades down your design in an exponential fashion. This is the cruel part of rocketry -- that exponential mass explosion makes it imperative to be meticulous about minimizing the mass of your upper stages.

A Skipper plus a few Kickbacks can usually do the job of a Mainsail -- which saves a lot of mass.

Also, the more gigantic your rocket, the more important it is to do a really nice gravity turn -- your launch profile sounds very inefficient. For example, that extremely high initial Ap -- you never said that you make any effort to time that Ap so it points at the Mun's future location.

 

Edited May 14, 2017 by bewing

[ArmchairPhysicist]

It's not letting me get the picture, sorry. I'm mainly wanting tips on small rocket design 

[Capt. Hunt]

for the pic, you need to host it on a third party website like photobucket or IMGUR.

Edited May 14, 2017 by Capt. Hunt

[FullMetalMachinist]

11 minutes ago, ArmchairPhysicist said:

It's not letting me get the picture, sorry.

You need to upload the picture to an image hosting site (lots of people use imgur) and then post the link to it here. 

11 minutes ago, ArmchairPhysicist said:

I'm mainly wanting tips on small rocket design 

As @bewing said, make the thing that fully returns to Kerbin as small as possible. You don't have to return your science instruments to Kerbin. Take readings on the way to the Mun, then send a kerbal on EVA, get close to each one and right click it, click on 'take data', and then board the command pod. That will store the experiment results in the command pod and free up the instrument to be used again (except the mystery goo and science jr, those have to be reset by a scientist kerbal to be used again). 

Do that over and again all the way until you've landed on the Mun. Then (if your ship design has a stage that you're already leaving behind) ditch the science instruments, and return only with the command pod. 

[Capt. Hunt]

It also sounds like for less weight then the NERVA you could just add more fuel for the poodle and use that to bring it all the way back to kerbin.

[Draalo]

Your Lander is way too heavy. Caused by the NERV.

Yes, i love that engine too, but it weights 3to. Compared to, lets say, Pod + Science = 1.2to.

A Terrier with 0.5to. has the same thrust. NERV has a better ISP than the Terrier but the fact that weight is reduced by ~50% reduces the amount of needed fuel...

Try this:

First and second stage will lift into LKO with ~400m/s spare dV.

Lander is able to transfer into Mun SOI, circularize, land, takeoff, escape and return to Kerbin.

Also able to land on Minmus and hop 2 or 3 times to other biomes for a great amount of science.

Compare the weight with your Lander, this one has 7.8to. fueled.

[bewing]

I totally disagree with the people who complain about the mass of the NERV, of course. If you just do a test run with a NERV-based rocket to the Mun, vs a terrier -- you find the terrier needs at least 3 tons more fuel than the NERV does, which totally eliminates any perceived benefit.

 

Small rocket design tips:

#1 most important tip: Minimize the number of kerbals.

Use your most efficient MK1 engine for your upper stage, which should be all MK1 parts.

Don't simply assume that you need separate lander and return stages. See if your return stage can also function as a lander.

Make sure each stage is 50% to 80% fuel.

Do a good gravity turn.

Aim your Ap to point at your destination, before you do your injection burn.

Burn all your fuel in each stage.

Don't overengineer your upper stages.

Work to minimize the number of parts and their mass.

 

Edited May 14, 2017 by bewing

[FullMetalMachinist]

1 hour ago, bewing said:

Don't overengineer your upper stages.

Quoted for truth. @ArmchairPhysicistthis is perhaps the most important lesson to learn about designing rocket stages. Every gram of extra mass in your top stage propagates down the rocket and exponentially increases the amount of total fuel needed. 

Design the rocket for the mission in reverse. And try to make it so that each stage has just enough fuel/supplies to do the part of the mission it's designed for. 

[Spricigo]

11 hours ago, bewing said:

I totally disagree with the people who complain about the mass of the NERV, of course. If you just do a test run with a NERV-based rocket to the Mun, vs a terrier -- you find the terrier needs at least 3 tons more fuel than the NERV does, which totally eliminates any perceived benefit.

May I propose the migth Spark and a couple of Spider as a 3rd alternative:

Spoiler

 

Launched with SAS off, 2xKickbacks throw the apoapsis to 115km and 1,7km/s (just hit the spacebar and watch) . Terrier from there until circularization in LMO, the lander go down but only the capsule come up and return to kerbin. Main issues: need to be careful to maintain the solar panel in the sunlight, requires atention with landing trajectory (not much TWR for a suicide burn).

Its possible to cut a few hundreds kilograms here and there to make it more "efficient". But I'd rather try hopping to an extra biomes.

 

 

9 hours ago, FullMetalMachinist said:

Design the rocket for the mission in reverse. And try to make it so that each stage has just enough fuel/supplies to do the part of the mission it's designed for. 

I'd say fuel/supplies/hardware. Oversized engines are prime examples of deadweight we often carry in our rockets.

[Wanderfound]

Seconding everyone else: too big, too heavy, overcomplicated. Nukes are massive overkill for a short hop to the Mun. Once you're in orbit, the only engines you'd see on a sensibly sized Munlander would typically be Terriers or Sparks.

A smaller demonstrator, built to have enough dV to hop between a few biomes while on the Mun:

That's still a bit on the beefy and over-expensive side by my standards. These days, I'd be more likely to do the lander like this:

Landing on Moho in that pic, but a similar concept is fine for the Mun and Minmus. The narrower form factor of the 0.625m radial tanks makes it much easier to fit inside a cargo bay or under a fairing, removing the need to streamline the whole ship.

Edited May 15, 2017 by Wanderfound

[Reusables]

If you consider roving mun, this is enough. It is even fully reusable.

Nervs will shave the mass further. (If I recall correctly, this is before I got Nerv tech)

[GoSlash27]

What @FullMetalMachinist said.

 The trick to building light, cheap efficient rockets is to design backwards from the end of the mission to the beginning. Plan out what your DV and t/w requirements are for each phase of the mission and then design each stage to those requirements.

 Don't put in more t/w than you need and while a little more DV for a safety margin is okay, don't overdo it.

Design your payload to be as light as possible while still getting the job done. For upper stages, light is better than cheap. For booster stages, cheap is better than light.

 The process is outlined in @Norcalplanner's tutorial here

 

The math behind the process can be found here:

 

Best,
-Slashy

[HvP]

On 5/14/2017 at 1:22 PM, bewing said:

I totally disagree with the people who complain about the mass of the NERV, of course. If you just do a test run with a NERV-based rocket to the Mun, vs a terrier -- you find the terrier needs at least 3 tons more fuel than the NERV does, which totally eliminates any perceived benefit.

Keep in mind that the OP requested a cheaper rocket, not just a lighter one. The Terrier costs just 390 funds, the NERV costs 10,000, and he's just going to be ditching that engine upon reentry.

[bewing]

1 minute ago, HvP said:

Keep in mind that the OP requested a cheaper rocket, not just a lighter one. The Terrier costs just 390 funds, the NERV costs 10,000, and he's just going to be ditching that engine upon reentry.

True. But the best way to make it cheaper is to make it reusable. Use a nuke, make it refuelable, leave it in LKO -- and have a personnel bus to do the KSC<->LKO run.

 

[StrandedonEarth]

On 5/14/2017 at 11:22 AM, bewing said:

Burn all your fuel in each stage.

Sigly agrees.

On 5/14/2017 at 11:22 AM, bewing said:

Don't overengineer your upper stages.

Oh... Oops

Edited May 16, 2017 by StrandedonEarth

[Spricigo]

3 hours ago, bewing said:

True. But the best way to make it cheaper is to make it reusable. Use a nuke, make it refuelable, leave it in LKO -- and have a personnel bus to do the KSC<->LKO run.

 

 

No, the best way to make it cheaper is to do the task at hand with minimal expense of resources. Good luck trying to design a nuclear Mun lander that operate cheaper than something like this:

It collect science of 2-3 biomes and come back to LMO for refuel, total fuel cost of a full tank is \F28.

While it is capable to do so (after refuelling), It never return to kerbin, that is what ESU are for.

[bewing]

You can buy fuel in LMO for KSC prices? I didn't know that.

 

[Blaarkies]

2 hours ago, Spricigo said:

 

No, the best way to make it cheaper is to do the task at hand with minimal expense of resources. Good luck trying to design a nuclear Mun lander that operate cheaper than something like this:

It collect science of 2-3 biomes and come back to LMO for refuel, total fuel cost of a full tank is \F28.

While it is capable to do so (after refuelling), It never return to kerbin, that is what ESU are for.

Really?OP posted this: "Noob needs rocket science lessons.", then you come along and brag about your garden chair science fare project?

It is clear OP is "noob" and admits that, but they want to learn more via this thread and that is awesome (we have all been there, with oversized wobbly rockets)...so teach/explain some useful tips to them (what is the idea behind your design, why is it so efficient, pros/cons,...) 

How do you get Crew Reports with that anyway?

[Spricigo]

11 hours ago, bewing said:

You can buy fuel in LMO for KSC prices? I didn't know that.

 

A valid point... if that extra fuel actually weighted more than a nerv... or if someone had a nuclear design that moved itself around for less. 

10 hours ago, Blaarkies said:

Really?OP posted this: "Noob needs rocket science lessons.", then you come along and brag about your garden chair science fare project?

It is clear OP is "noob" and admits that, but they want to learn more via this thread and that is awesome (we have all been there, with oversized wobbly rockets)...so teach/explain some useful tips to them (what is the idea behind your design, why is it so efficient, pros/cons,...) 

How do you get Crew Reports with that anyway?

I pointed out "the best way to make it cheaper is to do the task at hand with minimal expense of resources". To exemplify it I strapped together some scientific instruments and a scientist in the smallest engine in the game and some fuel.  But all you noticed it's that it don't have a closed pod?

Yes,  I decided to not be able to collect the less valuable scientific experiment and halve the mass of the lander.  I consider a good trade, but if you consider a pod mandatory just include it,  add some fuel and another ant. Still a capable vessel for the task at hand weighting about 1,5t.

Anyway don't just say I'm bragging about my design, show an alternative or point what you think can be improved. I really will appreciate constructive criticism,  and even more if you contribution make me notice something I'm overlooking.