parts [1.12.x] Sounding Rockets! Start small. Dream big!

[Wisq]

NIIIICE, but um how do?

The key is a) radial decouplers that you never decouple and rockets that you never actually light off (acting as structure) but that still have a tiny amount of fuel (so they don't automatically jettison). Basically, I started with one nosecone and (structural) SRM-S at the core, added radials (70% with TweakScale, but 100% should work okay), added four more (structural) SRM-Ses, then added four nosecones, four (actually fuelled) SRM-Ses, and four SRM-XLs (with adapters).

The result is a quad-rocket that can make it to about 100km apo if fired straight up (with FAR + KIDS "FAR to stock" profile), or about 63km if fired at the angle in that picture. The stages are, big SRMs, adapters, little SRMs, parachutes, and then stage 0 (above the parachutes, never fired) is the decouplers and final five SRMs.

Parachute velocity is low enough to protect the payloads, but high enough that it does still break up when it hits the ground. Typically it breaks up into five pieces, so I recover the core, then recover the four packages with the Tracking Station view. On one occasion, it broke into only three pieces (core+2 attached, 2 separate).

The nosecone in the middle is key, since it needs all five parachutes to (barely) safely land. Also probably helps aerodynamics.

[hakan]

Also, I miiiiight be abusing this a bit.

You know you are evil, right?

[UnanimousCoward]

Without FAR

With KJR

EDIT: Ah, frigglenuggets. Just reread the previous page, it seems KJR is what's causing my issues. I might just have to use a really tiny decoupler.

Had the same issue with the booster not separating. Strangely, when I forgot to set up staging in the VAB, so I had to do it on the launchpad, the booster separated just fine. Haven't done much (well, any) testing of this, but every sounding rocket I've sent up after setting the staging on the launchpad has worked great.

[Wisq]

If anyone's interested, I just started blogging a Kerbal science campaign, starting with 31 iterations of sounding rockets (with diagrams), most of them being variations on my quad design: http://wisq.net/2014/12/31/kadet-1/

[Black-Talon]

Update on Failure to separate booster issue -

Tested with FAR & Module Manager Only - Issue Persists

Given that I've also experienced the issue with and without FAR with no change, I'm left suspecting ModuleManager or the root cause to be in the behavior of USI_DropTank Module.

New Theory!!

- Since I'm using ModuleManager to add the USI_DropTank Module to my stock RT-10 Solid Booster AND I'm cutting the fuel down to 216.5 liters I happen to notice that my solid rocket booster sometimes flames out with 0.2 liters left in the tank. I have no idea if this is normal or why it sometimes happens (or is it always and I haven't noticed?). But hey, it's possible that USI_DropTank is waiting for the Fuel to reach "empty" in order to trigger a separation/ejection/jettison. And it's possible that 0.2 isn't considered empty. And it's possible that my setup of using USI_DropTank on the RT-10 AND cutting the fuel to 216.5 liters (50% of full) exacerbates the chances that flameout occurs with a few drops of fuel left in the tank.

Thoughts?

- - - Updated - - -

Had the same issue with the booster not separating. Strangely, when I forgot to set up staging in the VAB, so I had to do it on the launchpad, the booster separated just fine. Haven't done much (well, any) testing of this, but every sounding rocket I've sent up after setting the staging on the launchpad has worked great.

BTW - I've tested and confirmed this trick/work-around does not prevent the separation issue in my case. You had me hoping though!

[RoverDude]

FYI - here's what's coming up next

[Black-Talon]

FYI - here's what's coming up next

http://i.imgur.com/1HNA7BZ.png

Awesome!

Any leads on what causes FAR to be upset with the current parts that you've been able to avoid in the new modular parts?

And any thoughts on the belief that booster ejection fails in those rare occasions where a fraction of SolidFuel is left in the tank unburned?

[hakan]

FYI - here's what's coming up next

http://i.imgur.com/1HNA7BZ.png

Extremely cool

[RoverDude]

Awesome!

Any leads on what causes FAR to be upset with the current parts that you've been able to avoid in the new modular parts?

And any thoughts on the belief that booster ejection fails in those rare occasions where a fraction of SolidFuel is left in the tank unburned?

Talked to Ferram - it's the presence of a chute. Note that in the new version, said nosecone cannot be the root part.

I'll also likely drop in a mini decoupler given that we already get decouplers at 5 science anyway.

[Black-Talon]

And any thoughts on the belief that booster ejection fails in those rare occasions where a fraction of SolidFuel is left in the tank unburned?

I'll also likely drop in a mini decoupler given that we already get decouplers at 5 science anyway.

I now see that USI_DropTank does have a "threshold" value set in the part. Adjusting this should at least let me test my theory.

- - - Updated - - -

Setting the USI_DropTank threshold = 0.015 does the trick for me. Setting it to 0.01 did not. Fuel in one of my test cases was frequently cutting the engine with 0.01 units shown in the UI as remaining in the tank.

I have some ideas on how USI_DropTank logic could be changed so threshold can be set higher by default without negative impacts. But one thing at a time... Going to finish these darn test cases even though they may not be all that useful with the new version (which looks super awesome!) in the works!

[Benjamin Space Program]

Could I include this in Future Reloaded?

[RoverDude]

I have no idea what Future Reloaded is. But the license terms are covered on the first post.

[AntiMatter001]

it pays to read... i just spent the last hour looking for a decoupler/seperator for this rocket...

god damn my lazy bum...

[Black-Talon]

RoverDude, what do you think about the approach below for a DropTank that drops even when it has a little fuel left in it? The issue being that the tank shouldn't be dropped until the fuel can no longer be burned but we don't know what exact amount could be left unburnable and sitting in the tank. Your current implementation allows for fuel to be left in the tank and still drop by setting a "threshold" of fuel remaining which is considered empty. It has the drawback of requiring a fixed amount of fuel to be defined as empty, and as was the case with my test cases, the threshold may be set too low and prevent the tank from decoupling. But set the threshold too high and the tank will decouple while a Solid Rocket Motor is still burning. With the approach below the fuel remaining is checked, once the threshold is crossed, to verify that fuel is no longer being consumed from the tank prior to decoupling. This allows the threshold to be set higher without risk that the tank will drop before being empty. In turn, this solves my issue of the tanks not decoupling because the threshold is set too low for a particular tank/engine.

Your thoughts?

    public class USI_DropTank : PartModule
    {
        //tank decouples when out of resources.  May have hilarious results.

        [KSPField]
        public bool explode = true;

        [KSPField]
        public float threshold = 0.01f;

        private Dictionary<String, double> previousAmounts = new Dictionary<String, double>();

        public override void OnUpdate()
        {
            if (!HighLogic.LoadedSceneIsFlight)
                return;
            if (vessel != null)
            {
                bool drop = true;
                double previousAmount = 0;
                foreach (var res in part.Resources.list)
                {
                    if (res.amount / res.maxAmount > threshold)
                    {
                        drop = false;
                    }
                    else
                    {
                        if (previousAmounts.TryGetValue(res.resourceName, out previousAmount))
                        {
                            if (previousAmount != res.amount) drop = false;
                        }
                        else drop = false;

                        previousAmounts[res.resourceName] = res.amount;
                    }
                }
                if (drop)
                {
                    if (part.parent != null)
                    {
                        part.decouple();
                    }
                    if (explode)
                    {
                        part.explode();
                    }
                }
            }
        }
    }

- - - Updated - - -

Sounding Rockets + FAR

I have more or less completed the testing on the ModuleManager Configs I made for Sounding Rockets v0.1.1 to better work with FAR and I'm pretty happy with them!

Data if you want it - https://docs.google.com/spreadsheets/d/1K9UVrXre7bIuG8vGBbowGKdB5ZK6XZxiga1wa1N4vmM/edit?usp=sharing

Short Story:

With these configs the performance of Sounding Rockets is inline with the performance of stock boosters/parts when using FAR. It adds drag to the Nosecone and correct lifting surfaces and thus drag models to the fins. You should expect the performance of Sounding Rockets while using FAR to be enhanced to the same degree that Stock parts performance is - thrust/isp of the rockets was not adjusted.

Known issues:

FAR and Sounding Rockets don't work well together because of the Nosecone with a command module and a parachute in the same part. This results in:

- FAR Flight GUI is not able to be opened when the Nosecone is on the craft

- When applying drag using my configs, a Revert to Launch with a craft equipped with the Sounding Rockets Nosecone will cause FAR to freak out and no drag will be applied to the craft for that next flight. Reverting to VAB/SPH instead prevents this. I believe this is fixed in FAR v0.14.6.

Download Configs (these are identical to those I shared previously): https://www.dropbox.com/s/6kgtqncsyxuui3a/FAR%20ModuleManager%20Configs.zip?dl=0

And it sounds like RoverDude will Soon have a new version of Sounding Rockets that avoids the FAR Flight GUI issue! *cheer*

-Talon

Edited January 3, 2015 by Black-Talon
Added title to second post since it combined them

[Black-Talon]

I also added a time delay to the DropTank

- It was pretty simple to add a configurable time delay and once I was in the code so I added that feature to my copy. It makes the drop a little cleaner since the forces on the parts are kinda going crazy at that moment of engine cutoff. Happy to share suggestion though if for any reason you want it. I should probably figure out this GitHub thing in order to suggest code...

[LucidPixels]

FYI - here's what's coming up next

http://i.imgur.com/1HNA7BZ.png

Very much looking forward to that. I was just wishing for the ability to fit more science/stock science modules into these.

[SimplyCosmic]

Thank you, RoverDude! I have started loving Sounding Rockets as the "model rocket simulator" inside my "cartoon rocket simulator", so I look forward to every new update.

I'm glad to see you're using slightly slanted fins on your rocket. After watching real-world NASA sounding rockets, I noticed they use tiny thrusters just after launch to start them spinning, but this doesn't work as well in KSP. So I went with slanted fins, and they work reasonably well at helping keep the unguided rockets on track.

[Cleric2145]

An easy fix to the FAR/parachute/drag issue would be to cut the nosecone into two pieces, the cylinder beneath the cone becoming the "Gyroscopic Guidance Package" (probe core) and the actual nosecone becoming the non-rootable parachute part. That way you have one part creating drag. This would also open the door to using the nose cone parachutes on the solid rockets as boosters without actually adding probe control cores to them (mainly for aesthetics, since they are too cheap to care about recovering.)

[Black-Talon]

An easy fix to the FAR/parachute/drag issue would be to cut the nosecone into two pieces, the cylinder beneath the cone becoming the "Gyroscopic Guidance Package" (probe core) and the actual nosecone becoming the non-rootable parachute part. That way you have one part creating drag. This would also open the door to using the nose cone parachutes on the solid rockets as boosters without actually adding probe control cores to them (mainly for aesthetics, since they are too cheap to care about recovering.)

I get the impression that is what RoverDude is doing.

Though it currently appears that the only thing FAR gets upset with is the Command & Parachute Modules being on the same part. In FAR v14.6 it *seems* like I can manually add drag to parachutes without creating an issue. So nosecone (with parachute) on top of a "Gyroscopic Guidance Package" (with command module) would work perfectly AND both parts could have FAR drag on them!

[RoverDude]

Yep. Well, close.

Your root part is a payload truss that you affix your probe core, batteries, and experiments to. The nosecone atop it includes a chute. There's also a nifty radially attached packed parachute you can plunk inside your payload packages to add extra chutes if needed.

And yeah I am totally adding in a little spin-stabilizing SRB. Basically twin thrusters that you can use to provide a quick burst of spin as needed, and stage accordingly.

[RoverDude]

A pre-release is up. THIS IS A PRE-RELEASE AND YOU MUST DELETE YOUR OLD SOUNDING ROCKETS FOLDER!

Note: This is just on Dropbox for now as this is not the official release. File is SoundingRockets_0.2.0_PRE.zip

Tech tree is fully integrated. Your starting bit should be a payload truss (you can find it and it's fairings in structural). Best is to turn off snap and surface attach your batteries, probe core, and science packages. The gyro and chute have their attachment point in the middle as they should be center-attached on the payload truss and fit neatly inside of a 0.35m payload package.

Parts include:

Winglets in three sizes

0.625m LFO tank

0.35m LFO tank

0.35m RCS stabilizer

0.625m Solid Fuel rocket

Short/Long 0.36m Solid Fuel rockets

Radially attached probe core

0.625m and 0.35m payload trusses with matching fairings

Four radially attached science payloads

Radially attached parachute

Nosecones with parachutes in 0.35 and 0.625m sizes

Launch stick

0.35m gyroscope (acts as an SAS)

0.35m stack decoupler

Radially attached mini battery (lousy EC/weight ratio but good for a starter one)

0.35m-0.625m adapter (hollow, so it does not block engine thrust)

0.35m aerospike LFO engine

Enjoy!

Edited January 4, 2015 by RoverDude

[Deltervees]

Nice Prerelease!

Does this mod use the same StoredCharge resource as the capacitors in Near Future?

[Nitrous Oxide]

The new stuff looks fantastic.

[RoverDude]

Switched to ElectricCharge and gave you a lousy battery instead

[Nitrous Oxide]

Switched to ElectricCharge and gave you a lousy battery instead

Sweet, now if you were to add a tiny solar panel (not powerful, so that you'll need more than one for a single Communitron 16), this could easily become the best mod for early game satellites (like establishing a rudimentary COMSAT network in RemoteTech)... cheap, light, and early enough in the tech tree that each early game satellite doesn't feel like a waste, while saving your money for the actual manned program. Not to mention how well they work with KCT, being a lot quicker to build and launch than typical stock rockets.

Of course, I can already achieve this with TweakScale and existing panels, but it's just a suggestion.