Stock Payload Fraction Challenge {1.3.x Reboot}

[Laie]

3 hours ago, AeroGav said:

Lots of very long and thin mk1 sized rockets out there, but do people know that 2.5m fuel tanks have the best capacity / drag ratio of all ?   Maybe an orange tank based lifter, with a 2.5m tricoupler and some aerospikes/vectors ?

I didn't know but found out over the week. This isn't just a marginal benefit, btw, but quite substantial. Simply replacing the Mammoths with vector quadcouplers / orange stacks takes the vessel across the 28% boundary.

Considering that to be boring, I tried to leap to 29% right away... but cannot quite make it work. Not yet at any rate, not sure if ever.

 

[SanderB]

I'm not interested enough to improve Laie's latest craft. I think 0.3% improvement is definitely possible. Beyond that, i don't know and I don't want to spend the time to find out.

[Fauble2000]

On 11/13/2017 at 11:04 AM, ATEC said:

- The payload MUST be a DEAD WEIGHT until decoupling and after that only the usage of RCS and/or SAS is allowed to fine-tune the orbit

This seems unclear. The rule seems to allow RCS to raise the Pe to the requisite orbit. What parts are and are not allowed?

Are batteries , solar panels, or RTGs allowed?
Are reaction wheels allowed?
What about Probe bodies?
Nose cones / fairings?
Control fins?
Passive fins?
If a probe body is allowed, then what about using said probe body's reaction wheels to provide torque?

Personally, I would clarify by prohibiting everything but unused fuel tanks, passive aerodynamics / fairings, and decouplers, and one probe body (KER gets weird if you don't allow this).

Edited December 2, 2017 by Fauble2000

[swjr-swis]

4 hours ago, Fauble2000 said:

Personally, I would clarify by prohibiting everything but unused fuel tanks, passive aerodynamics / fairings, and decouplers, and one probe body (KER gets weird if you don't allow this).

Personally, I would ask that we please not restrict payloads any further, forcing them to be even more dead weight than they already are.

 

I do realize it's a challenge with a very specific scoring objective, and there's really no substitute for the sheer density of ore tanks in stock, so of course most entries are going to be lobbing up ore. I also get that the competitive pressure is not conducive to greebling up a payload, when there are mass fractions and gravity turns to perfect. All understood and accepted.

But the rules already explicitly disallow any kind of ship/transport/lander (no engines!). Let's please not restrict it so far that we can't even lift an independently operating module/station/satellite if we choose to, something that can interact in/from orbit in some other way than just as a collision hazard.

 

 

[Fauble2000]

New record: 28.407%
So freaking hard.

Album a/USWqE will appear when post is submitted

Edited December 3, 2017 by Fauble2000
Sorry, not used to posting imgur.

[SanderB]

4 hours ago, Fauble2000 said:

New record: 28.407%
So freaking hard.

Album a/USWqE will appear when post is submitted

 

How do you decouple the side boosters with vertical decouplers? I don't see it from the screenshots. Could you post a craft file? I'm having difficulty reproducing this craft.

[Laie]

Utilizing nukes, discarding the Rhino, and taking some advice from @AeroGav, have lead me to this vessel:

small gallery -> craft file <-

Note that the vessel has a kind of staging that neither KER nor MJ can handle. I made some modifications for the screenshot, the dV and burn times as pictured are very close to the real thing (1st stage actually runs 53sec, overall dV is more like 2950m/s). If you D/L and try the craft file, do not be alarmed if it appears to be nonsense -- just stage at burnout and everything will be fine.

Payload is 290t, the whole vessel is just a hint shy of 1000t, and in four attempts I managed to make orbit twice. I think I can in good conscience claim to have exceeded a 29% payload fraction.

Usage: point over, not too quickly. Actual prograde should be about 60° at 150m/s, then follow through. If you have a 50-second time-to-apoapsis at burnout of the first stage, you're on the right track: just stick to prograde. Important notice: do not revert to launchpad. On subsequent attempts, MJ reports substantially higher drag and the performance will be noticably worse (cargo bay not working, perhaps?). Reverting to VAB and rolling it out again works around that issue.

Design-wise, I don't think I left much on the floor. At least when it comes to marginal improvements.However, considering the low drag, it might be worthwhile to discard most of the aerospikes and go all-in on Vectors instead.

Edited December 3, 2017 by Laie

[Fauble2000]

5 hours ago, SanderB said:

How do you decouple the side boosters with vertical decouplers? I don't see it from the screenshots. Could you post a craft file? I'm having difficulty reproducing this craft.

How do I upload the craft file?
In the mean time, the decouplers were actually TR-V2s mounted to octagonal struts. Also, everything was stitched together using advanced tweakables auto-strut with rigid attachment enabled to improve rigidity.

[SanderB]

10 minutes ago, Fauble2000 said:

How do I upload the craft file?
In the mean time, the decouplers were actually TR-V2s mounted to octagonal struts. Also, everything was stitched together using advanced tweakables auto-strut with rigid attachment enabled to improve rigidity.

for small files http://www.filedropper.com/ is very convenient

Edited December 3, 2017 by SanderB

[Fauble2000]

2 hours ago, SanderB said:

for small files (<10MB) http://www.filedropper.com/ is very convenient

Awesome, http://www.filedropper.com/missingsomething

Edited December 3, 2017 by Fauble2000

[SanderB]

1 hour ago, Fauble2000 said:

Awesome, <a href=http://www.filedropper.com/missingsomething><img src=http://www.filedropper.com/download_button.png width=127 height=145 border=0/></a><br /><div style=font-size:9px;font-family:Arial, Helvetica, sans-serif;width:127px;font-color:#44a854;> <a href=http://www.filedropper.com >file upload storage</a></div>

just post the link normally

Laie's craft:

. There is 2 tons of unspent liquid fuel in 90kn LKO so it is theoretically possible at least to get 0.2% more.

Edited December 3, 2017 by SanderB

[Laie]

@SanderB: Whoa, that's neat. On my best attempt I had like 10m/s left, on the others I could barely eke out 90x90 on vapors. Considering the number of "almost but not quite" designs that preceded this, I dare say that 0.2% is a great improvement.

Also: Whenever I take screenshots it becomes a failed attempt; I do have video of a successful launch, but have spent the last hour trying to figure out a workable compression that's still intelligible (I'm hosting my own files, hence bandwidth is a bit of a topic). Your upload solves that nicely. Thanks!  And anyway, independent confirmation is always great. Double thanks!

[SanderB]

41 minutes ago, Laie said:

@SanderB: Whoa, that's neat. On my best attempt I had like 10m/s left, on the others I could barely eke out 90x90 on vapors. Considering the number of "almost but not quite" designs that preceded this, I dare say that 0.2% is a great improvement.

Also: Whenever I take screenshots it becomes a failed attempt; I do have video of a successful launch, but have spent the last hour trying to figure out a workable compression that's still intelligible (I'm hosting my own files, hence bandwidth is a bit of a topic). Your upload solves that nicely. Thanks!  And anyway, independent confirmation is always great. Double thanks!

All I did to improve the scores was to not fly your rockets sideways on into the airstream .

Actually after getting the right ascent trajectory and hitting prograde SAS, all I do is stage. It took me about 10 attemps to get this result.

I use Open Broadcaster Software. The files that it produces are only about 9MB per minute with my settings and they upload handily and very quickly to youtube within 20 min of recording.

Edited December 3, 2017 by SanderB

[AeroGav]

1 hour ago, SanderB said:

just post the link normally

Laie's craft:

. There is 2 tons of unspent liquid fuel in 90kn LKO so it is theoretically possible at least to get 0.2% more.

That's some pretty neat flying !

Some thoughts

1. Given that 2.5m parts have much lower drag than mk3,   why wrap the payload in a CRG100 bay?  Just have the 2.5m ore tanks naked to the wind on the final stage
2. i notice on the second stage there is a single Vector surrounded by Darts on each booster.    This is often well over 20 km before being dropped - given how poor Vector vacuum ISP is, surely it is better to drop all vectors by a lower altitude?
3. similarly, can the nukes be staged in any earlier.  By 10km they are close to full vacuum ISP,  in fact by 5km they have more ISP than chemical engines, so might as well turn them on ?
4. tricoupler is the lowest drag - i notice you use a 1 - way adapter at the bottom of some stacks,   with a load of radially attached engines.   I dont know if this is as good from a drag point of view but it does pack the engines more densely within the 2.5m radius.   I suppose by this point drag is relatively unimportant compared with getting the right TWR for each stage,  so your design might still be the best..

[Laie]

2 hours ago, AeroGav said:

That's some pretty neat flying !

Some thoughts

1. [EDIT: see above]
2.  ...

1. No good reason. I started my design with a 3m stack and a Rhino below... afterwards I still kept the cargo bay to avoid the wet noodle effect of stacking 20 tanks. (Didn't even try, maybe it's not bad at all).

2.The bundle you see is a compromise, my solution to wanting a better ISP but still needing a high TWR. While I could give you a good earful about why I did things just that way, the sad truth is that it was all a waste of time: in the end, TWR trumps everything. I just did a quick revision of the rocket, going all vector on the boosters, and it works even better than the one posted above. 30% seems quite possible, with a much simpler design.

I don't like how this turns out, but it appears that no upper-stage shenanigans can compensate for the gravity losses of a steep ascent. The spread between 315s and 340s just isn't large enough.

The key to a good payload fraction is to fly as fast and as shallow as you possibly can without burning up. No footnotes, no caveats.

3. The nukes *are* running from the ground up. Or at any rate, they're supposed to. If they don't, I put up the wrong craft file. Ideally one should touch them off at ~20seconds, but what the hell. The nosecone tanks on the lower stages provide LF for the nukes. I can't precisely pin down just how worthwhile they are, or if they do any good at all. When seen in isolation as a last stage only, they are hardly better than a LV-909 would be -- whatever value they add, it has to happen before. Like,  in my case, raising the last stages' ISP to ~370s.

4. on a per-engine basis, it's worse than tricouplers but still better than the quadcoupler. It's also much better than using (say) two tricouplers on two stacks. That lesson might even be valid for spaceplane use: even the evil quadcoupler can arguably be worthwhile if it spares you from adding a whole new stack.

 

Edited December 3, 2017 by Laie

[SanderB]

29.2% achieved. This rocket uses 14 vectors 2 nerv engines and I completely made it up myself after reading Laie's posts. I can launch it pretty consistently.by pointing it prograde at 50m/s airspeed. Some tilting up is required to maintain enough upward vertical velocity and you do actually need to do a 2nd pass through the atmosphere, but... improvement has been achieved. Or at least, I have attained the fraction that I said could theoretically be attained with Laie's craft.

craft file: http://www.filedropper.com/004fpayloadfractionchallenge

And I still had 19m/s dV to spare at the end, so further improvement yet again is possible...

[SanderB]

29.4% (184.852/629.243) has been achieved by adding a little bit of payload and rounding up from 29.38%. And yet again, improvement is possible because dV is left over. The only difference between this craft and the former is a bit of added payload.

craft file: http://www.filedropper.com/004fpayloadfractionchallenge_1

[ATEC]

@Fauble2000 I agree with @swjr-swis about the dead-weight rule.

I think that if someone exploits rcs to get the craft up i'll probably dis-allow it but if you want rcs just to get some usability out of the craft that's ok

@Fauble2000

Would also say that Fuel-tanks are functional parts... You clipped a lot of them so i won't be able to allow your entry Sorry

@Laie Congrats on the first entry to the unlimited category  Yer up on the board

@SanderB I haven't put your newest entry up on the board because this is a design challenge and you used the design from someone else. In the noteworthies tho (:

[SanderB]

My 29.4% entry was inspired by Laie's craft but I made the design up myself and I think it is substantially different, and it performed better than anything else posted in this thread by someone else.

[Laie]

43 minutes ago, SanderB said:

think it is substantially different

Indeed it is. @ATEC, would you please reconsider?

1 hour ago, ATEC said:

@Laie Congrats on the first entry to the unlimited category

Come again? There's plenty of intakes (because those are the better nosecones), but no jets. Rocketry all the way.

[ATEC]

@Laie Fixed the rules, Nerv's now also count towards unlimited and not rocket only

[SanderB]

27 minutes ago, ATEC said:

@Laie Fixed the rules, Nerv's now also count towards unlimited and not rocket only

Can you compare my 29.4% craft and Laie's craft? I think they are easily different enough to qualify as different submissions. My payload is made of unshielded ore tanks with a fairing on top, I use only 2 engine types, have fewer stages, my vessel is released from launch clamps and has very different starting mass; to name a few differences.

[Fauble2000]

6 hours ago, ATEC said:

@Laie Fixed the rules, Nerv's now also count towards unlimited and not rocket only

So what's the current non-nuke rocket record?

[swjr-swis]

7 hours ago, ATEC said:

Fixed the rules, Nerv's now also count towards unlimited and not rocket only

Were some posts deleted from the thread? I've gone through the whole thread again and I can't see what prompted this change. What is the reasoning... what does this fix?

 

[Fauble2000]

28.23% Rocket Only (no nukes)

I figured out how to build an open-ended fairing and stuffed most of the components into that.

The advantages of the open ended fairing are enormous. They allow you to cluster up to twelve 1.25m engines around a central core, which makes up part of the payload. these engines will have thrust vectors directed just outside the outer edge of a 2.5m lofting stage. This enabled me to house 12 Darts and 3 Vectors under one fairing just a touch wider than 3.75m. No need for nosecones other than the forward fairing, (which may benefit from replacement with stock adapters, though it seems to me that a well tapered fairing offers superior aerodynamics).

I could probably push the fraction well above 29% by getting rid of the big orange tank, adding a fourth tier of engine-less tanks and optimizing the staging and flight profile. one might also consider dropping vector engines one at a time mid flight. Though this will introduce thrust asymetry, it's nothing the vectors can't compensate for.
 

Album a/up2eT will appear when post is submitted

Edited December 5, 2017 by Fauble2000
typo