Another Mk3 lift problem topic

[Merandix]

Ok, I'm basically looking for some inspiration. I know more people are battling to build a shiny new Mk3 heavy lifter.

Lifting the infamous full orange tank up seems to be some sort of community benchmark for awesomeness...

But, having done that with surprising ease in a rather large 250+ part count craft (which ended up independently looking like Meslkin's Garuda though far less efficient, so I won't post it).

Turns out building it long is a good answer... but part count will rapidly soar. At 250+ parts, I ran into a lag wall I'm unwilling to deal with (over 3 seconds of real time per second of game time). Maybe I'm impatient, but it has to stay fun, and lag that bad rapidly deteriorates into annoyance. Not a feeling I'm playing a game for.

So this HAS to be possible using much less parts? Let's say, 150ish max? And to stay 'kind', I set my own bar to 27 tonnes to LKO. That sort of challenge is what I play games for So a limitation became a gameplay element lol.

I keep running out of room to put the wings on, while still making sure the craft has proper handling both with and without payload.

Anyone had a breakthrough with this? Or to put it differently: Where the bloody heck do I put all those wings? A 76 tonne craft requires around 38 static lift on each side (read that somewhere)... considering the best wings we have are 2 lift... yikes.

Very curious to seeing how other people cope with this. When I get something new that 'sort of' flies, I'll try to remember to make a screenshot.

Edited January 18, 2015 by Merandix

[jman508]

i guess you could stack your wings in a biplane type design.

Edited January 15, 2015 by jman508
doesn't sound smart enough

[Alshain]

This is a common problem in the current stock game. The issue is not that the wings don't lift enough, it is that they are designed to lift Mk2 planes. I suggest using the B9 mod Wings with all your Mk3 planes. You can extract just the wings if you don't want all the rest of the mod.

Until Squad gives us some larger wings, that is going to really be the only option to building Mk3 planes without massive part counts on the wings... unless you want to resort to welding.

Edited January 14, 2015 by Alshain

[eddiew]

Until Squad gives us some larger wings, that is going to really be the only option to building Mk3 planes without massive part counts on the wings... unless you want to resort to welding.

I look forward to that day!

IIRC welding wings doesn't currently work very well. Orientation of the panels tends to go amiss. That and I don't think FAR understands them properly, so you may not get the right lift/drag from it.

[Norpo]

Hm, might TweakScale be an option? I'm no expert on spaceplanes, though.

[Alshain]

Hm, might TweakScale be an option? I'm no expert on spaceplanes, though.

It might, to be honest I'm really not sure, I only use TweakScale on structural parts (girders, etc.) so I can't say I've ever tried it.

[Wahgineer]

This is a common problem in the current stock game. The issue is not that the wings don't lift enough, it is that they are designed to lift Mk2 planes. I suggest using the B9 mod Wings with all your Mk3 planes. You can extract just the wings if you don't want all the rest of the mod.

Until Squad gives us some larger wings, that is going to really be the only option to building Mk3 planes without massive part counts on the wings... unless you want to resort to welding.

Well, according to the latest blog (on aerodynamics), the lift revamp may allow you to produce more lift with less wings and/or speed.

[Foxster]

Perhaps I've got the wrong end of the stick but... you don't need any wings at all for a lifter, or are you specifically talking about space planes here?

There is an issue with the stock engines at the moment though - they aren't big enough. To get a heavy load (like an Eve lifter or a space station) into space requires a big messy asparagus of engines. We just need some big ass engines with 3x or 4x the thrust of a S3 KS-25x4 Engine Cluster.

I do get a bit tired of the shonky looking rockets that you have to build to achieve anything other than the simplest scenarios. There is limited scope for elegant designs.

[Alshain]

Perhaps I've got the wrong end of the stick but... you don't need any wings at all for a lifter, or are you specifically talking about space planes here?

con·text

/ˈkäntekst/

noun

noun: context; plural noun: contexts

the circumstances that form the setting for an event, statement, or idea, and in terms of which it can be fully understood and assessed.

He's talking about finding room to put wings, handling with and without payload, and static symmetrical lift, and you can't figure out he's talking about a plane?

[capi3101]

It might, to be honest I'm really not sure, I only use TweakScale on structural parts (girders, etc.) so I can't say I've ever tried it.

You can do the same sort of thing manually - I myself did it the other day to give myself landing gear better suited for the Mk3 size than the Small Gear Bay. Even wound up getting a test contract for the things.

I'll have to go find the link to the procedure, though. Only reason why I didn't post on this thread until now...

[Anquietas314]

I suggest using the B9 mod Wings with all your Mk3 planes. You can extract just the wings if you don't want all the rest of the mod.

Until Squad gives us some larger wings, that is going to really be the only option to building Mk3 planes without massive part counts on the wings... unless you want to resort to welding.

There is an alternative solution, which incidentally is handy if you use B9 and don't want all of its colossal memory footprint: Procedural Dynamics, AKA Procedural Wings AKA pWing . It is however a bit more complicated to use than normal parts, but I found it very effective for building a Mk3 spaceplane that could lift a full orange tank (and a bit more) to orbit. Sadly I don't have a screenshot of it to show you (unless Pecan wants to dig it out of his inbox - sent items never seems to work and my gaming machine is broken atm ).

[Foxster]

He's talking about finding room to put wings, handling with and without payload, and static symmetrical lift, and you can't figure out he's talking about a plane?

No need for the sarcasm - we are all friends here

He started talking about a Mk3 heavy lifter then he drifted into talking about wings. For all I knew he thought you had to add wings to get a MK3-based craft into space.

[Alshain]

No need for the sarcasm - we are all friends here

He started talking about a Mk3 heavy lifter then he drifted into talking about wings. For all I knew he thought you had to add wings to get a MK3-based craft into space.

Planes can't be lifters? I didn't mean to go off on you but we get that a lot. Someone puts the abbr. "SSTO" in the title and then writes a first post describing a plane and the next post is someone posting pictures of rockets knowing full well what the OP meant. I apologize if I was rude.

There is an alternative solution, which incidentally is handy if you use B9 and don't want all of its colossal memory footprint: Procedural Dynamics, AKA Procedural Wings AKA pWing . It is however a bit more complicated to use than normal parts, but I found it very effective for building a Mk3 spaceplane that could lift a full orange tank (and a bit more) to orbit. Sadly I don't have a screenshot of it to show you (unless Pecan wants to dig it out of his inbox - sent items never seems to work and my gaming machine is broken atm ).

Yeah, there are many wing options. B9 is the most complete modular wings outside of stock. pWings is ok but can be very frustrating to design because the interface sucks, unless it's changed recently, the mouse cursor doesn't lock in place when resizing the wings and you end up resizing something you didn't want to with no undo. That's the main reason I don't use it.

There are others that are less versatile but perfectly usable too. B9 shuttle wings for example (which is NOT B9) look just fine with the Mk3 parts, your just limited to that design.

Edited January 14, 2015 by Alshain

[capi3101]

Here we go - the procedure for "re-sizing" stock parts by hand. You could easily do something similar with any of the wings. To get them to generate more lift, you'd affect the line that says "deflectionLiftCoeff = 2.0" to some higher value. (Actually, just scaling up the model should make a wing generate more lift by definition, but I have no idea if that would work in and of itself or not).

[Merandix]

GAH. I'm so close! She handles a bit wobbly, and I acidentally let a cluster of four rapiers overheat and go boom when I was finally reaching 26 km and switching to rocket-mode. That's what I get for not paying attention. Relocated them to the wing pods, and put the turbojets on the rear. Hopefully that goes better.

She flies erm... ok... ish apart from the below problems, 'only' 188 parts, and she gets off the runway quite easily with a nice 27,2 tonnes of cargo in the rear cargohold (the front one is just there to extend the fuselage).

I'm having serious Angle of Attack problems that get worse between say 1,5 to 15 km... and then slowly get better again as I speed up. At worst the Angle of Attack is a toe-curling 35 degrees. I like my planes better when they fly at max 10 degrees AoA or so. But planes this big are hard, so any advice will be greatly appreciated.

I'm still fairly new at this game, so a picture for the Masters of Spaceplanery to judge. I'd very much like to improve this, as this can probably get into space. Will resume trying now.

188 parts

98,1 tonnes (or 96,6 according to RCS build aid) take-off weight

70,8 (or 69,4 tonnes according to RCS build aid) without payload

32,4 tonnes dry weight according to RCS build aid (without payload)

6,5 metres high

26,1 metres long

53,1 metres wide (!)

[Alshain]

What's wrong with 35 degrees? That's typical of a plane. The only way to get a higher AoA is more speed which means more power. Ideally you want an ascent angle that will get you up to maximum speed just as the engine loses it's air. If you do it right you won't have to change that angle the whole way up, if you don't you have to level off to build speed, which is still ok.

EDIT: Blast it, I just realized you're not using FAR. Ok, well that advice was for FAR users. I'm afraid I can't help much with stock, I just can't stand flying in the soup-o-sphere so I've forgotten anything I once knew about it.

[Anquietas314]

GAH. I'm so close! She handles a bit wobbly, and I acidentally let a cluster of four rapiers overheat and go boom when I was finally reaching 26 km and switching to rocket-mode. That's what I get for not paying attention. Relocated them to the wing pods, and put the turbojets on the rear. Hopefully that goes better.

I'm having serious Angle of Attack problems that get worse between say 1,5 to 15 km... and then slowly get better again as I speed up. At worst the Angle of Attack is a toe-curling 35 degrees. I like my planes better when they fly at max 10 degrees AoA or so. But planes this big are hard, so any advice will be greatly appreciated.

-snip-

98,1 tonnes (or 96,6 according to RCS build aid) take-off weight

Two things:

1) If you're going to use turbojets anyway, you'll quite possibly be better off ditching the RAPIERs entirely and using only turbojets to get to space, then using a high-ISP engine (nukes, poodle, etc) once you get there. Turbos have higher thrust than RAPIERs.

2) You don't really have enough thrust for the mass of that plane; that's most likely the reason you need such a high AoA between 1.5 and 15km. Roughly speaking you need a minimum of one engine per 13 tonnes of aircraft, which you have, but you want more than that to make it fuel efficient for getting to space and easier to fly. When you say it gets off the runway, does it lift off while still on the runway, or go flying off the end of it, but stay in the air afterwards? You should probably balance your intakes better too; right now the turbojets are getting a decent bit more than the RAPIERs, assuming intake and engine placement was done using the ordering trick and the intakes nearest the engines are for those engines.

Also, 26km isn't really high enough to be switching to rockets. You want to be doing that at more like 36-40km if you're using RAPIERs. If you use only turbojets, you can actually just get out of the atmosphere using those and not switch to rockets at all until you're in space. This probably means you need to add a few intakes

Edited January 14, 2015 by armagheddonsgw

[capi3101]

Stock aero craft, eh? Then my general guidelines apply.

-13 tonnes maximum take off mass per RAPIER (or 15 tonnes maximum take off mass per Turbojet).

-.035 minimum intake area per engine.

-1:1 Lift Coefficient to Mass Ratio.

-40 units of Liquid Fuel and 25 units of Oxidizer times the Maximum take off mass for the number of engines required (if using RAPIERS - you can get by with less oxidizer with a Turbojet/rocket setup).

-Plan for a 25% payload fraction for the craft.

The big one I'm seeing is insufficient intake area - the more you've got, the higher your atmospheric ceiling. Take a look at DocMoriarty's guide (the one in the link I provided) if you decide to stick with the RAPIERs - it's old data but the principles are still sound.

[Merandix]

@Alshain

Thanks for trying to help out anyway. I'm sticking with stock for now, this challenge is just too much fun Also, 35 degrees angle of attack is past 25. 25 is optimal lift, but also already generating tons of drag. Even in the slightly underperforming KSP aero model, as I've been told by sources long since forgotten (I hate forgetting sources).

IIRC, most large planes are falling out of the sky WAY earlier than 35 degrees. I think I've read 18-20 degrees AoA is pretty much a guaranteed stall for a 747 somewhere.

@armhageddonsgw

Maybe a good idea. Would 2 Toroidal Aerospikes and 8 turbojets be a good alternative? I shouldn't but I feel like I'm lacking rocket thrust, while in fact I should have more than enough already.

And yeah, it takes off at about 3/4ths of the runway, between 90 and 100 m/s. Could probably get off the ground quicker, but I have to be VERY gentle on the keyboard (no stick here) or I'll tailstrike her. So she spends I think a good 2-3 seconds on the ground with only the maingear still touching. And yeah, that's with the full 27,2 tonne payload. That's why I'm so boggled that I'm missing thrust high up... I have over twice the thrust at 22 km compared to when I'm taking off.... At 26 km or so it's finally dropping below take-off thrust.

@Capi3101, thanks for the suggestions! I think the payload fraction is probably the biggest culprit unfortunately.

- I'm using 4 turbojets and 4 rapiers. so that's 14 tonnes average per engine allowed, I'm at 12,26 including 27,2 tonne payload.

- 0.035 minimum intake area per engine. The RAPIERS get 0.036 (3x clipped shock cone each). The turbo's get however much 2 shock cones, a Mk1 structural intake and a REVERSED shock cone is (put it there for looks, it's lighter than a nosecone AND provides air (theoretically 0,042). If you look closely you can see the three rings making the three shock-cones on the screenshot. I like keeping them selectable for action groups. But apparently I'm JUST above minimum. So that may need some work.

- 1:1 lift coefficient. I'm using slightly more than 2:3 as indicated by THIS link. It also indicates a TTW of 0,9 should be sufficient, rather than 1,5.

- Ok, I'm at 36% payload fraction. That might be a bit optimistic.

Ok, recalculating everything for an 18,2 tonne payload (total weight goes from 98,1 to 89,1)

- TTW increased to a whopping 1,93 at optimum thrust. My lighter Mk2 planes can go vertical and still accelerate faster than terminal with a TTW like this <.< Something weird is happening here.

- Assuming I'll add 8 more shock cones (*cringes at increased part count*) I should be safe. That's 4 shock cones per engine and some extra for good measure.

- Lift coëfficient increases with reduced payload weight. I believe I have 72 static lift (excluding control surfaces). So that would be just over a 4:5 lift ratio.

- That decrease in payload puts me at a 20,4% payload fraction.

I'll call it a day for now, I'll test the 18,2 tonne payload run tomorrow, and maybe will tinker a bit with it. 18,2 tonnes is still nice May check in on the forum a few times, but enough KSP for today

Edited January 14, 2015 by Merandix

[Anquietas314]

@armagheddonsgw (fixed)

Maybe a good idea. Would 2 Toroidal Aerospikes and 8 turbojets be a good alternative? I shouldn't but I feel like I'm lacking rocket thrust, while in fact I should have more than enough already.

You're going to need more than 8 jet engines here, especially with the added mass of dedicated rockets. The spaceplane I built was a similar size to yours and used at total of 12 turbojets plus I think a skipper as the main rocket engine. It could lift an orange tank and the big RCS tank at the same time, full (but it didn't quite remain stable after dropping off the payload :/ problem was mass distribution).

And yeah, it takes off at about 3/4ths of the runway, between 90 and 100 m/s.

-snip-

That's why I'm so boggled that I'm missing thrust high up... I have over twice the thrust at 22 km compared to when I'm taking off.... At 26 km or so it's finally dropping below take-off thrust.

if it's taking you 3/4 of the runway to reach 90m/s, you don't have enough thrust. You do however have plenty of lift - too much even - which is why you can take off with the payload at what is quite a low takeoff speed. Ditching some of the wings will increase minimum takeoff velocity, but also decrease drag. If you have enough thrust you should still be able to take off. Also remember that high up, you're moving much faster, which means drag is significantly higher, even in spite of the thinner atmosphere. This tails off above ~30km; by 40-50km it's almost nonexistent, which allows you to maintain a low throttle (to ensure you get out of the atmosphere) and keep accelerating using turbojets if you have enough intakes.

I think the payload fraction is probably the biggest culprit unfortunately.

Your payload fraction is probably fine. As I said, my spaceplane was a similar size and could lift a payload of around 40 tonnes (orange tank + rcs + 2x large docking ports). Your main issue is insufficient thrust/intakes.

[Merandix]

@all the people suggesting many awesome mods

I love your suggestions, and might check them out in the future, but my stubborn Frisian self has taken up the challenge to make this happen in stock. I'm only using RCS-aid for now. I sort of refuse to believe it is impossible, just hard.

I also totally forgot to mention my sort of failed test-flight yesterday. I failed to bring up the -full- payload up into orbit. I ran out of gas on the way up and because it took it's merry time (nearly half an hour) I decided to unlock the payload's fuel. I did insert what was left of the payload into orbit after transferring the remaining fuel into my plane. So yeah, not a total failure, but enough of a failure to rethink.

Surprisingly, while the design needs more air, it seems my piloting is also a major fault in this story . The extreme AoA was caused by being pretty much borderline stalling and me ignoring it. Yeah... I'm a horrible pilot I suspect with a few more intakes, and a gentler ascent path she'll do much better. I will also stick to my 27,2 tonne target a bit longer.

And armagheddonsgw... I may or may not experiment with more thrust. More engines upset the balanced on an extremely well balanced plane, so that would likely entail a full redesign of the wings and tail. Considering the fuselage is under 30 parts, that's basically redesigning my plane. Which feels... unnecessary considering how far I came yesterday, and how wasteful my piloting was. I do fully agree that it needs more air. But I'm not ready yet for a full redesign

Did another take-off today, slightly more aggressive than yesterday (watch that tail!), and it basically gets up at around halfway the runway. It's not the acceleration to 90-100 m/s that is taking so long, it's the actual rotation that takes up a lot of space due to me being careful of tailstrike. Also, what would you consider a good take-off speed? Because I feel like this isn't slow at all with terminal velocity being so bloody close.

- - - Updated - - -

Yay! Success!!! 27,2 tonnes in a 74 by 76 km orbit with over 500 fuel and even more oxidiser left (boo, need to leave even more ox at home, and that's the plane's tanks, mind you).

What did I change? Mostly my piloting. It's still nothing to write home about, and it feels like clawing my way up into space. And if I were an actually good pilot, I could probably manage far more speed out of it than 1900 m/s at 34 km before kicking in the rockets. This could have easily been improved by having working action groups, gone up into space with open intakes. For some reason, my intake action groups did not work. I also need to figure something out to prevent overheating of my rapiers. Those things pack far more punch than the 80% max thrust I can run them at. That thrust would be appreciated in the early bit directly after the transitioning.

Other than that I added some batteries to right in front of the RAPIERS (in the hopes of solving that overheating problem, but alas, didn't work), and each engine now has AT LEAST 4 shock cones servicing it. It's still not as smooth as I like it, but yeah. A little tweaking goes a long way. Better piloting shaved 10 degrees off the max AoA. Still quite a lot at 25 degrees, but less jarring than 35. It feels a bit airhoggy though with four shock-cones per engine. But I probably shouldn't whine.

I'm still having some minor yaw-issues when climbing just after takeoff though. Anyone know how to solve that? Bigger vertical stabilisers? More dihedral on the wingtips?

[capi3101]

Moar rudder. DocMoriarty recommends anywhere from 1 to 2.5 rudder parts per engine used; last I checked you've got five rudders and no fewer than seven engines. Now, that's not terribly specific - I don't know which part he's actually talking about when he says "rudder". I'm guessing Standard Canards, but I've used AV-R8 winglets as rudders to good effect. Rudders can also help you in asymmetric thrust situations - I had one design where a engine completely flamed out but the craft held its course due to its rudders (though the fact that the remaining engine wasn't outputting a great deal of thrust at that point might've had a lot to do with it).

1900 m/s at 34k though is pretty darn good regardless of what kind of engine setup you've got - personally, I wouldn't knock it.

As far as your intake action group is concerned, you may have been hit by the old symmetry bug, especially if there was some kind of mishap during the application of the intakes. Try de-selecting the intakes from the action group, then re-applying them, and see if that solves the problem or not (might want to test it out while you're still on the runway...).

Edited January 15, 2015 by capi3101

[Anquietas314]

@all the people suggesting many awesome mods

I love your suggestions, and might check them out in the future, but my stubborn Frisian self has taken up the challenge to make this happen in stock. I'm only using RCS-aid for now. I sort of refuse to believe it is impossible, just hard.

Another useful mod I don't think anyone's mentioned yet: Intake build aid. It basically lets you see (and specify) how intakes are mapped to engines. There's also a feature to automatically balance them, but I found it misbehaves slightly - it works, but it assigns the intakes quite arbitrarily.

And armagheddonsgw... I may or may not experiment with more thrust. More engines upset the balanced on an extremely well balanced plane, so that would likely entail a full redesign of the wings and tail.

-snip-

But I'm not ready yet for a full redesign

Heh, and now you see why so many people shy away from spaceplanes . Small changes can mean major redesign . The pWing mod helps a lot with that since you pretty much only have two big wing pieces to worry about instead of 30, but as mentioned the interface could do with some improvement.

Also, what would you consider a good take-off speed? Because I feel like this isn't slow at all with terminal velocity being so bloody close.

The thing with planes is that you actually want to be slightly above terminal velocity, at least at takeoff and low in the atmosphere. In my experience 100-120m/s is a pretty decent takeoff speed, and you want to be able to reach it within the first 2/3rds of the runway, roughly. You can of course manage to do it at lower speeds, and in some cases that's desirable, but with a plane designed to carry payloads - especially heavy ones - to orbit you want the full plane with payload to be nearish to its performance limits - which means high take-off speed - because it'll be much lighter (and thus need far less lift) when you drop off the payload. Too much lift can make the plane a little hypersensitive to control input (even in fine mode) when it's time to land, which obviously makes life difficult .

And if I were an actually good pilot, I could probably manage far more speed out of it than 1900 m/s at 34 km before kicking in the rockets. This could have easily been improved by having working action groups, gone up into space with open intakes. For some reason, my intake action groups did not work.

Using RAPIERs, 1900m/s at 34km is probably about as good as it's going to get; that's getting close to their maximum speed. I don't generally use turbos in combination with RAPIERs, so you might be able to push it just a bit further, but I don't think it'll make much difference. There's a known bug where action groups can misbehave for symmetrically placed parts if you move things around. Always double check the action groups are mapped correctly after you've modified the plane - be especially careful to check that all parts in a symmetry group are mapped to the action group and not just one.

I also need to figure something out to prevent overheating of my rapiers.

-snip-

Other than that I added some batteries to right in front of the RAPIERS (in the hopes of solving that overheating problem, but alas, didn't work), and each engine now has AT LEAST 4 shock cones servicing it. It's still not as smooth as I like it, but yeah.

You probably have the RAPIERs too close together. In particular they overheat if you put them on stack quadcouplers. I don't know about tricouplers, but they should be okay on bicouplers. You can try adding parts like struts directly to the engines to help with the overheating, but it's more of a bandage than an actual fix

I'm still having some minor yaw-issues when climbing just after takeoff though. Anyone know how to solve that? Bigger vertical stabilisers? More dihedral on the wingtips?

It might help if you posted your updated design, but bigger vertical stabilizers will probably help assuming the problem is not that you're being overly aggressive with the ascent

[capi3101]

They'll blow on tri-couplers; that happened to me with my own "lift a full orange tank to orbit" spaceplane design...

I also had the overheating problem. I found, though, that the plane was still able to make orbit even two marks back on the throttle, so I didn't attempt to do anything to correct it.

Edited January 15, 2015 by capi3101

[Anquietas314]

http://i.imgur.com/shTk6cq.png

O.o that's an insane amount of reaction wheels you have there... why? You shouldn't need anywhere near that many to keep it controllable in space, so I can only assume they exist to counteract a serious aerodynamic flaw.