fuel flow problem - bug?

[kurja]

So, I put some liquid boosters on my rocket, and they seem to suck fuel from tanks they're not connected to. Craft file attached, all stock parts. Look at the booster rocket on the "near side", closest to the camera, it has flt 800 and 400 tanks above an lv-t30 engine, and on the solid booster next to it, there's one more flt 400 tank from which a fuel line comes to this stack. There are no other fuel lines coming to or from these tanks. In flight, the engine runs but fuel is not used from any of these three tanks - it seems that fuel is instead drawn from the center orange tank, which in turn does -not- give fuel to the center stack engines, they only get fuel from the thin tank directly above them.

I've spent the last 20 minutes trying to see what I've done wrong... But I can't find anything, please try this out and see if you get the same, or if I just missed something.

craft file here http://ubuntuone.com/5VXeuw8Bdu3hxC76FGcyFf

Edited August 25, 2013 by kurja

[WhiteWeasel]

Can't download anything right now, a screenshot would be helpful.

[Johnno]

please try this out and see if you get the same, or if I just missed something.

Two of your boosters are missing fuel lines leading to the core tank. If you have '4-symmetry' so that each FL-T400 ontop of a SRB feeds into the liquid booster and the liquid boosters into the core then it all works nicely.

[sicarius]

To address your main problem, when I flew it the center orange tank wasn't draining from the external tanks, so it could be a bug. Off-topic, I would suggest dropping the srbs as their own stage because right now they run out before the liquid boosters, meaning you have to carry the srbs until your liquid boosters run out.

EDIT: while what Johnno says is true, because your core tank ignites after you drop your boosters, I believe presence of fuel lines shouldn't affect fuel flow, in other words, they don't do anything

Edited August 25, 2013 by sicarius

[Johnno]

I guess we're confused on what the intended design/result it supposed to be. There's no point in feeding the core tank from the outer tanks unless you light the core engines, in which case those fuel lines would be redundant and are actually causing the inconsistency. If we go with what I thought was the intended result, as in feeding the core from the outside and dropping the boosters when they're empty, then you need to add two fuel lines.

As it was 'originally', you have two boosters feeding the core, if you light the core engines the core will feed from those tanks. However the two last boosters will feed from the furthest tanks from them, being the thin core tank. Hence their own tanks won't be used.

So in this particular case, they're working just as intended, it's just a poor design. If you're going to have boosters AND core engines, you should either light everything and have them independent of eachother, with no fuel lines, OR you feed from the boosters into the core with all engines running, dropping the boosters once dry.

kurja's design is actually quite a smart asparagus style setup, if that was the intended result, which I thought it was. In which case you need the fuel lines that are missing. 4x SRB liquid tank -> liquid booster -> core, light all engines at start.

[kurja]

Hello Johnno & Sicarius,

it was not supposed to be completely symmetrical (four way symmetry). I didn't post this for design tips, but just because it doesn't seems to work like it should and I thought it could be a bug in the game. Anyway, it was my intention to drop the two pairs of liquid boosters separately. Tanks on top of solid boosters are "drop tanks", anything in them is supposed to have been spent by the time the solid boosters run out, which is the first stage to be jettisoned.

Johnno, did you try launching the rocket? The center engines do light up at launch.

Issue is, that I have engines running that do not draw fuel from any tank they're connected to. Are you seeing the same? No?

[Sirine]

The fuel tank T-400 x4 (above the T-800) are "damage", replace it and everything should work as it is.

[Johnno]

Johnno, did you try launching the rocket? The center engines do light up at launch.

Issue is, that I have engines running that do not draw fuel from any tank they're connected to. Are you seeing the same? No?

Ah, sorry for jumping to conclusions, I didn't mean to give 'design tips' as such, I wrongfully assumed what the intended design was supposed to be, my apologies.

As the staging was set up in the file you have available for download, the center engines weren't staged to light up on launch. I did change that before launching (another assumption, yay).

The fuel logic is a bit confusing and you need to understand how it works to fully utilize it. Radially placed engines look first to the center stack, and draw fuel from whatever's connected to the center stack first, before their own fuel. They don't even have to be connected, which is why your design works the way it does.

The two boosters that aren't connected to the center stack look to the center stack and first draw fuel from the two other boosters. Once those are dry they will draw fuel from the tank furthest away on the centerstack, the bottom small tank. After that they'll draw from the orange tank. Not until the orange tank is dry will they use their own fuel.

This is all enabled once you place fuel lines mind you, if you were to remove the fuel lines from the two boosters that aren't connected to the center stack then they'd use their own fuel and burn out.

[kurja]

The fuel tank T-400 x4 (above the T-800) are "damage", replace it and everything should work as it is.

Thanks, I'll try that. But how do you mean, "damage"? Is that something you see in the craft file, I can't find string "damage" in .craft?!?

[Sirine]

Thanks, I'll try that. But how do you mean, "damage"? Is that something you see in the craft file, I can't find string "damage" in .craft?!?

Well, Game Bugs.

Sometimes if you put the fuel tank, detach and put it to another connection it will become crazy. So, most of the time now, I will get the component fresh from the component list (left side). And not reuse it.

[kurja]

Well, Game Bugs.

Sometimes if you put the fuel tank, detach and put it to another connection it will become crazy. So, most of the time now, I will get the component fresh from the component list (left side). And not reuse it.

Yay, that did the trick, thanks!

[kurja]

Ah, sorry for jumping to conclusions, I didn't mean to give 'design tips' as such, I wrongfully assumed what the intended design was supposed to be, my apologies.

As the staging was set up in the file you have available for download, the center engines weren't staged to light up on launch. I did change that before launching (another assumption, yay).

The fuel logic is a bit confusing and you need to understand how it works to fully utilize it. Radially placed engines look first to the center stack, and draw fuel from whatever's connected to the center stack first, before their own fuel. They don't even have to be connected, which is why your design works the way it does.

The two boosters that aren't connected to the center stack look to the center stack and first draw fuel from the two other boosters. Once those are dry they will draw fuel from the tank furthest away on the centerstack, the bottom small tank. After that they'll draw from the orange tank. Not until the orange tank is dry will they use their own fuel.

This is all enabled once you place fuel lines mind you, if you were to remove the fuel lines from the two boosters that aren't connected to the center stack then they'd use their own fuel and burn out.

Oops, the center stage was meant to ignite at launch but apparently I cocked up the staging in that save, sorry about that

About the highlighted part. I don't understand, that really doesn't seem to be the case. Make a rocket like this, and you'll see that each engine uses fuel from it's own tank:

[Johnno]

About the highlighted part. I don't understand, that really doesn't seem to be the case. Make a rocket like this, and you'll see that each engine uses fuel from it's own tank

You missed this part in my reply: This is all enabled once you place fuel lines mind you, if you were to remove the fuel lines from the two boosters that aren't connected to the center stack then they'd use their own fuel and burn out.

So yes, the design you have there works fine. However, as your boosters were designed:

The standalone LV-T45 isn't drawing from the T400 tank above the SRB, nor its own stack. It's drawing from the X200-8 tank at the bottom of the center stack.

I did a bunch of extra testing and actually the solid booster being on the decoupler is the big culprit. Move the liquid tank onto the decoupler and the SRB under it, and things work.

So yeah, the fuel logic is not only confusing, but it confuses itself at times. Which is why it's so much better to stick to simple things like asparagus or onion feeding.

Edit: Btw I'd highly recommend you swap that skipper for a poodle. Saves mass and adds efficiency.

Edited August 25, 2013 by Johnno

[kurja]

You missed this part in my reply: This is all enabled once you place fuel lines mind you, if you were to remove the fuel lines from the two boosters that aren't connected to the center stack then they'd use their own fuel and burn out.

Ah right. I saw that line but couldn't get over that engines would draw from tanks they're not connected to, because... well, that just shouldn't happen.

The standalone LV-T45 isn't drawing from the T400 tank above the SRB, nor its own stack. It's drawing from the X200-8 tank at the bottom of the center stack.

...

I did a bunch of extra testing and actually the solid booster being on the decoupler is the big culprit. Move the liquid tank onto the decoupler and the SRB under it, and things work.

So yeah, the fuel logic is not only confusing, but it confuses itself at times. Which is why it's so much better to stick to simple things like asparagus or onion feeding.

That's... confusing is a much too mild word.

Edit: Btw I'd highly recommend you swap that skipper for a poodle. Saves mass and adds efficiency.

Yea, I got it working like I wanted to but could use just a bit more efficiency, working on that now...

edit - I'll repost this in the bug forum anyway, silly fuel logic is a known issue but engines getting fuel from tanks they should have no access to whatsoever is a bug.

edit 2 - aaand I just realised that radial decouplers are fuel crossfeed capable, unlike stack decouplers. Did not know that before...

Edited August 25, 2013 by kurja