Asparagus set-up, fuel tanks draining unevenly

[odeonshiva]

Hello everyone,

I've been using the Asparagus staging method for my rockets in KSP for quite some time and it has always served me well. However, after getting back to KSP yesterday after a pretty long break, I keep running in to a strange occurrence.

Currently my rocket has four stacks of fuel-tanks+engines symmetrically placed around the center stack. Attached to each of those are two more symmetrical stacks. Totaling four stacks in the first ring, eight stacks in the second ring. Below I have posted a picture of how I hooked up my fuel lines. The red lines are where I connected the decouplers. Now, I double checked this about 20 times, there's no loops or double fuel lines. Also I rebuilt the entire rig from scratch about four times so far. The setup is as it shows in the picture. All stacks have the same amount of fueltanks and the same engines.

The problem is that when I launch my rocket, one side of the asparagus drains slightly faster. Resulting in a massive wobble and a deadly crash. Can someone please tell me what I'm doing wrong? I'm going crazy quadruple-checking every fuel line. Is there perhaps a problem with this specific asparagus setup? What could cause one side of the asparagus to drain half a second faster?

Paint picture made by me to indicate my asparagus setup.

I really hope you can help me, thanks in advance!

Edited September 6, 2013 by odeonshiva

[luchelibre]

It'd be far better if you could provide the craft file for us to test.

[pluto101]

Make sure during launch you're not encountering any structural failures in your tanks, and that your engines aren't overheating. Double and triple check your fuel lines. Sometimes Control-Z can cause fuel lines to disconnect. Your design, if the picture above is accurate, should work. Things like this can happen if you have a fuel line loop, which is easy to accidentally introduce when you're setting up an asparagus system.

[Amoun]

Which side and at what stage?

Or does that start from launch?

Post a few screenies and yeah, if its a stock craft, post the craft file, maybe we can check it out.

But that is odd....didnt have that happen before. Only thing I would think is fuel consumption.

ANOTHER thing, is make sure you attach the fuel tanks together. The only similar experience I had, was when I had the fuel tanks dis-attach while in flight. That resulted in uneven fuel consumption but I could not tell because the rocket was still pushing on the fuel tank so it LOOKED like it was still attached to the whole craft. As soon as I put struts on them, it was fine.

[Masked Turk]

I've noticed that where you attach your pipes is important. Fuel seems to drain downwards as it is burnt, so placing pipes at the top will result in less available fuel for that engine.

I.e. if your pipes are connected at different heights to their fuel tanks, your engines will have different amounts of fuel to work with.

So: try and place the fuel pipe connections to your tanks at the exact same levels.

[SRV Ron]

Asparagus staged rockets are tricky to brace especially when you use a design as complex as you have diagrammed. Massive wobble is not from uneven drainage. Uneven drainage would cause your rocket to tip over. Massive wobbling is from the stress of acceleration and changing weight and occurs when there is insufficient bracing to counter those forces. Better to design onion stage a dual ring setup. Much easier to brace the tanks against the stresses of launch.

Edited September 6, 2013 by SRV Ron

[KerbMav]

I've noticed that where you attach your pipes is important. Fuel seems to drain downwards as it is burnt, so placing pipes at the top will result in less available fuel for that engine.

I.e. if your pipes are connected at different heights to their fuel tanks, your engines will have different amounts of fuel to work with.

That will be the first time I read about such a thing.

[SRV Ron]

That will be the first time I read about such a thing.

While pipes connected at the top of tanks work OK, they often get disconnected upon reloading the ship into the VAB. That will cause endless problems of uneven or no fuel transfer during flight.

[Synapse]

You pressed undo at some point and one of your fuel lines isn't connected. Re-check the whole thing, line at a time.

[KerbMav]

While pipes connected at the top of tanks work OK, they often get disconnected upon reloading the ship into the VAB. That will cause endless problems of uneven or no fuel transfer during flight.

Which is not what Masked Turk wrote and I commented.

[allmappedout]

It will be a case of a fuel line either not loading or being disconnected in the VAB. carefully check the line flow, or alternatively take all the fuel lines off and put them on again. Make sure there are no fuel loops. The height of the line makes no difference on the fuel consumption.

[ThirdHorseman]

I am seeing a similar issue. Here is my config:

Red lines are couplers, arrows show fuel flow. As expected, tanks #3 drain first, as they are feeding into two other tanks. I drop them, and then I would expect tanks 2a and 2b to drain at the same rate and drop at the same time so I have the next stage activate the couplers for both 2a and 2b. However this does not happen, as 2a drains faster and when I drop both 2a and 2b, 2b goes flying up since they still have fuel and the engines are still active. So I change my staging, to have 2a drop first, and then 2b soon after.

It's not that simple, though. I launch again, but this time it's 2b that drains faster, and there is another big explosion. So what I end up doing is adjusting my staging on the fly, depending on which tank drains faster.

For example, on one launch tanks 2a will drain faster, and I'll drop them and then 2b right after. However, if I Revert to Launch and then re-launch, the next time tanks 2b will drain faster which means I have to scramble to adjust my coupler staging so I don't drop the wrong ones. I am not going back to the VAB, so there isn't any change in fuel lines. Launch 1, 2a drains first. Revert to Launch. Launch 2, 2b drains first.

Very strange behavior...and I'm pretty sure my fuel lines are all pretty close to the same height at the top of the tanks. Not sure if that has any impact, but since it was mentioned I figure I'll point that out.

[sojourner]

^Classic fuel loop problem. Don't go from one tank to two tanks to one again.

I've noticed that where you attach your pipes is important. Fuel seems to drain downwards as it is burnt, so placing pipes at the top will result in less available fuel for that engine.

I.e. if your pipes are connected at different heights to their fuel tanks, your engines will have different amounts of fuel to work with.

So: try and place the fuel pipe connections to your tanks at the exact same levels.

The height of the connection makes absolutely no difference on the fuel feed of the rocket. Unless of course you've accidentally put the connection above/below a decoupler when you didn't mean to.

Also, who is having problems with fuel lines when going back to the VAB? I have never encountered such an issue. Now, subassemblor will mess up fuel lines when loading a saved assembly if your not careful, but that's the closest I've seen to that phenomena.

Edited September 6, 2013 by sojourner

[odeonshiva]

First of all, thanks for the many replies. This community is amazing. With your help I have solved the problem with my craft. After that I did a little bit of testing.

^Classic fuel loop problem. Don't go from one tank to two tanks to one again.

The height of the connection makes absolutely no difference on the fuel feed of the rocket. Unless of course you've accidentally put the connection above/below a decoupler when you didn't mean to.

Also, who is having problems with fuel lines when going back to the VAB? I have never encountered such an issue. Now, subassemblor will mess up fuel lines when loading a saved assembly if your not careful, but that's the closest I've seen to that phenomena.

After reading your replies I tried a couple of things and discovered that the height of the connection can make a difference. It can result in strange discrepancies with the fuel drainage in your entire asparagus setup. However this only occurs when you're working with stacked fuel tanks.

In my case I had three grey medium sized rockomax tanks on top of each other to avoid overheating problems. I had consistently been placing a single fuel line on one side of the asparagus between the middle fuel tank on stage 2 to the bottom fuel tank on stage 1. I messed up the camera angles, making it hard to see. Somehow that uneven connection left that side of the asparagus with a little less fuel than the other side, wobbling my rocket and killing my kerbal friends. I'm still not quite sure what's best; connecting the fuel lines on top, middle or bottom of your stack. If I've got time to do some more experimenting I'll let you guys know. Cheers again for your replies.

[ThirdHorseman]

^Classic fuel loop problem. Don't go from one tank to two tanks to one again.

Any particular reason, besides just normal KSP bugs? I don't see any reason why the fuel shouldn't flow evenly regardless of how many tanks you are feeding into? Is this a known KSP issue?

[sojourner]

Any particular reason, besides just normal KSP bugs? I don't see any reason why the fuel shouldn't flow evenly regardless of how many tanks you are feeding into? Is this a known KSP issue?

Yes. It's a well known KSP issue. It's a problem with how the fuel flow logic is coded from what I have heard.

After some testing I need to amend my statement about fuel line heights. The height does not make a difference in overall drainage of fuel in a stack of tanks. It does, however, affect which tanks in that stack get drained first.

if your asparagus stage has 3 tanks and you attach the fuel line to the middle tank, it will drain the upper first while also pulling fuel from the lower tank at a reduced rate. Once the upper tank is empty fuel will be pulled from the middle tank while fuel continues to be pulled from the lower tank. Both of these tanks deplete at the same time.

If you attach the fuel line to the lower tank fuel drains from the top down. upper tank empties then middle, then lower.

Here's the weird part. If you attach it to the upper tank it behaves much like connecting to the middle tank did above. The only difference is the amount of fuel drained from the lower tank while the upper tank is draining is much more reduced. once the upper tank is empty, the middle and lower tanks both finish at the same time.

In conclusion, it would appear that for best COM during flight, to attach your fuel line to the lowest tank in an asparagus stage for draining purposes. Oh and of course, use symmetry. If you have a pair of asparagus stages draining from different tanks, well, you won't be going to space today.

Edited September 6, 2013 by sojourner

[odeonshiva]

After some testing I need to amend my statement about fuel line heights. The height does not make a difference in overall drainage of fuel in a stack of tanks. It does, however, affect which tanks in that stack get drained first.

if your asparagus stage has 3 tanks and you attach the fuel line to the middle tank, it will drain the upper first while also pulling fuel from the lower tank at a reduced rate. Once the upper tank is empty fuel will be pulled from the middle tank while fuel continues to be pulled from the lower tank. Both of these tanks deplete at the same time.

If you attach the fuel line to the lower tank fuel drains from the top down. upper tank empties then middle, then lower.

Here's the weird part. If you attach it to the upper tank it behaves much like connecting to the middle tank did above. The only difference is the amount of fuel drained from the lower tank while the upper tank is draining is much more reduced. once the upper tank is empty, the middle and lower tanks both finish at the same time.

In conclusion, it would appear that for best COM during flight, to attach your fuel line to the lowest tank in an asparagus stage for draining purposes. Oh and of course, use symmetry. If you have a pair of asparagus stages draining from different tanks, well, you won't be going to space today.

Thank you for posting your findings. What you say seems to be true and a very good thing to keep in mind when building rockets.

[Starglider]

Thank you for posting your findings. What you say seems to be true and a very good thing to keep in mind when building rockets.

What works for me, is this process:

1) Add clamps to the rocket (if you haven't already), and stage the rocket so the clamps don't release on "launch", but that the engines activate (on the lowest layer, that is). Remove all fuel lines (again, on the lowest layer).

2) Send it to the pad, and test-burn the engines. You get a good idea of the rocket's ability to handle heat, as well as whether the rocket's structure is stable enough. When the rocket settles down during the burn, revert to the VAB.

3) Add the first pair of fuel lines that will form your asparagus system, and re-test as above. This time, you're looking for pairs of engines to burn out together. This is actually easier with mainsails burning hot; as they burn out, the heat bars should drop away in pairs.

4) If there is uneven burning, you know which fuel line pair the problem is in.

5) Add the next pair of fuel lines and re-test the rocket with the new pair

6) Repeat #5 until your asparagus system is complete

7) Return to the VAB to work on staging. If you need to, you can test the asparagus staging in a similar manner, but staging issues are more obvious than fuel line issues so you probably won't have to.

8) Now that your asparagus system is working, you can continue on from there with confidence.

[running]

I've been suffering similar effects lately due to a rocket design that was just to unstable and kept shaking things apart.

F3 would show the link between fuel tanks being broken during launch/flight.

Right clicking on the specific fuel tank wouldn't work any more as it was still on the rocket but not connected anymore.

[SRV Ron]

I have found that when you stitch tanks together that are already attached in a ring using symmetry, not all the braces will go where they are suppose to. The fix is to build the tank stack in the center, stitch with symmetry, then take a copy of that tank set to use for the onion or asparagus stage. For asparagus, start with a pair from the set copied from the stitched center core, brace to the center core, and get a copy by selecting the tanks at the decoupler including the decoupler. When pasted as the second or third set, they should connect with all braces in their proper location.

You will know when you have got it right as the design will be stable during flight.

[Starlionblue]

Sorry abou the necrothread.

I'm having a similar weird issue with uneven tank drainage. Shown in the pic is my as yet unfinished Eve lander. It's a "layer cake" style asparagus with engines in all three layers firing at the same time (except center engines of course). Basically the asparagus starts by draining the large tanks at the bottom in pairs, then goes into the bottom ring and drains pairs, then joins in the bottom inner tank, then splits in two again to go up one level to the outer middle boosters, goes around the ring, joins in the middle, then does the same for the top level.

Here's the weird bit. One of the "first" pair of tanks in the middle layer (first tanks to be used up and jettisoned in middle layer) strats draining at launch. Even more weird is that it doesn't drain quickly as if it's not getting any fuel from upstream. It drains quite slowly but still noticeably, so there is fuel from going "through" it.

I have checked and double checked the fuel lines. I have rebuilt the entire stack from scratch (three times). I have tried different configurations. I have painstakingly placed the fuel lines in the exact same positions. I have tried doubling up fuel lines. I have tried a quadsparagus instead of a twinsparagus. I have yelled at the PC and invoked dark deities. I have drunk beer and I have made diagrams. None of that has worked.

"Solution": Remove the top layer from the asparagus, making it a "traditional" self-contained upper stage that only fires when the final middle booster separates. The tanks in the middle layer now drain evenly.

To be honest I don't lose a huge amount of delta-V from implementing the "solution", and the booster is already ridiculously efficient (I'm on track to make an Eve ascent vehicle weighing around 260 tons). It does bug me though, and I am pretty convinced that this actually is a bug.

Any ideas appreciated.

 

Edited February 29, 2016 by Starlionblue