Keep some engines off at takeoff?

[cwalrus]

Some times I find myself with asparagus setups that seem to indicate if I keep the center engine OFF during takeoff, my final Delta-V is better. This seems counter-intuitive to me - I figure, if it's pointing down, turn it on! Esp. if it's the last in the chain of the asparagus.

Why do my Delta-V stats say otherwise? Any ideas?

thanks!

Edited June 16, 2015 by cwalrus

[Arkalius]

It's probably got lower Isp than other engines. Having it off lets you consume more fuel with more efficient engines first, thus improving total delta-v. However if this gives you a poor TWR, then it may end up costing you efficiency in the end.

[Superfluous J]

Yeah there are engines you should run in atmo, and engines you should run in space (Generally the divider isn't "space" but 10-12km up). Ideally, you should never run "space" engines in the lower atmosphere.

You can find these engines by checking their stats in the VAB. If the Isp range is wide and really low on one end, don't use them in air. If the Isp values are close together you can use them anywhere, but they're better as atmospheric boosters than space.

[Teilnehmer]

Maybe, you exceed terminal velocity in lower atmosphere with all the engines on, and the drag kills the ÃŽâ€v?

[Snark]

It also depends on what tool you're using for calculating your dV-- how confident are you that it's correctly and accurately computing your dV?

After all, if your "intuition" is telling you that you should be getting better dV with it on, and your tool is saying otherwise, then the simplest thing to do is to back up your intuition and just do the calculation yourself to verify. That will either tell you that your tool is wrong, or else you'll have an "ohhhh, that's why" moment.

[Teilnehmer]

It also depends on what tool you're using for calculating your dV-- how confident are you that it's correctly and accurately computing your dV?

The more fuel the more ÃŽâ€v for the same vehicle. So you can just compare fuel left, can’t you?

[KerikBalm]

Yeah there are engines you should run in atmo, and engines you should run in space (Generally the divider isn't "space" but 10-12km up). Ideally, you should never run "space" engines in the lower atmosphere.

You can find these engines by checking their stats in the VAB. If the Isp range is wide and really low on one end, don't use them in air. If the Isp values are close together you can use them anywhere, but they're better as atmospheric boosters than space.

Yea, shut off a poodle/LV-909 on launch... why have it burn fuel and produce almost no thrust?

Or use the aerospike on your core stage(s). Almost has the best atmospheric dV, almost has the best vacuum dV (for a non nuclear/ion rocket)

[Tex_NL]

The more fuel the more ÃŽâ€v for the same vehicle. So you can just compare fuel left, can’t you?

This is pretty much true as long as the fuel amount is the only changing variable.

Unless your final orbital stage is nuclear and you're left with oxidizer. Any resource you won't need in the future should either be dumped or used ASAP.

[Snark]

The more fuel the more ÃŽâ€v for the same vehicle. So you can just compare fuel left, can’t you?

Yes, more fuel = more dV. However, the actual amount of dV is highly dependent on your setup. If you're running multiple engines of different types at the same time, you have to take into account their varying Isps and their relative rate of fuel consumption. If you're calculating dV across multiple stages, you need to calculate each stage's dV separately and then add them together. It's not rocket science (well, okay, I guess technically it is)... but it's fairly straightforward algebra. (Except for the thorny problem of engines whose Isp varies a lot from Isp to vacuum, usually have to do some sort of approximation there, though that shouldn't matter too much as long as you don't do something ill-advised like using a vacuum engine at liftoff.) But you do have to put the right numbers in the right places, which requires understanding your ship configuration.

The OP said,

...that seem to indicate if I keep the center engine OFF during takeoff, my final Delta-V is better...Why do my Delta-V stats say otherwise?

This clearly indicates that he's using some kind of tool to tell him what his dV is. He doesn't specify the tool (I'm guessing KER, that's the one I hear everyone talking about), but there's the open question of "is the tool smart enough to do the right thing for his particular setup"? e.g. is the tool actually reporting correct numbers, or is it possible that it got confused by a complex rocket setup (such as asparagus) and isn't actually doing the correct math?

Thus the suggestion to work it out manually and see what numbers come out, and compare that against what the tool says.

[cwalrus]

Thanks - all good points.

I'm using Mechjeb's Delta V stats, which i trust completely (shouldn't I?)

The one time my 'intuition' is kicking in is if I have 3 stages coupled radially, all with Skippers or Mainsails. In that case, pretty much the same engine, right? So you would think that having all of them on would produce the biggest TWR to get you out of the atmosphere the quickest. However, it seems consistent that if I move the center engine up in the stages (i.e. towards stage 2) and keep it off until later, it actually increases my Dv. (wait... trying to find the delta on my keyboard...AHA! Option J! ) This is with everything asparagus'd . You know what... let me check that....

∆∆∆∆∆∆

∆∆∆∆

∆∆

∆

Edited June 16, 2015 by cwalrus

[justidutch]

I just discovered this on the weekend too; I created an SSTO rocket with four Mammoth engines surrounding one Rhino. Keeping the Rhino off until higher up in the atmosphere saved some DV. Shutting the Mammoths off prematurely saved even more, even though my TWR took a big hit. I don't bother with all the math, but common sense and a couple test flights indicate that this is logical.

[Michael R]

It's hard to know what exactly is going on with your rocket without more info (or, especially a picture, or even a craft file), however, I think it's safe to say that if you decide you're better off not using certain engines at launch, you might want to consider whether you need them at all. If you are sure you'll need them later, then it's probably best to put the launch stages below them, rather than radially, to the extent possible, to minimize drag. The fewer the stalks of asparagus, the lower the drag.

Before 1.0, I believed in a design philosophy that said that as many of your engines as possible should be firing at launch, or you were hauling unnecessary weight and lowering your efficiency. This lead to a lot of asparagus stages with a lot of engines. With the changes since 1.0, both aero and engine ISP changes, I'm really seeing the benefits of designing with vertical staging rather than horizontal. And my rockets look better, too.

[Snark]

I just discovered this on the weekend too; I created an SSTO rocket with four Mammoth engines surrounding one Rhino. Keeping the Rhino off until higher up in the atmosphere saved some DV. Shutting the Mammoths off prematurely saved even more, even though my TWR took a big hit. I don't bother with all the math, but common sense and a couple test flights indicate that this is logical.

In this case it's pretty clear. The Rhino is a vacuum engine; its atmospheric Isp is abysmal, so if you're running it at takeoff, you're wasting lots of fuel. Best to wait until you're above 10-12 km before turning it on.

The original poster had a different problem-- he was talking not about what he actually observed the rocket doing, but what MechJeb says the dV is. I've never used MechJeb, but it seems to me that this would be a kind of hard problem-- if you have (for example) the issue of "I've got a vacuum engine, I should wait until I'm at higher altitude to turn it on," does MechJeb know what altitude you're going to be turning it on, and take that into account when it's giving you dV calculations? That's a pretty tall order.

[Red Iron Crown]

The original poster had a different problem-- he was talking not about what he actually observed the rocket doing, but what MechJeb says the dV is. I've never used MechJeb, but it seems to me that this would be a kind of hard problem-- if you have (for example) the issue of "I've got a vacuum engine, I should wait until I'm at higher altitude to turn it on," does MechJeb know what altitude you're going to be turning it on, and take that into account when it's giving you dV calculations? That's a pretty tall order.

I'm not clear whether the OP is saying that or saying that MJ is reporting more dV remaining after achieving orbit.

[Snark]

The one time my 'intuition' is kicking in is if I have 3 stages coupled radially, all with Skippers or Mainsails. In that case, pretty much the same engine, right? So you would think that having all of them on would produce the biggest TWR to get you out of the atmosphere the quickest. However, it seems consistent that if I move the center engine up in the stages (i.e. towards stage 2) and keep it off until later, it actually increases my Dv.

So, a couple of things:

1. TWR has nothing to do with dV. The math for dV depends only on the mass ratio (wet/dry) and on the Isp of the engines involved. TWR is completely irrelevant.

2. TWR is relevant to how much dV you need to accomplish a given task (such as getting to Kerbin orbit or whatever).

3. However, "higher TWR" does not automatically mean better. If no atmosphere is involved, then yes, you'd be better off with a higher TWR if doing so doesn't add dry mass to your ship (i.e. "if it points down, turn it on".) However, it's a bit more nuanced where atmosphere is involved: when you take off, you don't want to go too fast too soon, or you pass Mach 1 and your aerodynamic drag shoots way up and you end up wasting too much fuel trying to punch a hole in the sky. So you typically wait until you're reasonably high up (over 10-12 km) before really flooring it, and at that point you're already well into your gravity turn and TWR isn't as important for conserving dV. Essentially, as long as you have "enough" TWR at takeoff (around 1.5 is plenty, in my experience), having extra doesn't necessarily help you.

Can you exactly describe your asparagus setup? You mention Skippers and Mainsails attached radially; what's the one in the center that you're turning on (or not), which seems to affect your dV?

[Kryxal]

If you move it up in stages, it's not producing thrust until you have a lighter rocket, thus higher delta-v ... BUT you've reduced your TWR. If you can afford to do that with a sufficient TWR, it usually means you have too many engines / not enough fuel.

[Empiro]

Assuming an asparagus staging is properly set up, and all the engines have the same ISP, then not firing an engine only lowers your TWR. That's actually another thing that might be wrong -- are you sure the asparagus staging is properly set up? If not, then that could also be the cause of lower delta-V.

Otherwise, I concur with the others -- the likely explanation is that the TWR is too high. You don't need more than 1.6 for the first stage, and any more than 2.5 is likely to cause you to lose delta-V due to air drag.

[cwalrus]

Yes at this point I concur as well! I think it's calculating the terminal velocity factor. Basically this is a MechJeb question...

If I can find the exact set up I'll post.

Thanks!

[Laie]

Oh boy... the first answer was spot-on, I don't understand why this keeps going.

• rocket of mass X
  
• amount of fuel y
  
• ISP z
  

Put all three into an equation and out comes your delta-V. That's what MJ does (and KER too, for that matter). Any engine that lowers the average ISP will also lower the delta-V in the readout. It's really as simple as that.

Now, what you actually get out of it may be a different matter. If running that "inefficient" engine in addition to everything else will increase your TWR from very low to mediocre, it will almost certainly lead to a more efficient rocket launch overall. If your TWR is pretty decent even without that engine, you're probably better off if you don't ignite it until later.

[Kryxal]

If your TWR is pretty decent without it, a redesign might be a good idea ... lose a couple of engines and you're better off mass-wise.

[Wanderfound]

Yes, a Skipper surrounded by Mainsail boosters will report higher launch ÃŽâ€V with the Skipper off, because the Skipper has lower sea level isp (but higher vacuum isp). But the more important point (hinted at in the post above) is that if you can afford to leave the Skipper turned off then you have too many boosters. Drop at least one Mainsail.