Why is the estimated time for NERVA burns aways so wildly off?

[johnnyhandsome]

Ok, I've managed to build a launcher system that can put a big orange tank with two or three nerva engines, along with 20 ton rover and landing system, in orbit. So I plot long range trips to Jool or Dres and am given an estimated time for the length (via the nav ball). Yet, once I start the burn, from 80-100km in Kerbal orbit, the burn always lasts much longer, sometimes as much as twice as long as the original estimate.

Am I doing something wrong here?

[Sof]

The nodes work on the presumption that you are using the last engine fired. What tends to happen is that you use your first few stages with powerful engines to get yourself into orbit, then dispose of them.

At this point you create the node for Jool or where ever before firing the NERVA rockets.

The game still thinks you are using the really powerful engines you used to get into orbit. The NERVA engines have a much lower thrust so the estimated burn time is always much lower than the actual burn time that appears a few seconds after activating the NERVAs.

[soranno]

for some reason I was expecting you were going to say the opposite, that they are always largely over-estimated.

That is the more common occurance for me. Often a node will be estimated at burning for 4 and half minutes say, and once I light the engines it will drop to 2 and a half.

I`ve learned to estimate with my commonly used designs, but it is mostly worthwhile to light the engines for a second ahead of time to let the system adjust and give a proper estimate

[Festivefire]

yeah, i have the same problem, maneuver nodes are really bad at picking a correct burn time. just give those engines a one second pulse of full throttle and the burn time should re-calibrate to a more or less correct number.

[jheriko]

I find generally that the burn times are always off to some degree, even during a burn. The most noticeable is when doing long partial ion burns, say at 25%, and you can watch the timer countdown 1 second for every 4 seconds of real-time. Before starting the burn it is often much more wildly out... I've also seen the burn times be overestimated when using nodes during launch or in low orbit, when I still have super powerful engines attached.

My guess is that as well the time being based on full throttle, it doesn't account for mass loss from fuel either.

[Mokmo]

(And now i understand why it was a bad idea to change that stage right before running the node...)

[Luis]

Ultimately, this inaccuracy must be something that has been added deliberately for gameplay reasons. The total thrust, mass and ISP of all the engines on the current stage is known precisely by the game. If Squad wanted to provide an accurate figure for the burn time for a node, they could. I don't really see the point of not doing so, but they obviously feel it's a good ideafor some reason. Perhaps Mechjeb or KER could be upgraded to provide more accurate burn time estimates.

[Akalaël]

Actually MJ is quite accurate. As I recall, I was yelling at it at first times : "why do you start the burn now ? Can't you see it's a 5min burn, and you start at 7min ?"

And then noticed the estimate time have changed.

[rodion_herrera]

Don't you notice that MechJEB always divides the total time of the burn, and then starts the burn half the time to the "zero" mark, to compensate for the length of the burn? For instance, if the burn is to be 60 seconds, MJ will start the burn in -30 seconds to offset the +30 it has to do, thus completing (and balancing) the 60 second burn.

[MR4Y]

I learned to do 50/50 burns by myself, strangely enough.

[Tristavius]

My guess would be it calculates the burn time for if you were able to do the burn with the perfect heading, the same weight as when you started and with the same gravity as that one particular point. Taking into account the most efficient time to push your apoapsis out is at the periapsis, and you are only smack bang at your periapsis for one moment in time.

No burn is ever going to be quite like this, especially long ones where you may start many degrees further round the planet from where your node is, and continue for many degrees after - this will change the overall efficiency of the burn. Doing a 10 minute burn with 5 on either side of the node is going to cost a bit of extra delta-v as your going to start and end the burn in very different points of the orbit and will thus need a bit of extra burn time compared to the calculation. You of course also expend fuel making yourself lighter as you go which will cause you to need less burn time, though with NERVAs the weight reduction is likely to loose you a lot less burn time than the need for long burns adds on. Larger engines may end up the opposite, not sure - lot of variables.

[foamyesque]

Ok, I've managed to build a launcher system that can put a big orange tank with two or three nerva engines, along with 20 ton rover and landing system, in orbit. So I plot long range trips to Jool or Dres and am given an estimated time for the length (via the nav ball). Yet, once I start the burn, from 80-100km in Kerbal orbit, the burn always lasts much longer, sometimes as much as twice as long as the original estimate.

Am I doing something wrong here?

In your particular case, it's because you're using NERVAs and therefore have a long burn time required. Since you're not able to burn at or near the optimum point for much of it (which the node is assuming you do), your burn becomes much less efficient, and therefore necessarily longer.

In addition, the burn time estimates are based on your last thrust, and do incorporate the mass changes as you burn through your current stage, but do not account for staging events during the burn.

So if your last burn was a trickle of thrust on your main engines, you'll see very long burn times estimated until you throttle all the way back up.

[johnnyhandsome]

Ah, thanks y'all!

[Eric S]

Ultimately, this inaccuracy must be something that has been added deliberately for gameplay reasons.

No, they just use the acceleration of the last burn you did rather than compute what you can do now. It's simpler to do, and in the case of a ship that would break apart under full throttle, possibly more accurate. Yes, it's almost never really precise and is often horribly inaccurate after staging.

[johnnyhandsome]

"Ultimately, this inaccuracy must be something that has been added deliberately for gameplay reasons."

If true, it makes no sense, as being able to program in burn times and amounts has been possible for decades in real spacecraft. Is there anything more boring than having to watch the clock for a burn, other than roving around Eve?