[0.25] Engine Ignitor (Workaround for some bugs V3.4.1: Aug.31)

[Tyrador]

I've found out that there is no need for the Ullage, as when you rotate the craft for a while around 2-3 times in one direction the fuel will go to the bottom and up of the craft, due to Centrifugal force.

[NathanKell]

HoneyFox: I totally forgot that the latest EI bundles engine configs in the main install. For obvious reasons this conflicts *mightily* with any RF-based engine config that *also* adds EI modules and configs. Could I persuade you to go back to having them be a separate download?

Or, failing that, might it be possible for me to bundle a stripped-down version of EngineIgnitor with RF or RO?

[AbeS]

Shouldn't it work fine? because your configs say :final and EI ones don't

[HoneyFox]

HoneyFox: I totally forgot that the latest EI bundles engine configs in the main install. For obvious reasons this conflicts *mightily* with any RF-based engine config that *also* adds EI modules and configs. Could I persuade you to go back to having them be a separate download?

Or, failing that, might it be possible for me to bundle a stripped-down version of EngineIgnitor with RF or RO?

I think you can simply use !MODULE[ModuleEngineIgnitor] {} to remove any existing one before adding the one used by the default engine configuration of MFS. That's what i'm doing with my own local MFS' stock-alike engine cfg.

[HoneyFox]

I've found out that there is no need for the Ullage, as when you rotate the craft for a while around 2-3 times in one direction the fuel will go to the bottom and up of the craft, due to Centrifugal force.

It will. But this should be considered as an emergency operation if no other better solution is available.

Plus, this measure will make the engine easily go flame-out shortly after you ignite, because the fuel will get into an unstable state when the upper part of the fuel moves down and splash into the bottom part of the fuel. To avoid this, apply small throttle for the first several seconds and gradually increase the throttle.

[CoolBeer]

Love the mod, especially the ullage bit(which has bitten me quite a few times).

I do have a couple of suggestions/thoughts though:

-Set up a minimum thrust on the engines, this will make shutting the engine down a manual task, and running the engine at low thrust is easier as you don't need to fear accidentally shutting down the engine(wasting ignitors/hypergolics).

-No auto ignite when moving throttle, starting an engine is a major task and I want full control on when the start button is pressed.(This is a somewhat secondary point if the shutdown mechanics in the next point is implemented). (By major task I am referring mostly to the ullage mechanics).

-When engine stops(from whatever reason) shut it down right away. This makes starting the engine next time have the same startingpoint(engine is in shutdown, user have to press "Activate Enigne" or similar) also feeds into the previous point(and the first really, as if you manually have to shut down the engine it's off then and there).

I did download and compile the latest git version(commit bb489adb7cbaad9caf5a88ae8d7b3ddc157f6218 "Some edit for display.") and I'm happy to report that it seems the ModuleEnginesFX wrapper works.

One thing I noted though was that it will report "Engine State: IGNITED" even if it's out of fuel/oxidizer:

Keep up the good work, love the complexity added with restarting an engine in LKO.

-

Kolbjørn

[HoneyFox]

Love the mod, especially the ullage bit(which has bitten me quite a few times).

I do have a couple of suggestions/thoughts though:

-Set up a minimum thrust on the engines, this will make shutting the engine down a manual task, and running the engine at low thrust is easier as you don't need to fear accidentally shutting down the engine(wasting ignitors/hypergolics).

-No auto ignite when moving throttle, starting an engine is a major task and I want full control on when the start button is pressed.(This is a somewhat secondary point if the shutdown mechanics in the next point is implemented). (By major task I am referring mostly to the ullage mechanics).

-When engine stops(from whatever reason) shut it down right away. This makes starting the engine next time have the same startingpoint(engine is in shutdown, user have to press "Activate Enigne" or similar) also feeds into the previous point(and the first really, as if you manually have to shut down the engine it's off then and there).

I did download and compile the latest git version(commit bb489adb7cbaad9caf5a88ae8d7b3ddc157f6218 "Some edit for display.") and I'm happy to report that it seems the ModuleEnginesFX wrapper works.

One thing I noted though was that it will report "Engine State: IGNITED" even if it's out of fuel/oxidizer:

http://imageshack.com/a/img30/4468/vyia.jpg

Keep up the good work, love the complexity added with restarting an engine in LKO.

-

Kolbjørn

Hi, thanks for the suggestions and feedback about the ModuleEnginesFX wrapper and that "IGNITED" bug. I will see if I can fix it, though this bug won't give any real trouble.

You mentioned some sort of manual mode that both engine activation and shutdown require manual control. The design is a good one and similar ideas have been raised earlier (also in this thread).

While this might work well if you always control the rocket manually (it will still bring some trouble if you cannot shutdown the engine at accurate time), I'm quite worried about that if used with autopilot system like MechJeb or RemoteTech's flight computer, the engine which has a non-zero minimum thrust won't be shutdown by the autopilot system and that will cause more trouble than benefit.

One suggestion to you if you really worry about accidentally shutting down the engine, you can make use of the "ThrustLimiter" feature introduced in 0.23. Say if you put your throttle at 10%, which i think is safe enough, you then adjust the ThrustLimiter to around 20%, you will now actually have 2% of thrust. And i think that can provide enough accuracy.

P.S: Talking about engines that have non-zero minimum thrust, the "Real Fuels" mod currently owned by NathanKell has lots of engines with non-zero min thrust. That gave lots of trouble for these autopilot systems so he implemented a feature that will reduce the engine's min thrust to 0 temporarily (not by shutting down though) when the throttle is totally 0. So yes it still turns the engine into NOT_IGNITED state when the throttle is accidentally set to 0, but perhaps there would be no perfect solution unless all these autopilot systems are designed to really shutdown the engines after the execution of a burn.

[gleedadswell]

I'm thoroughly enjoying this mod, coupled with various other "realism" enhancements. Great work!

Has anyone else found that the KW Maverick-V engine doesn't ignite no matter what you do if you've got the Engine Ignitor mod? Yes, I've verified that it was connected to the right kind of fuel (I'm using RealFuels) and I had filled the tank (I'm using Stretchy Tanks). I noticed that the Maverick-V was the only engine whose ignition resource was liquid fuel + oxidizer. So I went into the .cfg and changed it so it uses Hypergolic Fluid. Now it works. Is there something wrong specifically with engines using their fuels as the resource used in ignition?

Cheers!

Geoff

[xZise]

O and U+.[...]

Then you are talking about something else. The KW Vesta VR1 is a U+ engine but does simulate ullage.

Fabian

[HoneyFox]

I'm thoroughly enjoying this mod, coupled with various other "realism" enhancements. Great work!

Has anyone else found that the KW Maverick-V engine doesn't ignite no matter what you do if you've got the Engine Ignitor mod? Yes, I've verified that it was connected to the right kind of fuel (I'm using RealFuels) and I had filled the tank (I'm using Stretchy Tanks). I noticed that the Maverick-V was the only engine whose ignition resource was liquid fuel + oxidizer. So I went into the .cfg and changed it so it uses Hypergolic Fluid. Now it works. Is there something wrong specifically with engines using their fuels as the resource used in ignition?

Cheers!

Geoff

Hi, it was the old design when i don't have Real Fuels installed.

It's used to emulate hypergolic fuel so that ignition consumes same types of resources as the normal burn does (which in stock KSP, is Liquid Fuel + Oxidizer). And Maverick-V should not be the only engine that uses that. IIRC the KW 2.5m SPS engine requires LiquidFuel+Oxidizer as well in my EI configs.

You can change that to use HypergolicFluid (which implies that your engine is a Kerolox one or something similar, instead of a hypergolic one), and remember to add a RESOURCE node in part.cfg too.

I'm hoping that the author(s) of "Stockalike"Engine/RealEngine/RftsEngine configs can adjust the EI configs in their mods so that the ignitor can request proper types of resources depending on what engine mode it is. e.g: For an engine that has 3 configs: MMH+N2O4, Kerosene+LiquidOxygen, LiquidH2+LiquidOxygen, I would hope that the first two configs require MMH+N2O4 as ignitor resources while the last one requires plenty of ElectricCharge (to emulate sparkplug).

Since there're lots of engines and there might be multiple sets of configs with each of them, it's really a heavy job to adjust all of them. Need to be patient.

Edited January 26, 2014 by HoneyFox

[NathanKell]

HoneyFox, have you thought about modeling propellant residuals? Normally the turbopump or pressurization system can't get every bit of fuel into the engine, leaving something like 0.5-2% of the fuel unobtainable when the engine cuts out for lack of fuel flow. Now, if ignitionThreshold would accept >1.0 values (or rather, if the code that uses it would), we'd be all set, but as it stands we have to do it via mod.

The annoying thing, of course, is that any dV calculator won't be aware of it, so maybe it makes more sense just to increase the dry mass and decrease the useable fuel fraction of tanks?

gleedadswell: which set of engine configs are you using with RealFuels?

[HoneyFox]

HoneyFox, have you thought about modeling propellant residuals? Normally the turbopump or pressurization system can't get every bit of fuel into the engine, leaving something like 0.5-2% of the fuel unobtainable when the engine cuts out for lack of fuel flow. Now, if ignitionThreshold would accept >1.0 values (or rather, if the code that uses it would), we'd be all set, but as it stands we have to do it via mod.

The annoying thing, of course, is that any dV calculator won't be aware of it, so maybe it makes more sense just to increase the dry mass and decrease the useable fuel fraction of tanks?

gleedadswell: which set of engine configs are you using with RealFuels?

The ignitionThreshold > 1.0 might work I think...

EDIT

{

No i guess it won't work well, because if you apply a pretty low throttle like 10%, the resource requirement will drop down also and if it falls lower than the ignitionThreshold, it can continue its burn without flame-out.

}

And you mentioned about the dV calculation in some mods (MJ, KER, VOID etc), it might be better to use dry-mass way, unless these mods already take that threshold into consideration (i don't think so ).

And one more thing about this residual feature, especially for stretchy fuel tanks: the residual might (i just guess, i don't know if it's realistic) depend on the height of the remaining fuel. For a given fuel volume, fuel tanks with a smaller diameter can have lower residual ratio while those with larger diameter have to keep more fuel so that gas won't be sucked into turbo-pumps.

Edited January 26, 2014 by HoneyFox

[xZise]

[...]The annoying thing, of course, is that any dV calculator won't be aware of it, so maybe it makes more sense just to increase the dry mass and decrease the useable fuel fraction of tanks?[...]

I'm not sure if that is really so annoying, because how exact are they anyway. At least if you do the maneuvers manually you usually don't reach the optimal efficiency so you have to consider if your rocket has a dV of X you can use only Y (< X). So when you need a bit of fuel left you simply choose a lower Y. Also isn't it hard to measure the exact amount of fuel in real-word? So if it shows you X l of fuel, you can't be sure if it is really X l which you can be used. Unfortunately you could manually transfer all of it so you can break that illusion.

Fabian

[gleedadswell]

gleedadswell: which set of engine configs are you using with RealFuels?

Not entirely sure how to answer that. Is there a particular config file I should look at? Inside the RealFuels folder there are, of course, a large number of .cfg files. I'm using KW, NP and AIES to give myself enough engines. So my engine configs would be whatever came in those packages (right?). The only thing I've changed is which ignition resource is used by the Maverick-V (edited IgnitorKW.cfg in the EngineIgnitor directory). I'm sure that hasn't answered your question. Can you clarify?

[NathanKell]

Check the second post of the RF thread. You need a set of engine configs as well as RF. Otherwise your engines won't actually be configured for RF.

[gleedadswell]

Check the second post of the RF thread. You need a set of engine configs as well as RF. Otherwise your engines won't actually be configured for RF.

Ah, that clarifies it. Yes, I'm using the Realism Overhaul. But looking back at it I think I see the problem. I had neglected to delete the .cfg from the root EngineIgnitor folder. I'll try that and see if it fixes the problem.

I totally understand why there isn't a single package we can download to get all of the mod configurations working in one go, but I also understand the grousing some people do about it on these forums. But if you're going to choose your own mods I guess reading the instructions carefully is part of the cost. ;-)

[gleedadswell]

Ah, that clarifies it. Yes, I'm using the Realism Overhaul. But looking back at it I think I see the problem. I had neglected to delete the .cfg from the root EngineIgnitor folder. I'll try that and see if it fixes the problem.

Hmm...nope, that wasn't the problem after all. I deleted all of the .cfg files from the root EngineIgnitor folder (e.g. IgnitorAIES.cfg, IgnitorKW.cfg, etc.). But that made it just like I didn't have EngineIgnitor installed at all as far as I could tell - no ignition resources displayed in the right click menu for engines, no ignition needs listed in red when I hover over the engine in the VAB, no ignition info listed in the engine information in the VAB, seem to be able to ignite and shutdown until the cows come home. So I'm not sure why the Realism Overhaul instructions say to delete those files. Am I misunderstanding the instruction? Was I supposed to delete a different set of files? Anyway, I restored those .cfg files to their originals. This once again resulted in the Wildcat-V (sorry, I said Maverick-V before but it isn't the problem) not being able to ignite (fuel flow lists as Very Stable, but even straight off the launch pad it won't ignite). So I redid my edit to IgnitorKW.cfg to switch the Wildcat-V's ignition resource over to hypergolic fluid and now it ignites.

Edited January 28, 2014 by gleedadswell

[NathanKell]

It depends on which set of engine configs you're using for RF. SFJackBauer hasn't had a chance to add Engine Ignitor settings to the RealEngines pack, and the Stockalike configs don't (they also don't limit throttle), as designed. RftSEngines *do* have EngineIgnitor configs.

[gleedadswell]

OK, and now I've found the AIES engines that use LiquidFuel+Oxidizer to ignite and they won't ignite either. So I've edited IgnitorAIES.cfg in the same way to get them to work.

[tygoo7]

With the realism overhaul I don't have any AIES engines needing an ignitor? Should I just keep the original AIES config in the folder or will that not work?

[NathanKell]

As the RF OP/readme/etc say, the engine ignitor configs that come with engine ignitor do not work with RF. That's why the RF engine config needs to have its own engine ignitor configs. At the moment only RftsEngines has Engine Ignitor support; if you're using Stockalike engines or the default SFJackBauer RealEngines, those folks involved haven't had time yet to add Engine Ignitor configs.

[gleedadswell]

It depends on which set of engine configs you're using for RF. SFJackBauer hasn't had a chance to add Engine Ignitor settings to the RealEngines pack, and the Stockalike configs don't (they also don't limit throttle), as designed. RftSEngines *do* have EngineIgnitor configs.

OK, I'm not using FftS (out of memory crashes...). But I'll have a look at them at some point. I'm happy to just go with my slight modifications to the ignitor .cfg files in KW and AIES. It still adds the element of limited ignitions, and sometimes needing to do something (like having some acceleration via other, easier igniting, engines) to stabilize fuel flow. So I'm not terribly concerned over the diversity of what ignitor resources are used by different engines.

[NathanKell]

Ok, long as it works for you, that's cool

[Chestburster]

Hey HoneyFox, I've thought about a better way to implement your Engine Ignitor with RF.

What if we could define different fuel mixtures per .cfg and all the inherent ability's coming with them (like UDMH+N2O4 are hypergolic and need no ignitors)? The module could be set up to look for the missing entry's and use values predefined in the config.

Let's say our engine is configured for KeroLox as a L type. The config provided by RF could look like this (missing lines, marked //, get filled with values from the config):

MODULE
 {
  name = ModuleEngineIgnitor
  ignitionsAvailable = 1
  // autoIgnitionTemperature = 800
  // ignitorType = Electric
  // useUllageSimulation = true
  //IGNITOR_RESOURCE
  //{
  // name = ElectricCharge
  // amount = 50
  //}
 }

A pressure fed hypergolic O engine with MMH+N2O4 could have this config:

MODULE
 {
  name = ModuleEngineIgnitor
  // ignitionsAvailable = 80
  // autoIgnitionTemperature = 700
  // ignitorType = Internal_Catalyst
  useUllageSimulation = false
  //IGNITOR_RESOURCE
  //{
  //}
 }

I hope you understand what i wanted to explain here. Don't know if this is possible or how much work it would take. It was just an idea i had in my mind for some time now.

[HoneyFox]

Hey HoneyFox, I've thought about a better way to implement your Engine Ignitor with RF.

What if we could define different fuel mixtures per .cfg and all the inherent ability's coming with them (like UDMH+N2O4 are hypergolic and need no ignitors)? The module could be set up to look for the missing entry's and use values predefined in the config.

Let's say our engine is configured for KeroLox as a L type. The config provided by RF could look like this (missing lines, marked //, get filled with values from the config):

MODULE
 {
  name = ModuleEngineIgnitor
  ignitionsAvailable = 1
  // autoIgnitionTemperature = 800
  // ignitorType = Electric
  // useUllageSimulation = true
  //IGNITOR_RESOURCE
  //{
  // name = ElectricCharge
  // amount = 50
  //}
 }

A pressure fed hypergolic O engine with MMH+N2O4 could have this config:

MODULE
 {
  name = ModuleEngineIgnitor
  // ignitionsAvailable = 80
  // autoIgnitionTemperature = 700
  // ignitorType = Internal_Catalyst
  useUllageSimulation = false
  //IGNITOR_RESOURCE
  //{
  //}
 }

I hope you understand what i wanted to explain here. Don't know if this is possible or how much work it would take. It was just an idea i had in my mind for some time now.

Umm... I'm not sure if I got it.

Let me try to restate it:

EI package has its own configs (those Ignitor_Stock.cfg, Ignitor_KW.cfg, ...), RF provide incremental config to override some values in EI's own config. e.g. change the ignitionsAvailable, change whether ullageSimulation is enabled, change whether it's pressureFed, etc.

Is that correct?