bug with ram air intakes?

[killernerd]

I'm trying out the new parts from the latest patch (which i'm loving btw) and took the time to upgrade some of my.... bulkier (read ugly mofuggas) designs with the new sexeh looking parts.

What i'm pretty much doing is rebuilding the old design with these new parts, since i only really fly at about 20 - 21 km (more or less) I shouldn't need more than 1 ram air intake per engine.

And on the old craft it works fine that way but on this design the engines will start flaming out somewhere between 5 and 6 km up (where the normal jet engines would start losing thrust).

I don't know what's causing it, i have them attached to the new splitter thingy (reversed at the front) and it does show they're working (they are receiving airflow) but it's like they're not passing it through to the engines in the back.

Any ideas?

ps: i'll ad a screenshot in a bit.

edit: screenshot:

edit edit: apparently this forum doesn't like steam screenshots?

here's the direct link then :http://steamcommunity.com/sharedfiles/filedetails/?id=325449412

Edited October 11, 2014 by killernerd

[Engineer of Stuff]

Try re-updating the game. That might work.

[killernerd]

other than a complet reïnstall i don't see how i would go back an update so i could reupdate the game?

There's certainly something strange going on though, i just tried to takeoff and my left intakes were giving less air than my right intakes. Somehow.

It went away after takeoff but came back at 3000m causing one of my left engines to suddenly flame out and putting me in a deathspiral to the ground. (i did close the normal air intakes so i could test the ram air ones).

[Geschosskopf]

I don't know what's causing it, i have them attached to the new splitter thingy (reversed at the front) and it does show they're working (they are receiving airflow) but it's like they're not passing it through to the engines in the back.

There's your problem. Resource flow (fuel, air, whatever) only works properly in 1 direction through a bicoupler, from the single side to the double side. IOW, bicouplers will take 1 big flow and turn into to small flows, but will not combine 2 small flows into 1 large flow. Whiich is what you're doing because the bicoupler has a single node on its rear side. So, in your plane, only 1 intake on each bicoupler is actually working.

Kasuha went through many tests to prove this if you care to dig for them. But you can see it for yourself with a simple test. Make a ship with 2 LFO tanks attached to the front of the bicoupler and 1 rocket engine on the back. Alt-right-click on both tanks so you can see how much fuel is in them both at once, and then start the engine. Watch how the fuel in the 2 tanks isn't consumed equally.

[killernerd]

i see... Then how do you go about combining these flows? And how does that explain the fact it worked fine on my other craft (with a very similar design)?

Also I did manage to get it to work by removing the left side intakes and reattaching them.

It's an interesting point you're presenting though.

[killernerd]

so i tried your theory but it doesn't hold up.

If what you say is correct i should be able to get as high with only 1 ram air intake for each side. So one air intake is feeding 2 engines.

But with that setup i flamed out at 16-ish km.

[LethalDose]

There's your problem. Resource flow (fuel, air, whatever) only works properly in 1 direction through a bicoupler, from the single side to the double side. IOW, bicouplers will take 1 big flow and turn into to small flows, but will not combine 2 small flows into 1 large flow. Whiich is what you're doing because the bicoupler has a single node on its rear side. So, in your plane, only 1 intake on each bicoupler is actually working.

Kasuha went through many tests to prove this if you care to dig for them. But you can see it for yourself with a simple test. Make a ship with 2 LFO tanks attached to the front of the bicoupler and 1 rocket engine on the back. Alt-right-click on both tanks so you can see how much fuel is in them both at once, and then start the engine. Watch how the fuel in the 2 tanks isn't consumed equally.

I don't think resource flow is the issue. I think the resource flow problem with the splitters and most engines is caused by the "STACK_PRIORITY_SEARCH" method used to handle LiquidFuel (and oxidizer, in LFO engines). However, IntakeAir uses the "ALL_VESSEL" resource flow, which should just look at the the total amount of resource in the vessel. The source shouldn't matter.

[Geschosskopf]

I don't think resource flow is the issue. I think the resource flow problem with the splitters and most engines is caused by the "STACK_PRIORITY_SEARCH" method used to handle LiquidFuel (and oxidizer, in LFO engines). However, IntakeAir uses the "ALL_VESSEL" resource flow, which should just look at the the total amount of resource in the vessel. The source shouldn't matter.

Ah yes, my bad. Sorry. I blame my persistent HUTA (Head Up The ... ) syndrome.

How about this? It's another thing Kasuha painstakingly researched.... The way multiple engines draw from multiple intakes is solely a function of the order the parts are attached to the plane, which determines the order in which the parts are listed in the .craft file. IOW, you have to attach the intakes for 1 engine, then that engine, then the intakes for the next engine, then that engine, etc. If you violate this rule, such as by attaching all the intakes, then all the engines, the 1st engine on the list will take all the air it can use from all the intakes listed above it in the .craft file and the engines after the 1st have to fight over what's left, so they'll flame out sooner. This usually means you'll have to NOT use symmetry when attaching intakes and engines.

[Kryxal]

Doesn't it work fairly well with symmetry on whole assemblies? All the intakes then all the engines can run into problems, but intake-engine-copy I think works better.

[Laie]

Doesn't it work fairly well with symmetry on whole assemblies? All the intakes then all the engines can run into problems, but intake-engine-copy I think works better.

Intake-engine-copy works fine.

I've run into problems with multicouplers before. The combination 2x2 intakes, quadcoupler, (stuff), bicoupler, two engines reliably has one engine flame out early. I always wanted to investigate that further someday, but never quite came around to it.

[killernerd]

that'd make more sense because that's exactly what i did. Attach all engines (using symmetry) then all the air intakes (again, using symmetry).

I'll try reattaching all of the engines and intakes using the copy method. I'll post the results.

It makes me wonder why they chose to use that search algorithm though. Surely they must've realized that the order in which the parts are saved can actually influence the craft's performance... (or perhaps they didn't?)

[killernerd]

so I tested it and there seems to be truth behind the "order is important" theory.

did the intake-engine copy method, slapped on 2 extra intakes because why not and a couple of minutes later we were cruising along happily at 24km above the ground doing a comfortable 1.7km/s