Kerbal Construction Time/StageRecovery Dev Thread

[Taki117]

I'm loving this mod! Only issue I'm having is that a mini version of my rocket (With all the effects active) appears horizontally over the R&D building, and the things with launch effects don't have launch effects.

[magico13]

I'm loving this mod! Only issue I'm having is that a mini version of my rocket (With all the effects active) appears horizontally over the R&D building, and the things with launch effects don't have launch effects.

The floating rockets over the R&D center is a known issue (discussed a few pages back) that I don't currently know how to fix. It's a side effect of having to load the ships to get to their parts, which I do only in the Space Center in order to avoid them being loaded (and causing real issues) in the flight scene. In the future I may write my own method for loading ships into parts, or I may find a better one. Until then the mini ships will appear, but should be harmless, just possibly annoying. You'll notice that every single ship in a build list or in storage will be piled on top of each other, as a result of the loading.

What do you mean by "the things with launch effects don't have launch effects"? Can you clarify that some more, perhaps with screenshots if it would help? Is this an issue on the launchpad, or is it with the floating ships? If it's the floating ghost ships, then I'm not going to worry about it, for the above mentioned reasons.

[JeffreyCor]

I still had it safe https://www.dropbox.com/s/afrjfgt8f1xr2ix/output_log.txt 4 large SRBs with a fitting on top to make the chute look good, with it set to slow to 6m/s in the realchutes editor. I had even used smartparts to make them deploy to try that even though the docs say it isn't needed. I think it just doesn't like me, or doesn't like some other mod in some way.

[CatastrophicFailure]

Ok got my log files, need to figure out how to post them when I get home tonight. I watched the bit in question all the way down, it's getting deleted at 22k which is standard for unloaded suborbital debris right?

and personally I think the rocket over the space center is kinda kewl. If you can figure how how/why it spawns where it does and just make it do so in a suitable building, now that would be even better!

[magico13]

I think it might have to do with another mod changing something. Looking at the logs, KCT doesn't even attempt to recover them (which is weird, even without parachutes it should display something as long as certain conditions are met.) I see that at least one of the boosters was set as the active vessel after separation, but am not sure if that's related. I also am only seeing two boosters actually being destroyed from the ship "Data Currier" due to being in atmosphere (and also some debris from a minmus satellite crashing into minmus, which entertained me for some reason). I'm going to list the requirements for it to attempt to recover the boosters, which may help somehow:

v is the vessel that is destroyed (the booster)

if (v != null && v.mainBody.bodyName == "Kerbin" && v.name.Contains("(Unloaded)") && v.altitude < 35000 && (v.situation == Vessel.Situations.FLYING || v.situation == Vessel.Situations.SUB_ORBITAL) && !v.isEVA && !v.isActiveVessel)

In pseudocode:

The vessel must not be null (it must exist)

It must be in Kerbin's SOI

It must be unloaded (not active/on rails, this happens when it gets outside of 2.5km)

It must be below 35km

It must be flying in atmosphere or at least on a suborbital trajectory (it has to crash)

It has to not be an EVA'd kerbal

It has to not be the active vessel

One issue could be restricting it to suborbital, as a booster with an AP of >70km and a PE of >0 is probably not considered suborbital, however your main vessel isn't even switched to sub orbital (from flying) until after the boosters are destroyed (so likely not the cause in this instance).

Could anyone experiencing issues with booster recovery please try this version and then get me their KSP.log and output_log.txt files after the boosters should be recovered? It adds some extra debug info that will possibly help me diagnose the issues (and also changes the height up to 70km instead of 35).

http://www./download/j7oc199q7th7v1t/KCT_Debug.zip

Thank you all for your help! Sorry about the issues :/

[JeffreyCor]

I had it going to active to see if the chute deployed. I was try to try all kinds of possibilities to figure out what might be going on. I'll get the test file and do a launch tonight and get you the log files. This is really a great mod adding a lot of flavor, and can see even greater advantages once the economy is implemented as well.

EDIT: Did it sooner than I thought I could, so did a launch just to get the logs for you ASAP.

KSP.log: https://www.dropbox.com/s/8pg7fz8wbc1ajyq/KSP.log

output_log.txt: https://www.dropbox.com/s/afrjfgt8f1xr2ix/output_log.txt

Edited March 28, 2014 by JeffreyCor
include log files after test

[Taki117]

SNIP

What do you mean by "the things with launch effects don't have launch effects"? Can you clarify that some more, perhaps with screenshots if it would help? Is this an issue on the launchpad, or is it with the floating ships? If it's the floating ghost ships, then I'm not going to worry about it, for the above mentioned reasons.

Specifically the sparklers from the Space Shuttle Main Engine mod after the ship has been loaded onto the launchpad and is ready to be launched. if you look below the main engine (Attached the the launch clamp) there is one of the sparklers from the Space Shuttle Main Engine mod that is supposed to be shooting sparks, but is not (Despite being in a pre-launch condition)

Edited March 28, 2014 by Taki117

[JeffreyCor]

I can see the confusion with your prior postings. That isn't part of the native game, that is part of a mod. I'm sure including the name of the mod for your tower, clamps, and effects would help with any narrowing down of the cause or how to correct.

[Taki117]

I can see the confusion with your prior postings. That isn't part of the native game, that is part of a mod. I'm sure including the name of the mod for your tower, clamps, and effects would help with any narrowing down of the cause or how to correct.

The Launch towers are from FASA, and the sparklers are from SSME mod. There are others (Including but not limited to NP, B9, and AIES) but I haven't had an issue before.

[magico13]

Did it sooner than I thought I could, so did a launch just to get the logs for you ASAP.

Ok, so this time you launched something called "Shuttle" right? What was attached to this one? I see some radial decouplers getting destroyed that for some reason are not unloaded when they get destroyed. These ones are not getting recovered because they aren't unloaded (their names don't include "(unloaded)"). Later on you stage again and there are at least two different pieces of debris. These both are being processed correctly (but don't appear to have any parachutes as their terminal velocity is 78 m/s) which means that the mod itself is still functioning correctly at that point (sometimes there's an issue and the mod just totally stops functioning afterward).

So the issue with this one seems to be that they aren't unloaded, but they were at something like 15km. Very strange. I'm going to try not using v.name and instead using v.vesselName. I'm not sure if that will change anything though. Can you get me the craft file? Maybe I can reproduce it on my end and then I can see if I can fix it, instead of continuously asking you to try things and then send me the logs

Specifically the sparklers from the Space Shuttle Main Engine mod after the ship has been loaded onto the launchpad and is ready to be launched. if you look below the main engine (Attached the the launch clamp) there is one of the sparklers from the Space Shuttle Main Engine mod that is supposed to be shooting sparks, but is not (Despite being in a pre-launch condition)

Hmm, I'm not sure what would be causing that. I'm going to do the same thing I ask of everyone and ask you for your logs. Perhaps there's (somehow) an issue with how I place Kerbals in the ship after the ship is on the pad, or with having to load the ship in the space center scene. If it's just a visual bug though, I'm not going to have it too high on my priority list (no offense, but right now I don't have the time to fix every little bug when this is still under heavy development. The game breaking bugs are my main priority atm)

Also, a reminder for everyone. I'm currently taking a small break from development as I am quite swamped with school. Please see this post for more information:

http://forum.kerbalspaceprogram.com/showthread.php?p=1031205#post1031205

Also, 0.23.5 adds timewarp to KSC and the Tracking Station, so the jump to complete will be gone and I can just timewarp to complete! I am very happy about that addition

[JeffreyCor]

Yes, the first stage was separated using an action group and smart part fuel sensor consisting of 4 SRBs with top mount chutes mounted by radial decouplers . Second stage triggered manually, 5 tank/engine cluster without chutes on a stack decoupler.

I am very glad to test anything, come up with different ships with any sort of specs needed to test different aspects. Here is also the craft file for my shuttle, though it does use some mods so that could be an issue going from one install to another. https://www.dropbox.com/s/o2dqg8175j8izw9/Shuttle.craft

[CatastrophicFailure]

Ack, apparently godaddy doesn't like these files. Can you PM me an email addy magico? (I hate dropbox)

[magico13]

Ack, apparently godaddy doesn't like these files. Can you PM me an email addy magico? (I hate dropbox)

I'm not seeing anything happening at all with that part, same as for JeffreyCor. One thing that I'm noticing about both of them is that it looks like they're being staged automatically using Smart Parts. I am starting to suspect it as the culprit, but am not yet sure. Were the fairings staged with smart parts, or was it the booster (or both)?

Wait, you didn't switch to it and follow it down that way did you? It won't be recovered if it's the active vessel. I can't tell if that's what got destroyed by deadly reentry or if that was something else.

I'm not entirely sure what's going on. It could be smart parts staging things differently and the boosters somehow not getting "(unloaded)" in their name, thereby not passing that check. I need that check to make sure they're being destroyed by being unloaded and not from crashing into each other (both use the same function). If you are using smart parts (JeffreyCor or CatastrophicFailure), could you try without it and just stage manually and see if that helps? If so, then I can focus my efforts on getting compatibility with it.

Could you also use the debug build I mentioned earlier, if it's not too much trouble, the extra bit of information it outputs will possibly help diagnose the issues. It's just the .dll file, if you extract it like you would normally and overwrite the old one it will be fine. Don't delete the old folder first. Thanks for your assistance. I wish I had more time to test things, since I don't really have a chance to try this out myself.

[JeffreyCor]

On the ones with fairings I have the smart parts for both boosters and fairings. I will launch the shuttle again to test for you with the smart parts removed and stage the SRB sep manually and upload the logs for you today so it can be as close to equal for comparing as possible. I have the debug build still installed and will keep it there so I have it ready to go for any testing you need.

Manual SRB sep done.

KSP.log: https://www.dropbox.com/s/8pg7fz8wbc1ajyq/KSP.log

output_lot.txt: https://www.dropbox.com/s/afrjfgt8f1xr2ix/output_log.txt

Edited March 28, 2014 by JeffreyCor

[magico13]

That having the same problem it seems. For some reason it's saying the vessel name is "radialDecoupler" instead of "[ship Name] Debris (Unloaded)" which is throwing it off. Would you kindly try this version and repeat the same procedure? (btw, you can just simulate the launch. It does the recovery then too, but when you revert it "forgets"). The only difference is what name I reference in the vessel (I also am printing each name I can access, there are 3 total, so hopefully one of them will be right).

http://www./download/fseste9fd933tsu/KCTDebug1.zip

This issue is really frustrating, but at least it seems like we know what the issue is, just not yet how to solve it. (the issue is just the name the debris is being given after decouple)

[JeffreyCor]

Glad to help! Put the new file in and did a simulated launch with the same ship and procedure as the last time. Here's the logs for you

KSP.log: https://www.dropbox.com/s/8pg7fz8wbc1ajyq/KSP.log

output_log.txt: https://www.dropbox.com/s/afrjfgt8f1xr2ix/output_log.txt

[magico13]

Well, darn. The different names I can apparently get are these:

[LOG 18:48:14.425] radialDecoupler

[LOG 18:48:14.426] Shuttle Debris

[LOG 18:48:14.427] Shuttle Debris

Which none of them are formatted correctly and only the first one ever gets the (unloaded) part added to their name. I think I'm going to have to come up with a different set of conditions. Instead of checking the name, I'm going to try to check if vessel.loaded is false. Another debug version for you, when you get time.

http://www./download/k887iawvok4sqvl/KCTDebug2.zip

[JeffreyCor]

Okey Dokey, another simulated launch with the same ship using the new file.

KSP.log: https://www.dropbox.com/s/8pg7fz8wbc1ajyq/KSP.log

output_log.txt: https://www.dropbox.com/s/afrjfgt8f1xr2ix/output_log.txt

[magico13]

For whatever reason it looks like they aren't being unloaded before being destroyed, even though the action of them being unloaded is what destroys them. I'll have to do some sort of workaround I guess. The one I'm thinking might work is checking (in addition to the other checks) that the altitude above the surface is greater than a buffer of 10 meters while also not being landed and is also below 35km. I fear that will "recover" boosters that collide with each other and are destroyed, which wouldn't make sense. Before trying that though, let me try seeing if the vessel is "packed". That may work better and I just thought of it while typing this.

This hopefully will be the last one... http://www./download/1wsw8l5bvc5o7g5/KCTDebug3.zip

Once again, thank you for your help. If I weren't busy with school right now I could do more of the testing myself. I did test that this one will recover boosters that normally are recovered, so this method might work. If the vessels aren't being packed before destruction though, it's going to be more difficult.

[JeffreyCor]

I am very happy to help all I can. I wish I knew more on programming to be of better assistance, but being in finals time it makes any additional learning unrelated to those subjects have to be on hold. I used the same ship again without any changes done to it to duplicate everything as before and noted SRB sep occurred just over 12k and they stayed visible until 5k away from the ship. Sometimes strange little things actually help.

KSP.log: https://www.dropbox.com/s/8pg7fz8wbc1ajyq/KSP.log

output_log.txt: https://www.dropbox.com/s/afrjfgt8f1xr2ix/output_log.txt

[magico13]

I would like to point you to this set of output from the KSP.log file:

[LOG 11:51:39.249] [KCT] Vessel being destroyed, listing information:

[LOG 11:51:39.250] radialDecoupler

[LOG 11:51:39.251] Shuttle Debris

[LOG 11:51:39.252] Shuttle Debris

[LOG 11:51:39.253] Kerbin

[LOG 11:51:39.254] 16167.6203737071

[LOG 11:51:39.255] -1

[LOG 11:51:39.256] False

[LOG 11:51:39.256] True

[LOG 11:51:39.257] True

[LOG 11:51:39.258] flying

[LOG 11:51:39.259] [KCT] Found realchute module on parachuteLarge

[LOG 11:51:39.260] chute area: 3525.652

[LOG 11:51:39.261] dragC: 1

[LOG 11:51:39.262] [KCT] Using RealChute Module! Drag/Mass ratio: 56.3295452308007

[LOG 11:51:39.264] [KCT] Recovered parts from Shuttle Debris

Which means it is working properly now! The last debug version should be fine to play with, but perhaps later tonight I will get an official update out. Assuming everything doesn't suddenly break.

Also, your chute area is huge! I only require a drag to mass ratio of like 8 or 9 to recover, which in my testing has been appropriate for a less than 10 m/s landing. When configuring the chutes were you using just the mass of the boosters? It certainly doesn't hurt to have larger chutes, but it's just something I noticed

If CatastrophicFailure could perhaps test the latest debug version and let me know if his issue has been fixed, it would be greatly appreciated! I'm assuming he is having the same issue as you are, JeffreyCor.

Edited March 29, 2014 by magico13
spelling

[JeffreyCor]

I believe it had read in the docs that it needed to be slower than 10m/s for recovery, so I wanted to make sure it had a buffer With the current version of RealChutes instead of entering the size, you enter the tons (or let it determine it itself) and speed and the thing figures out the size itself. Since that version came out I've never even looked at the size of the chutes For the booster chutes I used the listed mass on the parts list, rounded up, and set desired speed to 6m/s for a relatively gentle landing for a tossed off part.

I'll try the current test version out more during actual play later today both with and without smart parts and let you know if anything funky happens.

Update: The good news is it does indeed seem to be recovering the parts, yeay! However, it also seems to be recovering parts even that don't have chutes. I've been using some with just nosecones rather than chute cones and I'm seeing them in the parts list. Not too sure what may be triggering the recovery, if something else is slowing them down, or just what is going on with that.

Edited March 31, 2014 by JeffreyCor
Updated info

[magico13]

However, it also seems to be recovering parts even that don't have chutes.

Well, that's certainly weird. I didn't change the actual recover requirements, only the initial check for if a vessel is possibly recoverable. I'm going to ask you to do the thing you're so good at doing now: can you get me the logs? I'm curious as to if the stock drag is really high or something. It would have to be in the hundreds, and is usually 0.2 for something without parachutes.

[JeffreyCor]

If it's something weird, it must be me I'm putting up my logs from last night which I'm pretty sure should have something in them with it. I did at least 1 or 2 launches with parts that only had nosecones. I only noticed it when I was looking though my parts list and saw the engines and rechecked if I had put chutes on them. It could just be some really crazy thing happening to me, but certainly wanted to let you know since it seems related to the fix I was test out.

KSP.log: https://www.dropbox.com/s/8pg7fz8wbc1ajyq/KSP.log

output_log.txt: https://www.dropbox.com/s/afrjfgt8f1xr2ix/output_log.txt

[magico13]

If it's something weird, it must be me I'm putting up my logs from last night which I'm pretty sure should have something in them with it. I did at least 1 or 2 launches with parts that only had nosecones. I only noticed it when I was looking though my parts list and saw the engines and rechecked if I had put chutes on them. It could just be some really crazy thing happening to me, but certainly wanted to let you know since it seems related to the fix I was test out.

I looked through the KSP.log but couldn't find any instance where anything was being recovered. Did you cancel a vessel that uses parts from the inventory? Because that would give you back those parts. If you notice it happening again, let me know and send me the logs. Recovery of things that shouldn't be recovered is a serious bug if it's occurring. Thanks for testing this