[1.2.1] Taurus HCV - 3.75 m CSM System [v1.5.3 - July 1, 2014]

[wasmic]

Link? I'm intrigued by this, but can't find it.

On the third newest page of the NovaPunch thread, right here.

[jfull]

This pod inspired me to make what might be the best looking (and most realistic) rocket I've made in a while. I used procedural fairings both in the rocket and to make a smooth service module.

Javascript is disabled. View full album

However, I ran into some odd behavior when testing the launch escape system on the pad. The launch escape works fine while the rocket is in mid-flight, but not before launch. When I active the action group to decouple the pod and fire the escape rockets, the capsule separates and flies off, but the camera focuses on a point between the rocket and the capsule, as if they were still connected. Also, several parts that were connected to the decoupler now stay with the Capsule

After a little bit, the entire rocket explodes for some reason and the capsule is suddenly accelerated at high speed toward the launch pad, as if pulled back by a bungee cord.

Only very quick deployment of parachutes can interrupt this effect.

To get to the bottom of this, I tested decoupling the pod. For some reason, this did not separate the capsule, and I could still interact with ever part of the rocket. Activating the decoupler a second time did separate it but for some reason the parts mounted on the decoupler still remained connected to the capsule somehow.

I'm beginning to suspect that Procedural Fairings might actually be messing with the decoupler and staging somehow... but just in case the problem does have something to do with this mod, I thought I'd let you know.

Edited April 7, 2014 by jfull

[Eskandare]

I'm testing EL with the MM reference to command pods rem'd out to see if there is any difference.

[Railgunner2160]

Is it possible something about the way the engines and animation are tied together and set up that could be causing the issues?? I notice that even BEFORE the screen even loads that the engines are firing. Now I know that when in the VAB that the engine boxes appear as they should, could it be that the animation is getting turned around and starts active in the first place and the engines trip as a result???

(Wrote this right as I'm going to bed, please excuse any flawed logic, also I'm no expert on how the plugin works.....)

[Eskandare]

I'm testing EL with the MM reference to command pods rem'd out to see if there is any difference.

Just quoting myself from before.

Rem'ing out the EL command moduel reference in the EL/MM config file seems to have done the trick. Doing more testing.

Edit: Okay so that did work, but for some reason the capsule is asymmetrical in mass. All my test rockets list off to one direction.

This is what I did in the MM config in EL.

//@PART[*]:HAS[@MODULE[ModuleCommand],!MODULE[ExWorkshop]]
//{
//	MODULE {
//		name = ExWorkshop
		// command pods are a tad cramped and are really meant for other
		// purposes (like flying)
//		ProductivityFactor = 0.25
		// Unfortunately, due to MM limitations, probe cores get a workshop too
//		IgnoreCrewCapacity = false
//	}
//}

It seems to have fixed the problem.

Edit 2: There is either an asymmetrical drag or mass issue. I continued rigorous testing, trying various weights to see if the mass would get corrected. it is odd, because it seems to have an off balance of sorts that was only slightly fixed by a 0.03 ton object over the window side of the pod.

Edited April 7, 2014 by Eskandare

[bsquiklehausen]

Wow! Thanks for the help - we're looking into other ways to fix the problem, but it's great to know that there's an easy way to fix it already out there!

Just quoting myself from before.

Edit 2: There is either an asymmetrical drag or mass issue. I continued rigorous testing, trying various weights to see if the mass would get corrected. it is odd, because it seems to have an off balance of sorts that was only slightly fixed by a 0.03 ton object over the window side of the pod.

I'm slightly confused by this - is it from the LES system? That's designed to be asymmetrical to clear the pod from the rest of the rocket. If it's just from the stack, it should be totally symmetrical both in drag and mass.....

[Eskandare]

I'm slightly confused by this - is it from the LES system? That's designed to be asymmetrical to clear the pod from the rest of the rocket. If it's just from the stack, it should be totally symmetrical both in drag and mass.....

It's not from the LES to my knowledge, unless the LES is somehow providing thrust while still on the stack but not activating animation, FX, or sound. I'll keep looking into this and focus on the debug log to see if it is generating 'magical' thrust.

Edit: Just to add I felt it being extremely odd because my very first flight with the pod was a complete success. Then all the odd behaviors happened. I'll see if I can roll back EL to the previous version to see if something else in EL is completely screwing up command pods.

Edit 2: Okay, so I tested it with first my orange tank, main sail, and then with my 3 orange tank 3 main sail, test rockets. I, so far so good... but the log repeated

 [Log]: Skipped frame because GfxDevice is in invalid state (device lost) 

I doubt it has anything to do with Taurus HCV. There were a few other errors, possibly associated with other mods or KSP itself. So for, rolling back EL to the previous version has corrected the problems and keeping the ExWorkShop reference rem'ed out seems to make taurus work just fine. I'm going to test on my bigger rocket to see if it still wants to pitch down.

Edit 3: The log error also appears to be an anomaly, it hasn't come up again.

Edited April 7, 2014 by Eskandare

[jnrobinson]

The download now includes an alternate part.cfg with all the engine information removed. Eventually I will either solve this problem (good) or release a version without the LES (less good) but right now I want to focus on internals development.

Sorry for the slow update; I was out of town all weekend.

[wasmic]

I've also had the problem with asymmetrical drag, but I'm not sure if it happens only when the LES has been activated, or if it happens with an unactivated LES too. Anyway, the problem is not too large when using FAR - I can just pretend that it's on purpose and that I'm doing a lifting reentry.

[jnrobinson]

Dev update: This is a chair:

If it looks like the stock chair, that's because I basically traced the basic shape and then added my own detail to make sure it would fit the kerbals properly (but my chair has about 400 less tris. Ha! Efficiency!)

In other news, internals are really time consuming, but I've finally figured out everything I need to know to get them in game and working properly.

We've upped the number of Kerbals from 5 to 7 after realizing how insanely large this capsule actually is. However, this will come with a weight increase (which was probably needed anyway).

I expect to make lots of progress this weekend on the internal; I'll give you an update on how everything is going Sunday night. No promises for a release date though.

I'm spending a lot of time making sure everything is about as clean, efficient, and high quality as I can make it. I really want this to be an interior worthy of the capsule, so no corners are being cut. It will be worth the wait

PS: Here's what the UV for the chair looks like:

I've basically spent 6 straight hours pushing these dots around and listening to country music

Edited April 11, 2014 by jnrobinson

[jordanjay29]

This is pretty cool, a bit nicer than Novapunch's new Frejya pod. Nice work.

[Raptor831]

I did some testing to try and figure out where the phantom drag is coming from. I am experiencing the drag on both versions (with and without the LES). I've looked over the configs and there's nothing I can see there that would be causing anything odd. The only indication for me is the fact that the center of lift in the VAB is not in-line with the center of mass. The stock 3-man pod has them lined up vertically. (Note, this is all with FAR installed, I didn't check without)

So, whatever is causing the CoL to be a little "below" the centerline is causing these odd drag issues. I didn't notice anything odd when I looked at the log from FAR's drag model, but I may need to look at that again. Does anyone not running FAR have the phantom drag?

[jnrobinson]

I did some testing to try and figure out where the phantom drag is coming from. I am experiencing the drag on both versions (with and without the LES). I've looked over the configs and there's nothing I can see there that would be causing anything odd. The only indication for me is the fact that the center of lift in the VAB is not in-line with the center of mass. The stock 3-man pod has them lined up vertically. (Note, this is all with FAR installed, I didn't check without)

So, whatever is causing the CoL to be a little "below" the centerline is causing these odd drag issues. I didn't notice anything odd when I looked at the log from FAR's drag model, but I may need to look at that again. Does anyone not running FAR have the phantom drag?

I don't run FAR and I don't experience the phantom drag. I will make a point to do FAR compatibility sometime soon but because I have no idea what is involved in that process and I want to get these internals done (progress is very good, btw), it may be some time.

If possible, I recommend using FAR's stock 2.5m capsule config for this capsule (if that's something that you can do) until I have more time to work out these kinks.

[problemecium]

Keep up the good work! I intend to download this as soon as it has an IVA.

[Raptor831]

I don't run FAR and I don't experience the phantom drag. I will make a point to do FAR compatibility sometime soon but because I have no idea what is involved in that process and I want to get these internals done (progress is very good, btw), it may be some time.

If possible, I recommend using FAR's stock 2.5m capsule config for this capsule (if that's something that you can do) until I have more time to work out these kinks.

After a bit more research, it's FAR definitely. It's calculating the shape of your capsule as something other than a normal cone, and I don't know why. FAR calculates this based on node size (I think), so I don't know where the oddity is coming from. Ferram may have updated something that changes this, though. For reference, here's the ksp.log output for the stock 3-man capsule:

FAR drag model added; Size: (2.4, 1.8, 2.4), LD (2.4, 2.4, 0.0), UD (1.3, 1.3, 0.0)
Surface area: 10.89397
Fineness Ratio: 0.7550685
TaperRatio: 0.5355
Cross Sectional Area: 4.698075
Cross Sectional Tapered Area: 3.350854
Major-minor axis ratio: 0.9999999
Centroid: (0.0, 0.2, 0.0)

And here's the Taurus:

Taurus HCV: FAR drag model added; Size: (3.8, 2.7, 4.1), LD (3.8, 4.1, 0.0), UD (1.7, 1.7, 0.0)
Surface area: 24.32616
Fineness Ratio: 0.7017185
TaperRatio: 0.4426783
Cross Sectional Area: 11.99424
Cross Sectional Tapered Area: 9.660213
Major-minor axis ratio: 0.9208301
Centroid: (0.0, -0.4, 0.3)

In short, the Centroid is where the lift/drag effects are calculated from. The Taurus has the extra offset, which I believe is the issue. If you ask Ferram, he should be able to point you in the right direction. I am, unfortunately, not quite smart enough to figure it all out! If I get a config that works properly, I'll post it up if you haven't already fixed it.

The pod is great, btw, and I'm looking forward to the IVA.

[jnrobinson]

After a bit more research, it's FAR definitely. It's calculating the shape of your capsule as something other than a normal cone, and I don't know why. FAR calculates this based on node size (I think), so I don't know where the oddity is coming from. Ferram may have updated something that changes this, though. For reference, here's the ksp.log output for the stock 3-man capsule:

FAR drag model added; Size: (2.4, 1.8, 2.4), LD (2.4, 2.4, 0.0), UD (1.3, 1.3, 0.0)
Surface area: 10.89397
Fineness Ratio: 0.7550685
TaperRatio: 0.5355
Cross Sectional Area: 4.698075
Cross Sectional Tapered Area: 3.350854
Major-minor axis ratio: 0.9999999
Centroid: (0.0, 0.2, 0.0)

And here's the Taurus:

Taurus HCV: FAR drag model added; Size: (3.8, 2.7, 4.1), LD (3.8, 4.1, 0.0), UD (1.7, 1.7, 0.0)
Surface area: 24.32616
Fineness Ratio: 0.7017185
TaperRatio: 0.4426783
Cross Sectional Area: 11.99424
Cross Sectional Tapered Area: 9.660213
Major-minor axis ratio: 0.9208301
Centroid: (0.0, -0.4, 0.3)

In short, the Centroid is where the lift/drag effects are calculated from. The Taurus has the extra offset, which I believe is the issue. If you ask Ferram, he should be able to point you in the right direction. I am, unfortunately, not quite smart enough to figure it all out! If I get a config that works properly, I'll post it up if you haven't already fixed it.

The pod is great, btw, and I'm looking forward to the IVA.

It may have something to do with the fact that the collision mesh is not just a cone; it also has a small cylindrical section that extends behind the cone. Could you explain the asymmetrical drag more? When does it happen? What way does it torque? Is it just during reentry?

[Raptor831]

It may have something to do with the fact that the collision mesh is not just a cone; it also has a small cylindrical section that extends behind the cone. Could you explain the asymmetrical drag more? When does it happen? What way does it torque? Is it just during reentry?

The drag is a constant pull "down". As in, if the cockpit windows are up, the capsule wants to pull down constantly. The SAS could handle it really, and if I were to build an orbit-capable rocket the forces might get missed in the shuffle anyway. But on a test rocket (capsule, decoupler, KW 3.75m pancake tank, and the KW 3.75m orbital engine) it's very noticeable. If you use any yaw, the pod wants to roll around the CoL (which is normal if the CoL is in the right spot), which makes the start of the gravity turn interesting if you're not facing the right direction.

I have not tested a proper reentry yet, but in my suborbital tests with DeadlyReentry heatshields, the pod doesn't flip out really. The weight of the shield holds the position ok. If you dump the shield, you get odd movements, but nothing that causes everything to explode. I'll test with a proper reentry to make sure though.

If Ferram's using the collision mesh to figure the drag, as long as yours is rotationally symmetrical, I would think it would not pull to one side. If the bottom collision cylinder is offset in any way, then I'd start with that. Also, I can definitely say the LES engines and your plugin .DLL are not to blame, because this occurs with or without those.

If you have any other questions, let me know. Or if you have a test setup, I can give it a whirl.

[jnrobinson]

The drag is a constant pull "down". As in, if the cockpit windows are up, the capsule wants to pull down constantly. The SAS could handle it really, and if I were to build an orbit-capable rocket the forces might get missed in the shuffle anyway. But on a test rocket (capsule, decoupler, KW 3.75m pancake tank, and the KW 3.75m orbital engine) it's very noticeable. If you use any yaw, the pod wants to roll around the CoL (which is normal if the CoL is in the right spot), which makes the start of the gravity turn interesting if you're not facing the right direction.

I have not tested a proper reentry yet, but in my suborbital tests with DeadlyReentry heatshields, the pod doesn't flip out really. The weight of the shield holds the position ok. If you dump the shield, you get odd movements, but nothing that causes everything to explode. I'll test with a proper reentry to make sure though.

If Ferram's using the collision mesh to figure the drag, as long as yours is rotationally symmetrical, I would think it would not pull to one side. If the bottom collision cylinder is offset in any way, then I'd start with that. Also, I can definitely say the LES engines and your plugin .DLL are not to blame, because this occurs with or without those.

If you have any other questions, let me know. Or if you have a test setup, I can give it a whirl.

Ok, I'll spend some time investigating this before the next release. Thanks a bunch for all your help!

[Raptor831]

Ok, I'll spend some time investigating this before the next release. Thanks a bunch for all your help!

You're quite welcome!

For completeness, I did a test reentry from orbit with DRE/FAR. Works quite well, even with no SAS. The 3.75m default DRE heatshield masses a good amount, so my total mass for the capsule, shield, chutes, and docking port were almost 12 tons. But, she flew pretty much straight as an arrow the whole way. I imagine without the shield, it would have been a bit more exciting, since FAR wasn't calculating the pod to have much drag at that orientation anyway.

Also, you might consider making the bottom node for the capsule a size 3. FAR has calculated size/drag from the nodes, so given that you are 3.75 meters at the base, it would help to have the right node size. I did try this during testing and it didn't fix the odd drag, so no magic bullet there. And since stock now has size 3 parts, it would probably be beneficial to follow that.

I hope all this helps you out. I'll be using this for myself either way, as it'll be great for an Orion/Dragon-like pod and a station ferry!

[Mokmo]

I wonder if there's a way to take the LES out of the staging list but keep it available as an action group trigger. I just removed the solidfuel from it instead of using the alternative cfg file.

quick review: i like the size, i installed it using Procedural fairings like suggested in the first post, looked pretty neat, i hope the IVA is coming soon,

problems: my kerman does two rolls before grabbing the ladder when going on EVA (or just gets ejected forcefully), zooming in too close will make the LES thrusters disappear and i see inside (see screenshot) not sure what causes that (i'd expect it to disappear on a much closer zoom)

Overall a pretty good capsule.

[Senior Slaphead]

Live stream featuring this part here.

It's rather fabulous... looking forward to the interior.

[TheWampa]

I made my own service module for the bobcat orion pod:

http://i.imgur.com/UtUCV1L.jpg

Gold foil stretchytank and the round lionhead aerospace solar panels, 2m fatman nerva, but if you for some reason don't like the idea of a manned mission with a nuclear reactor on it you could use the KW 2m CSM engine, or any engine you like.

Hm, does anybody know where I can get that comm dish?

[Tangle]

comm dish?

AIES aerospace.

[bsquiklehausen]

I wonder if there's a way to take the LES out of the staging list but keep it available as an action group trigger.

problems: my kerman does two rolls before grabbing the ladder when going on EVA (or just gets ejected forcefully), zooming in too close will make the LES thrusters disappear and i see inside (see screenshot) not sure what causes that (i'd expect it to disappear on a much closer zoom)

http://i.imgur.com/K0cH0HG.png?1

1. We're looking into that - it's obviously the best solution to the way it works, but it may be tough with the way KSP handles engines.

2. The next update should have a new ladder hitbox which should fix the problem.

3. That's a known bug and will hopefully also be fixed with the next update.

Also - IVA soon - it's taking a bit longer than we expected, though the hardest parts are done.

[Rofl47]

I would like to report a glitch, for some reason, The launch escape system starts up when on the launchpad.