parts [1.10.x] SDHI Service Module System (V4.0.4 / 11 October 2020)

[Synthesis]

Gave the unfinished release a try--so far, working very fine, except it seems mounting parts on the service module under the ring is a little sensitive (move to far in either direction, and your parts will stick "outside" where the fairing would be, even if there is visible space.

Of course, it is unfinished, so I don't mind in the least--just wanted to report that, very excited to see this completed. I'm pretty sure this is the last 0.90 "must-have" mod I'm waiting to get updated to 1.02.

[sumghai]

I don't know why but i have extrem problems with DRE and the last masterbranch from github.

Installed mods are:

AnimatedDecouplers

DeadlyReentry

FASA Launch Clamps and Towers

Ferram-Aerospace-Research

ModuleManager-2.6.3

SDHI_ServiceModuleSystem

The Heatshield loses Ablator from launch to 20km high (ca 800/1000 left), and on reentry it dosn't loses any ablator, don't even get hot (6.8K). Instead the Dockingport gets hotter and hotter during reentry until it explodes. Any Idea why this happens?

For testing i have used the heatshield come with DRE. they dosn't loses any ablator during liftoff and losing only on reentry, but the dockingport still gets hotter untill it explodes.

I'm not sure what's going on there, either. Stawaster helped me tweak the SDHI heat shield.

Gave the unfinished release a try--so far, working very fine, except it seems mounting parts on the service module under the ring is a little sensitive (move to far in either direction, and your parts will stick "outside" where the fairing would be, even if there is visible space.

That's a relatively old issue that cropped up in 0.90, which I haven't been able to resolve.

[Starwaster]

I don't know why but i have extrem problems with DRE and the last masterbranch from github.

Installed mods are:

AnimatedDecouplers

DeadlyReentry

FASA Launch Clamps and Towers

Ferram-Aerospace-Research

ModuleManager-2.6.3

SDHI_ServiceModuleSystem

The Heatshield loses Ablator from launch to 20km high (ca 800/1000 left), and on reentry it dosn't loses any ablator, don't even get hot (6.8K). Instead the Dockingport gets hotter and hotter during reentry until it explodes. Any Idea why this happens?

For testing i have used the heatshield come with DRE. they dosn't loses any ablator during liftoff and losing only on reentry, but the dockingport still gets hotter untill it explodes.

It's a FAR issue. Get the dev version or wait for its next update.

[sumghai]

It's a FAR issue. Get the dev version or wait for its next update.

Ah, I see. Thanks for the heads-up

[Synthesis]

That's a relatively old issue that cropped up in 0.90, which I haven't been able to resolve.

No kidding? I hadn't noticed...then again, in 0.90, I used my craft files from 0.25, so I guess that would explain. Some sort of collision problem I guess--hardly devastating, you can skirt around it by mounting smaller parts in that area.

[sumghai]

Some sort of collision problem I guess--hardly devastating, you can skirt around it by mounting smaller parts in that area.

Yeah, it's probably changes to how colliders and bounding boxes for the Service Module and/or surface-attached parts interact.

I'll check my models tonight to see if I can shed any light on this.

[DianonForce]

It's a FAR issue. Get the dev version or wait for its next update.

Thanks for this Info. Seems i searched on the wrong end . I will try the new FAR dll today.

[Bomoo]

Here's the bodged solution I suggested earlier in the thread. Appears to be functional in terms of drag occlusion, but not many style points.

[DianonForce]

OK, i installed last FAR branch today, the dockingport dosn't explode anymore at reentry. But still get strangs behavior during liftoff. On half of all liftoff the heatshild loses all or nearly all of his ablative shielding on the first few meters. And still dosn't loses any ablative shilding during reentry.

output_bin.txt for analysis prupose

Installed mods are:

AnimatedDecouplers

DeadlyReentry

FASA Launch Clamps and Towers

Ferram-Aerospace-Research

ModuleManager-2.6.3

SDHI_ServiceModuleSystem

What i have done:

I have build a simple rocket SDHI Service Module on top of 2 LF stages and 6 SRB's. Launched all first time -> losing 60% of ablative shielding during takeoff. Revert flight to launchpad and tryed again -> dosn't losing any ablative shielding during liftof but also loses no AS during reentry. Revert flight an tryed again -> losing 80% of the AS during liftoff. Revert flight ... -> losing no AS during liftoff. Next revert ... -> losing all AS on the first 100m during liftoff.

Hope annyone understand whats going wrong.

[sumghai]

Are you, by any chance, using the pod cover and side fairings in your SDHI SMS stack?

If so, please note that they currently don't provide drag occlusion (and by extension, ascent shielding) because they are incompatible with the stock aero system. Starwaster will help me fix it once he's taken a breather from updating/maintaining DRE.

[Synthesis]

So, I've been trying to get a working abort stage together with the assembled service module--and I'm running into some obstacles.

Currently, even if you have Realchute installed, you can't edit the docking ring parachute in the action menu--which is a problem, since you can't set it to automatically deploy earlier.

That, in turn is a problem because if you just try and deploy it normally, the pod can be lifted by the escape tower effectively, but when you try and deploy the chute, you get the message "Can't deploy, parachute is in fairing" even though the fairing is already clear.

[sumghai]

So, I've been trying to get a working abort stage together with the assembled service module--and I'm running into some obstacles.

Currently, even if you have Realchute installed, you can't edit the docking ring parachute in the action menu--which is a problem, since you can't set it to automatically deploy earlier.

As per the FAQ, this is intentional - the parachutes were optimized specifically for return to Kerbin from LKO/Mun/Minimum. The last thing I want is for newbies to play around with the values, apply incorrect settings, then come crying to me that I've cold-bloodedly murdered their Kerbals.

That, in turn is a problem because if you just try and deploy it normally, the pod can be lifted by the escape tower effectively, but when you try and deploy the chute, you get the message "Can't deploy, parachute is in fairing" even though the fairing is already clear.

I've never had any issues with manual chute deployment, although that probably was back in 0.90 when I still had access to a computer capable of running KSP (I'm currently relegated to working from a laptop that can't run the game). After firing the LES, I simply waited until the pod had reached the apex of its trajectory and slowed down, before manually jettisoning the pod cover/LES combo and opening up the chutes.

As for the fairing issue, as I've explained numerous times previously (including in the OP), the pod cover and Service Module side fairing panels are incompatible with the new stock aero system, so odd behaviour is to be expected until Starwaster helps me figure out a fix.

[sumghai]

@Synthesis - with regards to the surfaced-attached parts issue, I took another look at the Service Module colliders, and while I couldn't see anything obvious, I ended up redoing them. Please try out the latest commit on GitHub and let me know if that helps.

FYI, here's what the Service Modules look like right now - the lower propulsion section is a convex mesh collider in the form of a single cylinder, while the upper ring is made up of 24 convex mesh colliders in the form of flat cuboids.

In any case, the colliders for the ring shouldn't be sticking down far enough to be causing any issues, so I suspect that the problem might be to do with the bounding boxes of certain RCS thrusters or solar panels from other add-ons being too big.

[Starwaster]

Re: AnimatedDecouplers. These were still functional without being updated, but they are now officially updated for KSP 1.0.2 (which really just means that I recompiled the plugin against the new KSP binaries)

[DianonForce]

Are you, by any chance, using the pod cover and side fairings in your SDHI SMS stack?

If so, please note that they currently don't provide drag occlusion (and by extension, ascent shielding) because they are incompatible with the stock aero system. Starwaster will help me fix it once he's taken a breather from updating/maintaining DRE.

Yes, i use the pod cover and fairings but i'm also using FAR, wich if i understand right, calulate drag on shape and not on values in a cfg. And also it work with same configuration with the headshields come with DRE. So i dont think it is a bug only with stock aero.

Imho, it looks like the shields are upside down, so they work on ascending and not on descending. I will try this today.

Edit: finaly it seems to work, i don't know why. I have throwing my test installation in the bin and updated a bunch of mods for my normal installation, after all, now it seems to work during reentry, still loses som AS during ascending, maybe because of an high AoA.

Edited May 21, 2015 by DianonForce

[Starwaster]

Yes, i use the pod cover and fairings but i'm also using FAR, wich if i understand right, calulate drag on shape and not on values in a cfg. And also it work with same configuration with the headshields come with DRE. So i dont think it is a bug only with stock aero.

Imho, it looks like the shields are upside down, so they work on ascending and not on descending. I will try this today.

Edit: finaly it seems to work, i don't know why. I have throwing my test installation in the bin and updated a bunch of mods for my normal installation, after all, now it seems to work during reentry, still loses som AS during ascending, maybe because of an high AoA.

Just an FYI, shields don't work the same as they did in previous versions of DRE.

If your shield ablates then it's because it's heating up past 400K and keeps ablating until the temperature drops below that. (or whatever the shield's threshold is)

[DianonForce]

it is right that the Shield has only 3 AS left after landing from 120km orbit to a 45km periapses descending?

[Synthesis]

As per the FAQ, this is intentional - the parachutes were optimized specifically for return to Kerbin from LKO/Mun/Minimum. The last thing I want is for newbies to play around with the values, apply incorrect settings, then come crying to me that I've cold-bloodedly murdered their Kerbals.

I've never had any issues with manual chute deployment, although that probably was back in 0.90 when I still had access to a computer capable of running KSP (I'm currently relegated to working from a laptop that can't run the game). After firing the LES, I simply waited until the pod had reached the apex of its trajectory and slowed down, before manually jettisoning the pod cover/LES combo and opening up the chutes.

As for the fairing issue, as I've explained numerous times previously (including in the OP), the pod cover and Service Module side fairing panels are incompatible with the new stock aero system, so odd behaviour is to be expected until Starwaster helps me figure out a fix.

I understand your reasoning behind the fixed parachute values--though I think as of 1.0 it's glitching in the other regard. I've set up a "escape from surface" test a few times where I test the escape tower from the launch pad. The parachute just doesn't want to deploy, either manually or by action group, even after you meet the criteria (negative velocity, etc.) because of the fairings.

In which case, then yes, probably a 1.0-related glitch. I don't want to sound ungrateful or anything, but bringing it to your attention if you were not aware of it. I'll try out your most recent release this evening and report my findings in the area of parts attachments.

Edited May 21, 2015 by Synthesis

[Synthesis]

@Synthesis - with regards to the surfaced-attached parts issue, I took another look at the Service Module colliders, and while I couldn't see anything obvious, I ended up redoing them. Please try out the latest commit on GitHub and let me know if that helps.

FYI, here's what the Service Modules look like right now - the lower propulsion section is a convex mesh collider in the form of a single cylinder, while the upper ring is made up of 24 convex mesh colliders in the form of flat cuboids.

http://i.imgur.com/JNzwb88.png

In any case, the colliders for the ring shouldn't be sticking down far enough to be causing any issues, so I suspect that the problem might be to do with the bounding boxes of certain RCS thrusters or solar panels from other add-ons being too big.

So, I tried the new MU file, and while the collision model does seem different, it's still having some glitches towards the middle of the cylinder (so there's a "sweet spot" for top and bottom third of the cylinder respectively). Just thought you might want to know, I'm sure these sort of things can take time, and some areas are beyond your control for now.

[sumghai]

So, I tried the new MU file, and while the collision model does seem different, it's still having some glitches towards the middle of the cylinder (so there's a "sweet spot" for top and bottom third of the cylinder respectively). Just thought you might want to know, I'm sure these sort of things can take time, and some areas are beyond your control for now.

I'm going to need to see screenshots describing those situations.

[Synthesis]

I'm going to need to see screenshots describing those situations.

Sure, sorry I didn't get to this sooner--hopefully these illustrate the issues. Objects that have a larger mounting base, like large batteries, just "hover" entirely.

[sumghai]

Hmm, that's darn peculiar.

Unfortunately, it seems that the only solution is to pick surface-attached parts with smaller bounding boxes.

[Bomoo]

As far as I've seen, that floating behaviour is only present in the clamp version of the SDHI service module. The clampless version allows junk to surface-attach normally. I also thought I remembered running into this issue and solving it by taking the bottom adapter/fairing mount off before attaching junk to the trunk (see what i did thur?), but this ended up not working. Nor did much of anything else including removing the LV-909, any RCS blocks, etc.

[sumghai]

As far as I've seen, that floating behaviour is only present in the clamp version of the SDHI service module. The clampless version allows junk to surface-attach normally. I also thought I remembered running into this issue and solving it by taking the bottom adapter/fairing mount off before attaching junk to the trunk (see what i did thur?), but this ended up not working. Nor did much of anything else including removing the LV-909, any RCS blocks, etc.

I may have an idea - although I no longer have the source Blender files for the V1.9 version (i.e. the one without the umbilical), I can try re-downloading V1.9 from the GitHub archives and re-importing its Service Module .MU model to see how I did the colliders back then.

No promises that this will be the magic bullet, though.

[StEligius]

Sure, sorry I didn't get to this sooner--hopefully these illustrate the issues. Objects that have a larger mounting base, like large batteries, just "hover" entirely.

I've had that problem since .90, figured it was just snapping to the 2.5m section. I just use the offset tool to move radial attach parts into place.

Edited May 23, 2015 by StEligius