Parts & Vessels Struts failing when redocking modules (or possibly when sideways docking).

[Alchemist]

Reported Version: v0.1.4 (latest) | Mods: none | Can replicate without mods? Yes 
OS: Windows 10 | CPU: Intel(R) Core(TM) i5-6400 CPU @ 2.70GHz | GPU: NVIDIA GeForce RTX 3060 | RAM: 32 GB

 

Version: 0.1.4.0

Severity: Medium (can render the vessel unusable, but can be circumvented by changing mission profile)

Consistency: Consistently reproducible, but under particulate circumstances

Here's the the relevant picture of the payload bay:

The intended mission profile at this point was as follows (the mission is STS-3 from Shuttle Challenge):

1.  redock the maneuvering modules (the ones with Kerbals in the seat) to solar panel assemblies (packed on the bottom on the bay)
2.  use the maneuvering modules to extract the solar panel assemblies and dock them to the telescope (released from the shuttle just before the screenshot was taken
3. return the maneuvering modules to the shuttle and go back to Kerbin

The problem appeared on the first step: the moment a maneuvering module was docked to the up-facing port on solar panel assembly (with control point of the resulting assembly becoming the command seat), all the struts on the shuttle were broken. No matter how many times I reloaded, the struts kept breaking.
And let's just say without the struts this shuttle falls apart under its own weight even on gentlest landing

But then I found a solution that worked:

1. Release the solar panel assemblies.
2. Catch them with the maneuvering modules and dock them to the telescope.
3. Return the maneuvering modules to their original place in the payload bay

If I do exactly this and return the maneuvering modules to the backwards ports in the bay (control scheme still switches to the seat, which is backwards relative to the shuttle) the struts remain intact. But if the maneuvering modules get docked to the upward-facing ports the solar panel assemblies were docked to - the struts immediately break. (No matter how many times the game was saved and reloaded/restarted between deployment and return of the maneuvering modules).

 

Probably needs double-checking/verifying on different vessels, but the craft file is in attachment if anyone wants to replicate it on this one. Just get it to orbit and experiment with the maneuvering modules (the RCS modules might need to be switched on through part manager)

 

Included Attachments:

HROSTS3.json

Edited September 3, 2023 by Anth12

[Alchemist]

OK, another test - and it's even weirder.

Same shuttle, different payload (this time probe-controlled). The payload is attached to 2.5m docking port on the bottom of the bay (so both root part and probe core of the payload are 90 degrees to the shuttle's cockpit)

The moment I undock the payload, the struts on the payload break (those are struts within the payload. incidentally there are no struts between the shuttle and the payload - the docking port can hold sideways 43 tons through ascent!)

 

Now, here are several options I tried after undocking the payload

test a:

• control from payload. flip it 90 (end ports) or 180 degrees (opposite middle port) and redock to the port on the bottom of the bay.
• struts on the shuttle are intact.
• but if the payload is then undocked again - struts on the shuttle break

 

test b:

• flip the payload the same as in test a
• dock to the same port, but controlling the process from the shuttle.
• struts on the shuttle break the moment the payload is docked
• outcome stays the same whether the shuttle is controlled from the cockpit or from the upwards port

 

test c:

• maintain the attitude
• control from the payload
• redock the payload (with end port) to the port in front part of the bay
• results are identical to test a

 

test d:

• maintain the attitude
• control from the shuttle
• translatejhh backwards and redock the payload (with end port) to the port in front part of the bay
• results are identical to test b - struts break on docking

 

test e (intended mission profile simulation):

• quicksave after releasing the payload and load
• control from the shuttle.
• redock the payload just as it was.
• struts on the shuttle break again

 

test f:

• quicksave after releasing the payload and load
• control from the payload.
• redock the payload just as it was.
• results identical to test A
• quicksaving and reloading after redocking doesn't help with struts breaking if the payload is undocked again

 

so far, it seems like this has something to do with how the structure is handled on docking/undocking and whether the root part orientation changes - but the logic is quite weird.