Spring Supports: reactions = 2x input forces

[flit-float-fly]

Describe the bug
I am trying to simulate a grade beam, so I have made a beam model that is supported by springs. I apply a simple UDL or Point load at the centre of the beam and the total sum of my reactions is 2x what my UDLxL or point load are. I have attached a screen shot of the statics check.

I double checked all of my inputs, and the best guess I have (apart from a bug) is that I initialized the spring supports incorrectly... but I can't see where.

To Reproduce
Import the attached module (as".py")
fdn_model_20240311_issue.txt

and call up a beam with this script:
input_dict = {"mat": "conc",
"L": 10000,
"w": 1000,
"E": 23000,
"A": 1e6,
"Iz": 83333333333,
"Iy": 83333333333,
"nu": 0.3,
"rho": 24e-6,
"J": 10e6,
"subgrade_modulus": 0.1,
"n_springs": 10,
"UDL": [1,1,0,10000],
"pt_loads": [(0,5000)]
}
gb = fdn.grade_beam(**input_dict)

gb.analyze(check_statics=True) <- this should produce the screenshot attached.

Expected behavior
I was expecting to have the sum of the reactions match my input forces.

Screenshots

[JWock82]

I've isolated the issue and have corrected it in the GitHub repo. I'll push this change out with version 0.0.93.

The program did not like having both a support and a spring at the same node. This helped me lock on to a bigger bug that was reporting phantom reactions at unsupported degrees of freedom that should not be showing any reactions. Thank you for reporting. That bug was a tricky one to find and only triggered under very rare circumstances.

[flit-float-fly]

Well this is great! This is my first time reporting anything, and I'm glad I could help; I'm also very glad you're so responsive. This is an impressive tool, and am looking forward to following its development.