stability issue with chain model

[johhell]

Tried to simulate a multiPendulum problem (to compare the results with an MTK implementation from another project). The model + results: D7xproblem.zip

model description

model of 1 segment:

Segment(;obj1, phi_start1=0.0) =  Model(
    temp1   = Object3D(),
    box     = Object3D(feature=Solid(massProperties=MassProperties(; mass=mass_per_seg, Ixx=0.01, Iyy=0.01, Izz=0.02))),
    frame1  = Object3D(parent=:box, translation=:[ len_per_segment, 0.0, 0.0 ]), # nach unten  z-Richtung
    rev1     = RevoluteWithFlange(obj1=obj1, obj2=:temp1, axis=2, phi=Var(init=phi_start1)),
    prism   = Prismatic(      obj1=:frame1, obj2=:temp1, axis=1 , s=Var(init=0.0), v=Var(init=0.0)),
    spring  = SpringDamperPtP(obj1=:frame1, obj2=:temp1, springForceLaw=:springForce, damperForceLaw=:damping),

    damper1 = Damper | Map(d=19.50),
    fixed  = Fixed,
    connect = :[
                (damper1.flange_b, rev1.flange),
                (damper1.flange_a, fixed.flange)     ]      )

results

In my simulation I used stopTime=3.73.

Longer duration e.g. stopTime>3.73 results in a stability problem of the solver. Maybe the problem has something to do with angle transformation. Below values of m4.rev.phi short before the simulation becomes unstable.

Full model, PNG and GIF can be found in the attached ZIP

julia> T7test.revPhi[4]
374-element Vector{Float64}:
  0.0
 -2.851810285609827e-7
 -1.8911534161141074e-6
  ⋮
 -0.056001209423869214
 -0.057929861281100024
 -0.04127514251024846
 -6.5751491644538245

[johhell]

Partly solved. My interpretation of the Prismatic joint was not correct. The answer to my problem I found in the test-examples. Modia3D is great!