Search
Close this search box.
Notifications
Clear all

Testing packing ring for model calibration

2 Posts
2 Users
0 Reactions
170 Views
Posts: 1
 RMH
Topic starter
(@rmh)
New Member
Joined: 5 months ago

Hello,

 

I'm trying to calibrate material model for a packing ring made of braided graphite. The main goal is predict radial forces, when the packing ring is under axial compression. The packing ring is used as a dynamic shaft seal, so it is inserted in a stuffing box, and the shaft goes through it, so the radial expansion of the ring is limited by the shaft and walls of the stuffing box.

What kind of tests should be done for calibrating the material model? I'm planning to use temperature-dependent three-network model, is that ok or is there better model for this application?

I have possibility to test the packing ring in a stuffing box, so it is kind of confined compression test. I tried it, but MCalibration seems not to be able to gain good results based on the strain-stress data. Another chance is to test the ring by compressing it between two plates. Could this method lead in better results?

 

I attached a .mcal file, where I tried to calibrate a three-network model with one dataset from a confined compression test, and two datasets from tests where the sample was compressed between two plates. The model is very accurate for the tests done between the plates, but not for the confined compression tests. Is there something wrong in my calibration?

Topic Tags
1 Reply
Posts: 3998
(@jorgen)
Member
Joined: 5 years ago

Hello RMH,

 

I don't know much about braided graphite, but I think your mcal-calibrated material model looks pretty good.

If  you want to improve the confined compression unloading predictions then you will need to switch to either a linear viscoelastic model with volumetric compression, or a crushable foam material model. Of course, these models have their own challenges...

/Jorgen

Reply
Share: