 | | Re: Mathematical assistance please —Philippe Hurbain
|
| | | Hello Kevin, (...) What about using 4-4cyls primitives placed to the end of hose segments? Scale their length to (hose diameter)*tan(angle/2). See example for a 30° angle: (URL) and (URL) know, this is not a full solution, you need to rotate the (...) (20 years ago, 6-Jan-05, to lugnet.cad.dev)
|
| | | |
| | | | | Re: Mathematical assistance please —Kevin L. Clague
|
| | | | (...) Hi Philo, Good to hear from you. My mistake.... I am using 4-4cyls.dat. rotating them to align them with the S0-S1-S2 plane is what I don't know how to do. Thanks, Kevin (20 years ago, 6-Jan-05, to lugnet.cad.dev)
|
| | | | |
| | | | | | Re: Mathematical assistance please —Ross Crawford
|
| | | | (...) It's hard work, but this page might help (URL) (20 years ago, 6-Jan-05, to lugnet.cad.dev)
|
| | | | |
| | | | | | Re: Mathematical assistance please —Kevin L. Clague
|
| | | | In lugnet.cad.dev, Ross Crawford wrote: <snip> (...) know how to come up with the rotation angle needed to twist the 4-4cyls.dat into the right position. One reason to push forward on my original solution is because the same knowledge can be used to (...) (20 years ago, 6-Jan-05, to lugnet.cad.dev)
|
| | | | |
| | | | | | Re: Mathematical assistance please —Ross Crawford
|
| | | | (...) Could you transform the co-ordinate system so that one of it's axes is along the centre-line of the 4-4cyls (in the plane S0-S1-S2), rotate it about that axis so that the longest sise is in the plane, then transform the co-ord system back? (...) (20 years ago, 6-Jan-05, to lugnet.cad.dev)
|
| | | | |
| | | | | | Re: Mathematical assistance please —Ross Crawford
|
| | | | (...) Or maybe another way to look at it (got this from old lecture notes): Goal: rotate about arbitrary vector A by alpha Idea: we know how to rotate about X,Y,Z So, rotate about Y by beta until A lies in the YZ plane Then rotate about X by gamma (...) (20 years ago, 6-Jan-05, to lugnet.cad.dev)
|
| | | | |
| | | | | | Re: Mathematical assistance please —Ross Crawford
|
| | | | | (...) Ooooops, I see now I was totally barking up the wrong tree, you need to WORK OUT the alpha angle, sorry. But transforming the co-ordinate system should still work. ROSCO (20 years ago, 6-Jan-05, to lugnet.cad.dev)
|
| | | | | |
| | | | | | Re: Mathematical assistance please —Ross Crawford
|
| | | | (...) OK, here's my VERY LAST word on the subject: Transforming both 4-4cyls S0-S1 and vector S1-S2: Rotate about Y by beta until S0-S1 lies in the YZ plane Rotate about X by gamma until S0-S1 coincides with +Z Rotate about Z by delta until S1-S2 (...) (20 years ago, 7-Jan-05, to lugnet.cad.dev, FTX)
|
| | | | |
