Rotate portion of design, Pivot keeping relationship?

have you tried using Orient3pt?

1 Like

I use that when I need to normalize the whole capture to the plane. If I use that for this small section, I still need to tell it where to go, right? I need 3 reference point, then 3 destination points which could be anywhere. BUT… hmm… if I reuse the two points from the drawn line and the 3rd on the plane, that may work. Is that what you were thinking?

I dont think that would work unless I’m not doing something correctly. I’ve attached the model to the post if you want to mess with it.

I’ll have to play with this more in the morning. This might work but I think order of reference and order of targets matter. It also appears I can only do one “point” at a time.

If I change the order around the new point can either be WAY off or if I do it in another order it offsets like an inch or so which is expected for a rotation arc.

Red is the new position which may or may not be right? /sigh (=

Grouping points to your objects often helps with future rotations. You can add extra points to create a hinge-like axis for rotation. The 3D rotate works well for this.

Orient3D is one of those commands I never get right the first try.

If accuracy is paramount, protect the original by making a copy of the object, and laying the copy down, and or make a back-up layer. Blocks also work for this, too.

By using points, and the [TAB] key for lines, temporary lines, and snaps, I have only once needed construction plains, and that was for laying out an control panel, on a angle.

1 Like

Someone else was asking about rotating an object and I posted detailed instructions here: Raising a shape with each corner at a different height - #4 by jeremy5

You may be able to adapt the concept to your model.


1 Like

If you want the “hinge” line to lie along the edge of the in-plane piece then you will have to twist the two points, i.e. rotate one more than the other.

1 Like

I would use Orient3Pt with Scale=No with three reference points on the object and three target points on the CPlane.

The first target point determines where the oriented object will be located on the CPlane.

The second target point determines how the object will be oriented on the CPlane. It’s distance from the first point does not matter, only the direction from the first to second points.

The third target point needs to be on the CPlane. It cannot be colinear with the first and second points. Which side of a line from the first to second point the third point is on determines whether the result appears to be “right side up” or “upside down”. Otherwhise the location of the third point does not matter.

Result using Orient3Pt: CushionsDC1.3dm (90.3 KB)
Use of Orient3Pt (Blue scribble shows location third reference point.)

1 Like

Orient3Pt will yield the same results in this situation.

Agreed, if we are simply orienting the two points onto the c-plane. But if we want to include an edge to match the existing one, ie to start building the parts for a 3-d cushion, then the twist is going to make life more complicated. (and renders my original suggestion worthless too, of course)

(I’d probably go old-school: lay fabric on the deck and mark the points, the put it on a flat table, measure the distances and transfer to the drawing.)


That maybe true using Orient, but Orient3Pt (different command) has an additional rotation using the third reference and target points.

Thank you. I’ll check out that article and see what I can gleam from it.

That’s why I think using Orient3Pt in the correct order 1 dot at a time might be the trick. Reading the other comments suggest lots of great options.

Quite, but using another boat analogy, if you apply any orient3pt transformation to a spinnaker you don’t get a flat sail. Here the two points and the edge do not lie in the same plane. If the relationship between the edge and the two points is important (and I am only speculating that it could be) then orient3pt is not going to provide a solution. We are into the dark art of surface flattening (although as we are working with flexible materials and the bend isn’t large, we may get away with it).

Sorry, but I don’t think that will help you. I misread your scenario at first reading

If we look at projecting a single snap (dot) based off either newly added temp dots or using the dots from the other snap position, I should be able to orient one dot at a time accurately, right? I believe this was already mentioned in the chat in a similar manner.


(although as we are working with flexible materials and the bend isn’t large, we may get away with it).

Completely agree… we’re not building a rocket. Its off season so I have time to learn some better practices so if I were busy, I’d just have a 2x2 snap patch and snap it on the boat. Which I’m still going to do more or less.

I think what I did in my recent screenshot post works and is basically like what you describe here. I’m still on v5 so I can open your 3dm but I get the gist I think. I dont have a scale option in v5 either it looks… just Copy=yes/no

Attached is a copy of the file in V5 format. Scale option was added in V6. In V5 the command is equivalent to V6 with Scale=No.
CushionsDC1.3dm (76.8 KB)

1 Like

Eric, do you work with curves or surfaces? If you use surfaces the UnrollSrf (for planar or developable surfaces), Squash and Squish commands will flatten the surfaces onto the CPlane.