Rotation / Mirror problem in revit

Hi,
Im taking data out from Autocad as X Y and rotation coordinates.
Importing these data in dynamo and putting out family on that points.

Problem is autocad rotation and revit is mirror image.
Im not getting any clarity how to should I modify the node. I tried mirror an element node, and few other.

Files to download:
dynamo 2.0.2
revit 2018.3
PlaceComponentsFromPointList 2.dyn (52.1 KB)

test file autocad to revit.rvt (3.3 MB)

sample 5 points rotate.xlsx (9.3 KB)

@andre.abotnes thank you for your dyn file. It helped me for XY coordinate.
https://forum.dynamobim.com/t/place-family-instances-by-coordinate-from-excel/18947

Try changing the code block containing the X rotation angle to 90-X.

Yes I had tried already. Then what happens is, it also rotates for element having 0degree. i.e. 0degree turns to 90degree.

Are both files using the default origin/ucs of 0,0,0? Otherwise you might need to use a coordinate system and Point.ByCartesianCoordinates

1 Like

Yes it is on same origin. It just have rotation problem. :sweat:

Does your Revit file use a different Angle to True North?

Orientation: True North.

I tried both ways. Project north and true north. same result.

I just did a test data set, and it worked fine.

Your initial angles are off so you will have to adjust your values though. In Revit family rotation is always clockwise from the X axis. That is if you previously had a the angle shown in your image, you’ll need to subtract the value from 90 or subtract 90 from the value to get the right angle.

Observe top 2 elements. for X and 90-X;

I think mirror would help but I cant see any way to fit. :sweat:

Elaborating above images:

1st image having X; 0degree has perfect alignment but not for 70degree.

2nd image having 90-X; 70degree has perfect alignment. but those 0converted to 90degree is not in alignment.

How about you rotate it the other direction by inputing -1*x?
It seems that the family’s local origin is at the top of the object, and thus the rotation will seem to be oposite. I believe it would fix the problem to either change the family or switch the direction of rotation by the *-1 method.