Deriving Unitless Parameter From Parameters With Units

[Jsoares1994]

I am designing a wooden simulation rig to hold my wheel, pedals, and shifter for my simracing hobby. I am attempting to make everything as modular and adjustable as possible for the sake of convenience. One such way I plan on doing this is by having regularly spaced holes along the bottom of the frame to allow for pieces such as the seat mount, steering wheel mount, pedal board, etc to be moved and adjusted independently. That said, I am not 100% sure of the final dimensions of the project. I am using parametric modelling to allow me to make easy changes to the dimensions of the design without wasting significant time, however I can not seem to come up with away to parametrically create the frame holes.

I am attempting to use a rectangular pattern to repeat a single hole (whose location is determined parametrically) along the length of the frame. The extent of this pattern needs to be RigLength-(HoleOffset+PlankThickness)*2. This formula works because I am defining a length in terms of other lengths. Having my holes spaced at 1" intervals, I need RigLength-(HoleOffset+PlankThickness)*2+1 holes. Attempting to plug this into the "number" field of the rectangular pattern, it turns red and I am unable to finish the pattern, likely due to the type mismatch of no unit vs length. The same situation occurs if I try to create unitless parameter "NumHoles" with the same formula.

I have attached a (poorly) annotated picture of my design so far to help illustrate the geometry in question.

I have tried quite hard to find a solution to my problem but have had no success. Can anyone help point me in the right direction?

Thanks!

[jeff_strater]

@Jsoares1994 - the Fusion units system can be a pain to deal with.  It is very literal.  Think back to high school science and "units cancellation".  If you want to end up with a unitless expression, sometimes you will have to divide by a length parameter.  For instance, if you have these parameters:

• x = 3 mm
• y = 4 mm
• z = 5 mm
• a = 1 ul     (ul is "unitless")

Then, if you have an expression such as: "x + y*2 + z", the units of the result will be "mm".  So, to get a unitless result from that you would have to change the expression to "(x + y*2 + z)/ 1mm"

If your expression is "x*z", the result is "mm**2"  (millimeters squared), so if you want to use this in a dimension (which requires a length value), you would have to change it to "(x*z)/1 mm", and if you want to use it as a unitless value (in a count), you would have to use "(x*z)/(1 mm * 1 mm)"

[drisler8109]

Hello I found a way to make this work, but it is ugly.

If you want to scale a length - mm value (as an example) by a unit less constant, you can do this if you also divide the value by another unit less variable (set it to “1”).

The reason this works is that (unit less-variable / unit less—variable) results in a truly unit less sub-expression, which can be freely multiplied by a length - mm value.

This is very stupid, “unit - less” data type should not require any conversions.

For example, the built-in value “PI” (which is a ratio of length values) somehow can be multiplied by “length-mm” values, without any crazy stuff. Whatever code magic was done to make “PI” work in equations - this is what is needed.

The same thing applies to “percent - %” data type; which should always be “unit - less”.

There are plenty of constants in physics with units, and there are also plenty (such as PI ) without any units. If a constant has no units (it is unit - less) you should be able to freely scale ( multiply or divide) a value by such a constant.