Reporting angular parameter error - host references

[jnoordzij]

I am trying to set up a reporting angular parameter in an adaptive component family. 

When I use the reporting parameter in a formula I get this error:

"A reporting parameter can be used in a formula only if it's dimension references are all to host elements in the family"

What I tried at first is to make sure that when I set up the angular dimension it is referenced to reference lines. But still I get the error. 

After that I found this piece of info on Revit Forum with Alfredo Medina saying in post 3 that: 

"

angular reporting parameters (...) can detect and report the angle between (...) the angle of a reference line and one of the planes of an adaptive point

" 

So I tried that as well. But it still doesnt work. 

I will attach the family. The reporting parameter 'hoek_reporting' is already set up. 

Now I want to use it in a formula for parameter 'seg', with the formula being: seg = Lx1 / sin(hoek_reporting). 

I keep getting the error. Hope somebody can help me out!

[bin]

Try to get this angle.

[jnoordzij]

---

@bin wrote:

Try to get this angle.

---

Thanks for you reply! It looks like you created an angular dimension between a placement point plane and a reference line 'end'-plane, am i correct?

Sadly I am not able to reproduce that 😞  My placement point planes are visible, but my reference line 'end'-plane is not. If I set work plane to reference line 'end'-plane I cannot reference to placement point plane. 

Can you describe to me exactly the what you did to create yours?  Many thanks in advance!

[bin]

it Looks like that work plane(perpendicular to the reference line) only exist when there is only one reference line defined by that control point. 

What you can do is to place a point on that placement point and use the new point to define the other reference lines. 

Let me know if this works or not. I’m also new to this. 

[jnoordzij]

Next thing I tried: 

-create a dummy reference point on Z=0

-set the work plane to vertical plane front/back

-create a vertical dimension between PlacementPoint1 and Dummy Point (PP1_Z)

-create a vertical dimension between PlacementPoint2 and Dummy Point (PP2_Z)

-try to use these values in a formula: testlength = PP1_Z - PP2_Z

.. Still the same error. 

Placement Points and Reference Points are both 'Host Elements', right? 

I also tried referencing PP1 and PP2 towards 'Ref.Level', which imo definitely is a 'Host Element'. Also did not work.

[bin]

I think in your family, the only report parameter that you can use in your formula is the length between the 2 points and possibly the angle formed from the points and the perpendicular work plan(not even the reference line between).  

Its a bit late here and I’ll try it tomorrow.  🙂

[jnoordzij]

---

@bin wrote:

it Looks like that work plane(perpendicular to the reference line) only exist when there is only one reference line defined by that control point. 

What you can do is to place a point on that placement point and use the new point to define the other reference lines. 

Let me know if this works or not. I’m also new to this. 

---

You my hero! 

I started off with drawing a single reference line. As you said, as long as there is nothing else attached to it, then both work planes exist. So i was able to angular dimension those planes. And I also was able to use the dimension as a reporting parameter in my formulas. End result is fine!

Many thanks

[bin]

Good to hear. 

[jnoordzij]

@bin Scratch that .... 

As soon as i move Placement Point the angular dimension get's lost 😞 

[jnoordzij]

I have managed to create the dimension so that it does not disappear upon moving the placement points. 

But now it keeps flipping back and forth so I cant really use it. 

[bin]

I think it’s fine. 

Because cos(a)=-cos(180-a)=cos(180+a)=-cos(360-a)

You just need to use if() to make the final result a positive number. 

Just make sure that the angle does not equal to 90. 

[jnoordzij]

You're probably right, but I cant really get my head around that for a quickfix. Plus, sometimes angle will be 0. 

I did a workaround for now:

-I added two more placement points (pp1', pp2') which will be placed exactly below pp1 and pp2, and both on Z=0

-From there I measure height between pp1<>pp1' = pp1height, pp2<>pp2' = pp2height

-Then I compare pp1height and pp2height, which gives me height difference between pp1 and pp2 (this can be 0)

-That value, along with the line length between pp1 and pp2 will give me an angle. 

I can work with that for now. 

Thanks anyway @bin ! I might do further research some other time. 

[bin]

Since you are using the formula a=b/cos(A), although the dimensions of the angle flips around and give you the figure of A,180-A,180+A and 360-A randomly, the result of cos(A) will be either c or -c.

For example you move the points so the angle equals 60 degree and you expect to get the cos(A) as 1/2, but the dimension might flip to the other side of the reference planes and give you the figure of 120, 180 or 300. Cos(120)=-1/2 , cos(180)=1/2, cos(300)=-1/2.  So you can just use your formula and make sure it’s not negative. 

I did a search and looks like you can use abs() for this situation. 

Also this formula will not work when the angle A become 90 because you can’t divide a number by 0, what I can think of is to create another parameter c, so a=b/c, and c= if(A=90,x,cos(A)) something like this. 

I don’t think your 2 extra points workaround will work as expected because those 2 points will not alway stay right below the other 2 points. 

Please let me know whether it works or not, or if you have other ideas or better solutions. 

[bin]

For example, you move the points so the angle equals 60 degrees and you expect to get the cos(A) as 1/2, but the dimension might flip to the other side of the reference planes and give you the figure of 120, 180 or 300. Cos(120)=-1/2 , cos(180)=1/2, cos(300)=-1/2.  So you can just use your formula and make sure it’s not negative. 

Sorry, I did some mistakes, 180 should be 240, cos(240)=-1/2 and cos(300)=1/2.