Geometry Problem - Window Fitting

[Georgiou]

i am current drafting a Residential Home, that has a complicated window configuration, that needs to fit into a 5000mm Cavity in the Rumpus room.

 

(Please see attached image)

 

 

Requirements of the geometry:

1. The center window panel © needs to be 800mm from the external wall
   
2. All window Panels need to be equal in length. (the same size)
   
3. Window Panels (a),(b),©,(d),(e) are variables
   
4. Window fitting must have 5 panels with 6 angles

How can i determine the correct angles and window panels size to fit the 5000mm Cavity in the Rumpus room?

 

Any suggestion would be appreciated.

 

(Please see image to see what I'm getting at)

[Glen Smith]

I don't know if this is the 'right' way or even the easiest way but I would use the 3 point circle command and the offset command.

 

You have 3 fixed points in your drawing. The two points on the exterior wall at the edge of the 5000mm cavity and the center point of the window 800mm left of that. Draw a circle on those three points. Offset the circle (inwards) the thickness of the wall/windows. Now draw 2 rays from the center of the (concentric) circles to the exterior wall corner points. Measure this angle and divide it up into 5 window bays.

 

Glen

 

whups, sorry, that's really close but not right. I think that working with the midpoint of the wall edge and the midpoint (both x and y) of the window will give you what you need.

[BearDyugin]

[ReMark]

 

Why is everyone suddenly become intelligent, when no longer necessary?

 

Why do YOU ask?

[BearDyugin]

Why is everyone suddenly become intelligent, when no longer necessary?

Why do YOU ask?

Because sometimes comes on a forum (on any) the beginner and sets it would seem a silly or simple question. This question reads the guru, and it even cannot pick up words as to explain such elementary thing, or will name only to a system variable, and the person at times AutoCAD has started for the first time..., or will send in search, and the person more recently at a forum, and it is not guided yet where and what to search, it even formulates a problem plainly cannot, pick up key words for inquiry in search. But users of an average level try to help the beginner, and they offer a decision soya, let not always optimum and true. And here when the theme for a week nevertheless has expanded, here the guru by cannot pass, and starts to philosophise: - "All previous councils are not correct and are a work bad form. It was necessary to take advantage of the program.lsp also gives the reference "(though as a rule beginners of representation have no that with these programs to do and as them load) or "the Problem is solved for a long time and gives the reference to a theme" (and the theme consists of 457 messages of 2 annual prescription, and the name absolutely about other, but there really there is an answer in the message №374, but the reference not on this concrete message, and on all theme entirely)...

or

At the person draw in terms, and it is necessary for it to finish the project today, and it has no time to use F1 help, or search on the Internet. He creates a theme with the request for the urgent help, but all are silent... There Comes tomorrow, and here suddenly at all an enlightenment, and they start to give advice. But already late, yesterday has already passed... And today he where not to hurry up, and it certainly will easy come in F1 help, will look on the Internet and certainly it will understand itself, after all the engineer, but the yesterday's project it has ruined.

 

But I do not say that I well done! I same as you

[PS_Port]

Я бы это сделать так же, как geobuilder

[Georgiou]

thanks for help Geobuilder and Glen.

 

Using the current method i noticed i can't get all window panels the same length. However, for practical purpose I'm sure this solution will be fine.

 

(see bellow for current results)

 

In terms of geometry is it at all possible to get exact equal sized windows?

 

I will post any solutions i come up with..

 

thanks

[BearDyugin]

Using the current method i noticed i can't get all window panels the same length.

In terms of geometry is it at all possible to get exact equal sized windows?

thanks

Well!

But you have lost 33 millimetres instead of 800 at you 767

Look as I did

[SEANT]

All interesting stuff!

 

 

I was curious about a mathematical solution for this but found the equations got sticky pretty fast.

 

I took what I could to Excel – as an opportunity to give the Solver feature a try.

 

This spreadsheet (requires the Solver Addin) will calculate the necessary dimension for a 5 facet Bow Window of any depth or width. Input the desired depth (800 in the case of the OP), then go to Tools – Solver to input the desired width into the “Value of:” field.

 

Edit:

Oops.

I forgot I had hard coded one of my values. Attachment fixed,

FiveFacetWindowSolver.zip

[Georgiou]

Geobuilder,

 

you have been very helpful, however don't understand what your doing with ploy line when type (*0.8(/0.80.7670)

[eldon]

As Geobuilder seems to be master of the visual, perhaps I could explain in words what I think is going on.

 

The first arc was drawn so that the offset to the centre was the required 800. When this arc was divided into 5, and the chords were drawn, the offset to the chord was 0.7670, which is not what was required.

 

What then happened was that the 0.8 length was scaled up by the ratio of 0.8 divided by 0.767 which was calculated on the screen with a lisp expression (/ 0.8 0.767), which gives a factor of 1.043, and the arc was then stretched to this new offset.

 

I apologise if I have got it wrong, but that was the way that I saw it.

 

Graphically, there is no exact solution, but successive iteration gets near enough.

[BearDyugin]

however don't understand what your doing with ploy line when type (*0.8(/0.80.7670)

What for

When I approximated an arch distant from 800, We have received a line distant from 767. Having divided 800 on 767 (/0.8 0.767) I have learnt How long reduce distant 1.04302, have then increased received on necessary for us 800 (* 0.8 1.04302), has received 0.83442.

Thus if we approximate an arch distant from 834.42 we will receive a line distant from 800. Certainly it not is the mathematical formula, and true result we have received only because distance a little, and at bigger distance such focus will not be possible.

Mathematics you can find in a file attached by SEANT.

How

At constructions for a finding of the second point it is possible to enter distances both known, and calculated. For calculation we can use following functions: + - / * sqrt, syntax of record (functionspacenumber1space...spacenumbern).

[BearDyugin]

I apologise if I have got it wrong, but that was the way that I saw it.

You are absolutely right, while I wrote... translate... checked... made out..., you have outstripped me