Dynamic blocks

[NightMist]

Dear Community,

 

I am not an autocad professional and I am not trying to be one. But I produce windows and doors. and I reproduce the same drawing in the attachment a lot.

 

Is there a way to make it a dynamic block where I change the three dimensions in the drawing and the whole thing grows or shrinks symmetrically and proportionally?

 

Regards,

dynamic block.dwg

[nestly]

Symmetrically and proportionally? That would just be straight scaling I believe. I expect you want to be able to adjust the height and width independently, like the overly simple sample below, correct?

[NightMist]

True, I would like them to be independently adjustable like the picture you drew. But it has to have one more parameter... the arch's height. In your picture the arch's height is fixed

and instead of moving cursors, I would prefer to enter the size directly (or enter the size increment for example if it is 1000 and I want it 1250 I would add the 250)

regards,

[dbroada]

before looking too deeply a couple more questions

 

when the height of the arch doesn't equal half the width, what controls the shape? Do you just stretch the tip and let everything else follow?

 

do you have a limited number of set sizes or can they by anything you want?

[NightMist]

the shape must always be rectangular from the bottom and a connected arch on the top. the width of the arch always equals the width of the base.

size can be anything. the circle can be half the width, but also it can be less than that (in height I mean)

[dbroada]

I'm leaving this now for somebody else. I can't control the shape of the arch if its not a full semi-circle. Good luck.

[NightMist]

pros please help?

[neophoible]

Sorry, I'm not a pro, but if you would post several example results it would be very helpful to see precisely what you are hoping to accomplish. If dynamic blocks won't do it, then perhaps a LISP routine will. But without precise information, it's not worth looking at more than I already have. BTW, I don' t think the arch's height is fixed in nestly's example. If it were, then, when the width changes, it would look a lot different. The radius changes with the width.

[neophoible]

Symmetrically and proportionally? That would just be straight scaling I believe. I expect you want to be able to adjust the height and width independently, like the overly simple sample below, correct?

[ATTACH=CONFIG]42689[/ATTACH]

nestly, how did you get it to adjust without the values showing? When I set up such a block

using stretch parameter sets, it shows an editable value as I stretch, which I think is one of the criteria.

[nestly]

The first example is just a closed polygon with a couple of tangent geometric constraints. I had the full dynamic block almost done, (it took a ton of tangent, perpendicular, and concentric contraints since the two sides are not symmetric) but then the OP said the arch wouldn't necessarily always be tangent, so I called it a night and frankly forgot about it until now. Typically, when I need to control stretch dimension by entering values, I make them available on the properties palette and adjust the part from there, no grips needed.

[neophoible]

The first example is just a closed polygon with a couple of tangent geometric constraints. I had the full dynamic block almost done, (it took a ton of tangent, perpendicular, and concentric contraints since the two sides are not symmetric) but then the OP said the arch wouldn't necessarily always be tangent, so I called it a night and frankly forgot about it until now. Typically, when I need to control stretch dimension by entering values, I make them available on the properties palette and adjust the part from there, no grips needed.

Thanks for the explanation. I did notice that the block wasn't drawn perfectly symmetrically, but I think the intention was for it to be symmetric. I tried using the Autoconstrain option, and then set it up from there, but decided to abandon it as well, as the stretching didn't look so hot. Without some serious input from the OP, as I requested earlier, the pursuit isn't worth it to me. Too often, I see people try to help only to find out that the OP had something else in mind.

[dbroada]

Without some serious input from the OP, as I requested earlier, the pursuit isn't worth it to me. Too often, I see people try to help only to find out that the OP had something else in mind.

pretty much why I pulled out at post #6

[NightMist]

I drew a few examples where the base is an initial drawing then

picture A we change the width from 1000 to 1200

picture B we change the height from 1500 to 1650

picture C we change the height from 1650 to 1800 and the arch is half the width (just a special case of the possible drawings)

 

I hope you can help me now

[Chilidawg]

In those three drawings, is the stile the same height and all you are doing is changing the height of the arch?

[neophoible]

In those three drawings, is the stile the same height and all you are doing is changing the height of the arch?

The 1200 vertical dimension is the simplest to change, so not really an issue. However, the mitered arch is not so easily handled, it seems to me, and it can change in width and/or height. I'm not sure if anyone knows a way to do this with dynamic blocks, but I will certainly be interested to see. If this cannot be done via blocks, it could still be done with LISP. The main thing is to give the programmer all of the parameters up front so he knows what he's dealing with from the outset. Of course, what is shown here is just the basic frame. It omits quite a bit of detail from the original dwg. But at least now it's easier to see what is changing on the arch.

[Chilidawg]

It seems to me that there are likely to be some general contraints* in the manufacturing process that would limit the number of available arc radii for any given width. I dont think an infinite number of arcs is practical or desirable.

 

 

 

 

*not to mention asthetic considerations.

[nestly]

... the mitered arch is not so easily handled, it seems to me, and it can change in width and/or height.

 

The line representing a miter would just be a line with coincident constraints to keep the endpoints at the end of the arc segments. The reason I haven't invested any more time into this, is because the design criteria hasn't been define. ie there were no miter lines in the original dwg. Also, if this is just a generic window, it's not that difficult, but to build in individual adjustments for sash, jamb and sill gets pretty involved, and I'm pretty sure is going to be a required feature as the window gets bigger/smaller ... and then there's grill options....

[NightMist]

"If this cannot be done via blocks, it could still be done with LISP."

As long it can be done... Pls show me how to do it... I am not so attached to dynamic blocks.

 

As for the dimensions... Side height, total height and width can be anything. Of course some constraints apply but this is not our issue at the moment. The architects will not design a window outside the standard constraints.

[neophoible]

"If this cannot be done via blocks, it could still be done with LISP."

As long it can be done... Pls show me how to do it... I am not so attached to dynamic blocks.

 

As for the dimensions... Side height, total height and width can be anything. Of course some constraints apply but this is not our issue at the moment. The architects will not design a window outside the standard constraints.

So, if dynamic blocks are too difficult here, my recommendation is to ask the moderator to move this to the LISP forum and rename it custom window design (which it should have been in the first place, really). This way they can see what has already been discussed. Also, go ahead and complete the windows in the examples so that they are truly representative and clean up the drawing in the OP, unless it is not really supposed to be symmetric. And post the drawings as DWGs so that they can look at it closely. And don't ask them to show you how to do it, unless you already know LISP and want to learn how to apply it here. Just ask if anyone is willing to tackle the problem for you.

[nestly]

IMO, this is not a particularly difficult dynamic block, the problem is that not enough information has been provided to build it, and that will still exist in a lisp based solution, perhaps even more so.