Merging spaces vertically? or connect with holes?

[ColinR]

Hi,

Quick query about merging spaces to make them a double storey height.

Sometimes when two spaces stacked vertically are "merged", all sorts of intersection errors occur. If they are joined with a hole there are no errors. Is there any difference in how the building will perform when these two methods are used? Any difference in results?

Cheers,

C

[PCully]

Results wise there shouldn't be any problem with either approach but remember spaces separated by a hole are different from one large volume so think along the lines of how MacroFlo treats this or what quality you would need to set in a lighting analysis to calculate penetration through the holes.

A merge is better in terms of it trying to create a complete clean geometry but it is also harder for the software to get that right every time when the geometry gets more complex so I do recommend using it but as long as you are keeping an eye on the quality of the geometry (intersection etc) as you go and remember to undo when you see big problems being introduced so that you don't get to the end and have such a poor quality geometry that you can't proceed to the analysis part.

Phil

[ColinR]

Thanks for that Phil. Much appreciated.

[ColinR]

Hi Phil,

Sorry, I was just reading the post back and when I say join "with a hole", I mean joining them using the "connect spaces" tool, "remove partitions", and then pressing "Connect".

Will that give different results when running simulations than if choose "merge"?

Cheers,

C

[PCully]

It can do, Merge means the volume is a complete shape (where it is possible) so there is no hole in between parts of the volume. I'm just thinking along the lines of running with MacroFlo or quality settings on your Flucs analysis where you need to get to Medium before light from internal openings is taken into account.

In general there may be some small difference but it won't be an extreme discrepancy so modelling in either way is valid. I've never studied it in detail but some quick tests will give you a good feel for it

Phil

[ColinR]

Hi Phil,

Thanks for the information. I found that when running Part L calcs it makes a difference on Criterion 3 checks as it doesn't seem to consider the "connected" spaces as one whole space giving two different benchmarks and in my case, the upper area failed by 850% whilst the lower area passed!

I think the moral of the story is, get the geometry correct if you want an easy life

thanks again,

C