More...
Goal 1:Learning Solidworks
First off, learning solidworks, that part at least has worked a treat, I am much more proficient at drawing and trouble shooting problems that occur. However I'm running into a massive limitation, solidworks is a very heavy program.
This screenshot wasn't taken at peak times, just during normal drafting where each movement and input was taking 10-30 seconds sometimes upwards of 20mins just for a rebuild. Needless to say this is quite frustrating, during rendering images however the CPU load was at a constant 100% and the temperatures went to 59°c (3°c below failure point) so excess cooling had to be implemented.
So learning about the program was a big tick, the second goal, to see if it is capable of being an architectural design tool, that's a little harder to answer.
Because I have run into so many issues with the speed of drawing and the overall complexity of the model my second goal has really suffered. I had intended to create a full animation of the operable panels on the building along with some fair flow studies of how the space would ventilate, but the panels were much harder to draw than I anticipated. They would require me to draw each individual panel which if I were to do I would still be doing in December, so I'm going to have to rely on a diagram and a bit of imagination to get that element across. Even the flow studies would take a week or so to calculate on a model that large, I was considering doing a small detail of the flow study but haven't gotten to it yet.
The second problem I was hoping to tackle was the structural simulation of the domes, to see if they would stand up. This one I am really sad about not being able to get to work. The way solidworks computes these problems is to create a mesh of the model that breaks the components down into bite sized chunks of model that it can then analyse. Prior to this subject I had been doing some stress tests for a modular system I have been designing. I was applying a force to a 35mm SHS about 1.5m in length with an array of holes cut into it, when I finally got the mesh to create (it fails to create if the setting or model aren't quite right) it took 2 days to calculate the results.
Unfortunately I couldn't even get the mesh to create for my dome structure, the problem seems to be in the way I have drawn the domes. If I had have drawn each structural member as its own part (attempt shown in the next drawing post) then I think the model could have dealt with it. But Since I had drawn just one triangle element and replicated and scaled copies around to form the structure the meshing always fails to create, and therefore it cant run the structural simulation study.
Fortunately I have run into this problem before and this is one of the reasons that I decided to go for simple dome structures over complex irregular shapes, because even without the structural modelling simulation it is a tried and proven structural shape. The only sections I would be slightly concerned about would be the points where three domes intersect and the cut sections around the perimeter.
Another feature I was hoping to play with was the thermal transfer simulation, which could be used to calculate how the dome spaces would heat up throughout the day and how effective the insulation and ventilation could be at counteracting and stabilising the area at a comfortable level. This was the feature I had most been looking forward to applying to a building because for me it is basically the basis of designing buildings.
Unfortunately the complexity of the model was again hindering my implementation of this feature, it failed to run on the full model. Prior to this realisation though I ran a test on a basic model of Paddington central, the result failed to complete because "solid bodies have melted". This was surprising as the model was designated to be made of concrete with the heat element apparently set to normal sun levels, as far as I know concrete doesn't melt in sunlight. So I wasn't quite sure what to do about that one, I think the error lies in my implementation of the study and its parameters.
Aside from all the studies not working for me, I think the idea of using these features to aid in the design of a building is a sound concept, just the experimentation of it would best be confined to a smaller building concept.
Next time I'll get it to work.
Sources: [Original Works]
No comments:
Post a Comment