USER FORUM
(you are viewing a thread; or go back to list of threads)
SolveSpace operating capacity? (by Andrew McIntyre)
I find that as a drawing builds in density the solver gets to be more 'picky' about allowing further constrained elements to be added. Once that density is exceeded things that were acceptable early on no longer are and trigger a 'white-out' requiring the program to be closed and re-opened.
Is this a limitation of the solver, or of the pc's processor - or a bit of both?
I am mainly working with 2D sheet metal templates that will be fed to a laser cutter via the 2d .dxf export facility.
Is this a limitation of the solver, or of the pc's processor - or a bit of both?
I am mainly working with 2D sheet metal templates that will be fed to a laser cutter via the 2d .dxf export facility.
(no subject) (by whitequark)
The solver has several kinds of limits. It will be easier to answer your question if you post a file that triggers an error for you, along with instructions to reproduce.
(no subject) (by Andrew McIntyre)
Whitequark - thanks for that.
I attach the file that appears to be getting too dense. The object is to be able to design a variety of sheet-metal convector-boxes of different shapes and sizes as depicted on our website at http://www.backwoodsman-stoves.co.uk/installation/
At present we produce each box drawing in a standard 2d cad package and insert the appropriate woodburning stove from our library of pre-drawn symbols so that we can check that the required all-round clearances etc are being met. We then forward our drawing to the fabricator who uses it to generate a set of four profiles for his laser cutter.
My objective is to use parametrics to simplify these long-winded procedures so that we merely adjust a few parameters as each job comes in - then pass the fabricator a set of profiles he can feed straight to his laser.
Since Solvespace doesn't offer the option of importing library symbols I have settled for a simple cube that can be adjusted parametrically to match the dimensions of any stove, but this last 'wrinkle' threw up all kinds of problems. In particular trying to get Solvespace to accept the diameter and location of the flue connector of the stove. I achieved this at the first attempt on a separate 'blank sheet' but it took several attempts before I could persuade Solvespace to incorporate it into the existing drawing - and I'm not sure that I haven't destabilised the drawing by pushing things too far since almost any further amendments trigger a warning that the file can't be saved.
This first effort depicts one of our "standard" boxes but it would be useful if we could extend it to allow a box to be tapered towards the rear and/or have an arched top. However at this stage I don't want to push my luck . . .
Interested to have your views!
I attach the file that appears to be getting too dense. The object is to be able to design a variety of sheet-metal convector-boxes of different shapes and sizes as depicted on our website at http://www.backwoodsman-stoves.co.uk/installation/
At present we produce each box drawing in a standard 2d cad package and insert the appropriate woodburning stove from our library of pre-drawn symbols so that we can check that the required all-round clearances etc are being met. We then forward our drawing to the fabricator who uses it to generate a set of four profiles for his laser cutter.
My objective is to use parametrics to simplify these long-winded procedures so that we merely adjust a few parameters as each job comes in - then pass the fabricator a set of profiles he can feed straight to his laser.
Since Solvespace doesn't offer the option of importing library symbols I have settled for a simple cube that can be adjusted parametrically to match the dimensions of any stove, but this last 'wrinkle' threw up all kinds of problems. In particular trying to get Solvespace to accept the diameter and location of the flue connector of the stove. I achieved this at the first attempt on a separate 'blank sheet' but it took several attempts before I could persuade Solvespace to incorporate it into the existing drawing - and I'm not sure that I haven't destabilised the drawing by pushing things too far since almost any further amendments trigger a warning that the file can't be saved.
This first effort depicts one of our "standard" boxes but it would be useful if we could extend it to allow a box to be tapered towards the rear and/or have an arched top. However at this stage I don't want to push my luck . . .
Interested to have your views!
(no subject) (by Andrew McIntyre)
Apologies - for some reason the website link above requires you to scroll down to option 13 "Recessing a stove".
(no subject) (by EvilSpirit)
You have overconstrained the "problem element". Just remove H constraint on the line inside circle.
Some known specific reason SolveSpace can't analyze degrees of freedom for horizontal/vertical constraints.
Also, problems with file saving/loading can be because of using text entity. You can use Constrain->Comment instead it to make comments for your drawing.
Some known specific reason SolveSpace can't analyze degrees of freedom for horizontal/vertical constraints.
Also, problems with file saving/loading can be because of using text entity. You can use Constrain->Comment instead it to make comments for your drawing.
(no subject) (by Andrew McIntyre)
Thanks for that. All now seems to be well. I have not previously used Constrain>Comment - it clearly imposes much less on the program than the text entity. Thanks for the tip!
Post a reply to this comment: