BOLTS Open Library of Technical Specifications

[bernd]

Hi Johannes,

again regarding standards. I had a closer look. To be honest I could not find the REAL standard for the ibeams. The standards which I put in use the ibeams (steel construction standard). They are for sure the govern one for the material. I had a look in a few books but no real success. I don't know if there is a real standard for them. As I mentioned I'm just an engineer ...

They are mostly called european ibeams.

kind regards bernd

[jreinhardt]

Hey,

translations is on the todo list. But there are many many things that one could do, and I have only a limited amount of time, so it will take a while to get BOLTS closer to feature completion (if this even exists )

It is nice to be able to be very specific in the requirements that you have for a part, and standard numbers are a convenient way to express that, so I would like to include standard data where possible. On the other hand, often this data is not available or does not exist, so BOLTS should handle that.

In the case of the ibeams it seems to be not of so great importance in this field, so if in doubt just remove the data. If someone needs it and knows it exactly, it is easy to add it back.

Greetings

[bernd]

Johannes, Thanks for the informations. Found it: http://www.bauforumstahl.de/Querschnittswerte I'm gone change the ibeam.blt as soon as I find some time.

Regarding the year a standard was introduced. Would you take this on the feature request?

Bernd

EDIT
The year could be just behind the standard. It would be part of the name string? Good or bad idea?

[bernd]

@Johannes: I still stick with the fillets ... viewtopic.php?f=22&t=5575

[bernd]

Johannes, thanks for the "fish can" viewtopic.php?f=22&t=5575&start=10#p45864 I'm gone adapt that one for the ibeams.

Regarding file names. Would it be a good idea to call the files ibeam1 or ibeam_european or whatever?

As I sad, I have some imperial beams as well. The data is in metric unit (ca. 150 - 200 beams). Is it a good idea to put them in a different ibeam file? I also have data from some ibeams which do not have parallel flanges and more fillets. Thats why they have some more columns in the data table. Should they be in a different file as well?

Bernd

[jreinhardt]

There are no technical reasons for putting imperial or more complicated beams into a different file, there are even some against it. Splitting requires duplication for the base geometries (i.e. the python functions), which makes everythin harder to maintain. A long as the file is somehow manageable, you can keep it as one. And the files are not very complicated, just long. If at some point it is necessary to split it up, this is easily done, as the data is not interconnected.

Maybe I should try to whip up a syntax highlighting and code folding scheme for blt files. This would make it very easy to manage also huge blt files.

Regarding file names in general: I prefer to have descriptive collection id (=filename for blt file), so ibeam_european would be better than ibeam1.

You can have tables in inch or mm by choosing the types of the columns accordingly ("Length (in)" or "Length (mm)").

[bernd]

Thanks for the fast answer. Close to the borader line.

There are no technical reasons for putting imperial or more complicated beams into a different file, there are even some against it. Splitting requires duplication for the base geometries (i.e. the python functions), ...

Does it mean if there are ibeams which have complicated geometry and thus a totally different python function they have to be in a separate blt-file? Thus I would put all ibeams which use the same pythonfunction in one file? For the ibeams this would be all ibeams which have parallel flanges (metric, imperial, euroean, whatever ...). Means I would call the file ibeams_parallel_flange.blt

There are standard european ibeams which have sloped flanges. They will probably need a sketch in freecad if they should be modelled excactly. Thus they would be in a separate ibeam-file?!

There are c-beams, t-beams, z-beams and castellated beams (Wabenträger) as well. Again parallel and solped flanges. Would it be better to call the files beams*.blt as for the ibeams: beams-i-parallel_flange.blt and so on ?

EDIT
Mhh, there are even rectangle and tubes. All ibeams could be columns as well. People call them beams (in german and in englisch) but at the end technically they are all profiles. I don't know. I leave all this name stuff to you and call they ibeams.blt.

[jriegel]

BOLTS become a really fantastic project! I like the approach (CAD independent, meta information and generate the actual shape on the fly) very much!

Some questions:

Did you think about a Icon/picture per catalog? Makes it easier for the user.

Did you think about length sets? Most Bolts have a certain set of common lengths which are in stock. I'm not sure if this lengths are in the standard...

[jreinhardt]

@bernd:

No. You can have parts with different geometries in the same collection, actually most collections do this. The hex collection for example has several classes for both hex screws and hex bolts, which have different base geometries (e.g. python functions for FreeCAD, modules for OpenSCAD, ...). The structure of this can be seen here:

http://jreinhardt.github.io/BOLTS/html/ ... s/hex.html

Currently it works like that
A class (boxes in the middle) basically describes a mapping from user input to a bunch of parameter values (e.g. type HEA100 and length 1000 to values for the dimensions h, b, ...). It often happens that several classes have the same output parameters, because the parts they describe are similar, like in the case of beams. Then multiple classes can share the same base geometry (boxes on the right), but if that is not the case, each class can have its own base geometry. The idea is to map back as much as possible onto as few base geometries as possible. Because to bring BOLTS to a new target system, all that needs to be done is to provide base geometries.

A class can be exposed under multiple names/standards (boxes on the left). I think that BOLTS should be able to provide the same parts under different names, as someone might be more used to one or the other. But for this the data model is not really good now, I think I will change the blt file format a bit and separate the names and standards from the parameter mapping. This should make things cleaner conceptually, but on the other disperses the different aspects of a part a bit more to different places. I have to think about it a bit more.

@jriegel:

I am glad you like it. If you are interested, we can think about how to integrate BOLTS into FreeCAD more tightly, or if it makes sense to include it in the release packages and installer. Maybe not immediately, BOLTS is still very much in flow, but maybe in the medium term.

Icons/pictures: BOLTS contains and tracks drawings for parts, so it would be possible. The GUI in general can be improved a lot, and it is on my (very long) todo list. The online parts directory shows drawings for parts, e.g.:
http://jreinhardt.github.io/BOLTS/html/ ... O4035.html

Length sets are already implemented, but the FreeCAD gui does not expose them. But common combinations of the free parameters can be specified in the blt file, and are then used to generate the files for the IGES distribution, as there everything has to be known when the files are generated.

[bernd]

Got it, thanks again Johannes.

... I think that BOLTS should be able to provide the same parts under different names, ...

as well as in differnt languages.