Sheet Metal

robert_zschoche

How do I manage sheet metal parts with unfolding and bend allowances

Stuart_Tod

@robert_zschoche ,

Onshape can't yet do sheet metal. I believe a request is 'in' for the capability.

Stuart_Tod

@robert_zschoche ,

Onshape can't yet do sheet metal. I believe a request is 'in' for the capability.

robert_zschoche

Thanks.

brucebartlett

Vote for this as a improvement request

https://onshape.zendesk.com/hc/en-us/community/posts/202922330-Sheet-Metal-Features

robert_zschoche

Hi.  I have been away for a while.  just know in the "CAD" market again.,  No sheet metal is a deal breaker.  Has that functionality been added since I've been gone?

shanshan

robert_zschoche, sheet metal has not been added , maybe it has been under the development process,let us look forward to it together! 

scott_harris

Hi All,

What are the key sheet metal features or functions that are most critical to your workflow?

robin_mccaffrey

scott_harris said:

Hi All,

What are the key sheet metal features or functions that are most critical to your workflow?

In order of importance

Thickness
Flatten
Flanges
Cut across bend/normal cuts
Convert surfaces to sheet metal
Stretch allowance (k-factor, or +material thickness rule of thumb)
Bend Radius
Stepped breaks (i.e. stepped cones/etc)

francis_courchesne

Here would be my list, in order or importance

1. Thicken from sketch (base flange)
2. Flange/Contour flange, with the option to bend only a portion of an edge ("from" selector on both sides)
3. Flatten with stretch calculations (k-factor mostly for me)
4. Drawing flat view with bend axis reporting (print out all bend angles and positions)
5. Proper cut support over bends
6. Automatic corner seams (I mostly keep the same seam across bends)
7. Mirror
8. Pattern
9. Chamfer/radius
10. Convert solid to sheet metal
11. Unbend
12. Bend back (used often with punching)
13. Emboss/punch
14. Manual corner seams

I can be functional when functions up to (and including) item 6 are implemented.

andrew_troup

Just a small thing: PLEASE permit dialling down the number of decimal places for the (automatically inserted) bend angle.

Nothing looks more like the ravings of a rank amateur than a drawing for a  part from a thin material with lots of springback asking the operator to provide Bend Angles at the accuracy implied by "90.00 degrees". It's several orders of magnitude better than what's generally achievable, or what's needed.

I imagine Solidworks eventually addressed this but it was an irritant for many years, and in the meantime it degraded engineering mindsets by repeatedly suggesting to junior engineers that there is no such thing as specifying too much accuracy, and/or no such thing as implied accuracy.

brucebartlett

@andrew_troup, why not just dial it back on the drawing? Some parts I do have compound angles with parametric links to flanges,  so you naturally end up with angles that have alot of numbers after the decimal point. I just round these back to the nearest degree on the drawing to suit the process.

I never really been in the habit of rounding in the 3d model, I always do it on the drawing. I notices in your challenge you specified to have the dimension rounded is this something you would normally do?

andrew_troup

brucebartlett said:

@andrew_troup, why not just dial it back on the drawing? ....

I can't work out what I wrote to suggest I wanted to dial it back in the model.
I want to be able to specify the number of decimals on the drawing, for bend angles.
For many years Solidworks did not permit this; flat pattern drawings showed all angles to two decimals of a degree and could not be truncated.

brucebartlett said:

I never really been in the habit of rounding in the 3d model, I always do it on the drawing. I notices in your challenge you specified to have the dimension rounded is this something you would normally do?

There are two topics here which are not quite the same thing: one is truncating the number of significant digits (and possibly rounding the last one up or down) and the other is shoehorning the dimension into a preferred number series. In the random challenge, it was a case of the latter.
 
I would certainly round (to a sensible "preferred number") the key dimension specifying a cosmetic feature, like chamfer or a radius, in a parametrically size-variable model, rather than in the drawing.
Drawings produced from parametrically variable models should manage themselves automatically, and only require human intervention if they go haywire (at least, that's how I see it).

Partly what drives "preferred number" rounding is because it helps make it obvious to everyone throughout the production chain that the feature does not serve any functional purpose which would require it to be an exact size.
CHF 1 x 45º  has a very different implication for production and value engineering from CHF 1.414 x 50.2º

Furthermore, in many cases this class of cosmetic feature will be produced by standard tooling which will only be available in certain preferred sizes.

 It doesn't make sense to me to have features on the model sized incorrectly with respect to the production drawing, or (another way of saying the same thing) to have the drawing showing a feature which is not to scale.

Another reason to get the model representing reality is that someone else may end up producing the drawing, and I don't believe in having lengthy and complicated handover briefings and notes: I reckon the model should guide the draughtsman as far as possible. 

peter_hall

Really hoping a sheet metal and fabrication environment (like solidworks) comes in soon.

jim_7

Love the concept of Onshape and would make the move immediately. However, not having sheetmetal means another year with SW as we will be renewing soon.

andrew_troup

jim_7
Is it worth renewing, as opposed to simply hopping off the upgrade treadmill, but keep using it for a few more years?

pete_yodis

andrew_troup said:

jim_7
Is it worth renewing, as opposed to simply hopping off the upgrade treadmill, but keep using it for a few more years?

@andrew_troup Great suggestion and one he should contemplate... unless he works with others that force him into needing the 2016 file format support.  Ahh.... desktop CAD locking you in.  Get all your others together and decide to not upgrade and instead spend your money on Onshape accounts and collaborate far better 

I think that will be trend.  People will decide to get off the SolidWorks upgrades, and shift the money towards Onshape.  They will then have 2 CAD packages for less than the cost of the old one.  Call it a 2 for 1 deal.  The change in SolidWorks subscription policy is causing some additional doubt among the users about how long they want to keep paying.

@jim_7 What is it specifically that is holding you back in sheet metal.  Is it the whole sheet metal package, or just a biggie... like flatten?  I'm curious.

paul_stephenson

Flatten is essential step 1 to be able to create 2D DXF for laser cutting real world parts. Other advanced sheet metal functions eg changing k function can be added later. I can not give up SW or even consider OnShape until there is good sheet metal functionality.
Creating a new sheet metal tabs off an existing sheet metal part is the other commonly used tool also placing a bend in a flat sheet, these two features should be added ASAP 

paul_stephenson

Mirroring Sheet metal parts was not added to SW for many years, make sure you consider this as a MUST feature, even if not straight away, but if you don't consider it now, it may be really hard later.

brucebartlett

No.1 for me is the ability to Flatten. Without any other sheet metal functionality this could be used to flatten both imported parts (originally modelled in another sheet metal package) and parts modelled in Onshape part studio as if they are sheet metal, but obliviously with standard Onshape tools. 

Here's a sheet metal project I did in Onshape, I used a very crude method to get a flat pattern. It went something like this 1. derive part to new studio 2. transform by mate connector to the origin and in the correct orientation (more for my drawing than dxf) 3. de-feature bend radius's with remove faces 4. split out flanges with split part, 5. transform the flanges via transform rotate, 6. boolean merge, 7. extrude remove to get edges normal 8. general clean up direct edit tools and extrude 9. export face as a dxf. 

It would be nice to have a tool to do step 3 to 7 in 1 go.  
Screenshot 2015-11-13 21.04.56.png

pete_yodis

Nicely done Bruce.  Impressive none the less.

sistema_perry

Partly what drives "preferred number" rounding is because it helps make it obvious to everyone throughout the production chain that the feature does not serve any functional purpose which would require it to be an exact size. 
CHF 1 x 45º  has a very different implication for production and value engineering from CHF 1.414 x 50.2º  AMS

sistema_perry

I would certainly round (to a sensible "preferred number") the key dimension specifying a cosmetic feature, like chamfer or a radius, in a parametrically size-variable model, rather than in the drawing AMS. 
Drawings produced from parametrically variable models should manage themselves automatically, and only require human intervention if they go haywire (at least, that's how I see it).