6 ms·

I had a tremendously embarrassing failure in a sheet metal part designed in Solidworks due to a bug in the way bend allowance is calculated. The long story shor

by sammcf 10y ago

I had a tremendously embarrassing failure in a sheet metal part designed in Solidworks due to a bug in the way bend allowance is calculated. The long story short is that the flat pattern developed by SW was about 10% dimensionally inaccurate which resulted in a rotating part binding up in the field and going blammo. This was years and years ago as a fresh young designer, so I reported the issue to my reseller and the response was essentially "yeah, you're gonna have to take that into account." I've since learned that this sort of idiosyncratic difference in calculating certain values, particularly things like bend radii and bend allowance, things like beam extensions on structural frames, etc, is par for the course for CAD packages. Pretty much every engineer I know has learnt this lesson the hard way. I guess it's not really a bug per se, just hidden behaviour that you have to learn about and take into consideration.

rtpg 10y ago
I don't get this though... Like if the software calculates bend radii that will break things, isn't that a bug? What's the explanation for that even happening? Is it just that the formula involved means that you can't always get a "right" answer, so the software will be like "well, this is the best we can do, figure out the rest"?
- coldtea 10y ago
  >Like if the software calculates bend radii that will break things, isn't that a bug? Well, it's not like all companies fixes their bugs or care about them. If, as the parent says, all CAD packages have such issue, what you gonna do?
  - chris_wot 10y ago
    The difference in LibreOffice not rendering an EMF correctly and a bug that causes wrong dimensions on a drawing in a CAD tool is that LibreOffice gives you a bad looking document and a CAD program costs you money when the material starts to be cut up and it all had to be thrown away as scrap. Mostly I can fix the LibreOffice bug and make the document render correctly (unless it physically corrupts the document when saving), once physical material gets an error it's often impossible to fix.
- chris_wot 10y ago
  So what are you gonna do? Move to another CAD package? Take a look at the reasonably large number of CAD packages listed on Wikioedua here: https://en.m.wikipedia.org/wiki/Comparison_of_computer-aided_design_editors https://en.m.wikipedia.org/wiki/Comparison_of_computer-aided... Virtually all of them are proprietary. They are too expensive to move away from, and every one of them has flaws that firms workaround and build there workflows on. Retraining, proprietary file formats and the disruption of migrating an essential part of most businesses is prohibitive. Frankly, IMO the CAD marketplace is an example of how closed source really bites an industry. Funnily enough, most engineers - who would never accept inaccurate tolerances on items like screws or bolts - seem to accept this is the way it is and despite the stories we read here let the situation continue. I'm frankly amazed there aren't more failures in infrastructure from bugs in CAD software. That's more a testament to the fact that engineers are engineers though - when they realise there are problems, they dig in, figure out where the problem is and find a solution, and in the case of CAD software I really think they build or design their way around the problems. Overall, CAD software is, despite all the flaws of being coded source, still pretty amazing and vendors do fix issues eventually. We can do so many more things in society because of it, but as we evolve as a society it becomes apparent there are limitations. Eventually CAD software vendors will be disrupted, but we're a long way from that point right now!
  - rtpg 10y ago
    I mean sure, things have bugs, right? I get that. Sometimes autocorrect messes up things I type. It's not really a bug per-se, more just a side-effect of how things like autocorrect work, right? But sometimes Chrome crashes, and that's a bug... The OP seems to say that the bend radius thing is more of the first case than the second, and I'm wondering how true that is. I'm not really sure of the science behind it (maybe it's an impossible constraints problem).
    - chris_wot 10y ago
      Oh, no doubt. The issue is that the tools don't appear to document their limitations and constraints, and there is often no way of knowing until you get bitten. At least with OSS you can look at the code behind the program to work out how it calculates things like the bend radius. I'm no engineer, so I had to look it up (Wikipedia was no help whatsoever on this occasion) but I located this page which gives the formula and an explanation: "Calculating the correct flat pattern layout is crucial to getting a good quality finished part from your press brake. Yet, many CAD and CNC programmers have no idea how to calculate the required values. Years ago, the real experts created cheat sheets and tacked them to the wall. They only taught the new apprentice how to apply the results shown on the cheat sheet, not how to calculate the numbers. Well, now those experts have retired and it's time for a new generation to learn the right way to do the calculate the correct flat pattern layout." http://www.sheetmetalguy.com/bend-allowance.htm http://www.sheetmetalguy.com/bend-allowance.htm Now admittedly the guy does put the responsibility on the operator, but it appears CAD packages can indeed calculate this for you. So long as the right info was input into the program, the formula should be programmed correctly and the operator should be able to have confidence in the results!
- SixSigma 10y ago
  It is a formula but you need to know the physical characteristics of your metal.
- malka 10y ago
  Or it's on purpose. This way you can sell training courses and each user is locked on it's cad implementation due to the cost of retraining on another tool
kazinator 10y ago
> "yeah, you're gonna have to take that into account." What? But the software was supposed to take that into account, which is why it it supports bend allowance calculations in it! (I have a tiny bit of experience with this; I designed some sheet metal thing. I somehow clued into bend allowances before doing it, even though it was my first and only one! I figured, you can't just draw the naive shape; the metal has a bend radius. How the hell does that work? [half a day of Googling] Oh. I used a 2D drafting program that had no clue about this, so I manually drew the pattern with the allowance in it already, based on the thickness of the aluminum.)
- sparky_z 10y ago
  > But the software was supposed to take that into account, which is why it it supports bend allowance calculations in it! Structural engineers aren't allowed to treat software as a black box and trust that it works correctly. They're legally responsible for choosing their tools and verifying that they work correctly.
  - kazinator 10y ago
    I.e. the "that" in "that into account" means the tool being possibly wrong. There is no, "sorry the TV blew up in your face; my scope was out of calibration". :)
  - chris_wot 10y ago
    Uh, I don't think there is a better example of closed source than CAD packages.
    - afuchs 10y ago
      I don't think he was talking about verifying the software itself. Even if the software was open source, would it have been formally verified to make sure it has zero bugs? It is better to check the calculations by hand to make sure that no critical errors make it into the final design.
      - 10y ago
cm2187 10y ago
I don't know how expensive is the stuff you were making but if it was just about cutting metal it can't be that expensive. Isn't testing samples the 101 of engineering or have we gone 100% trusting CAD now? It looks to me that there are so many things that can go wrong in real life like deformations because of temperature, different calibrations on different models of machine, different composition of the material used, etc...
- sammcf 10y ago
  Metal isn't that expensive I guess but still, several hundred dollars and a turnaround time increase of a week+ minimum isn't always on the cards for every project. Sure, if I had unlimited time and budget, I'd physically prototype and field test everything. Client demands tend to take precedence. Part of being a good engineer is balancing those concerns, learning which parts of the design and manufacturing process can be trusted implicitly and which assumptions must be rigorously tested - suffice to say I've been a lot more stringent about sheet metal design in the years since.
PhasmaFelis 10y ago
> I reported the issue to my reseller and the response was essentially "yeah, you're gonna have to take that into account." > I guess it's not really a bug per se, just hidden behaviour that you have to learn about and take into consideration. I'm not that kind of engineer, so I may be missing something, but it sounds like the reseller snowed you into blaming yourself for their faulty software. I get what others have said, that a structural engineer is responsible for the things they produce and for doing whatever is necessary to verify correctness, but that doesn't mean that software engineers get to pass off responsibility for the things they produce.
- malka 10y ago
  Usually there is a lire in the eula for this purpose. I have no idea how this would plan in front of à judge though
- sammcf 10y ago
  I mean, it's a really easy issue to work around (if occasionally time consuming), most people who use the software are familiar with the issue, it's not enough to make me want to change CAD package given the massive sunk costs one rapidly accumulates using a particular package...I guess there's no real incentive for the software guys to devote time to fixing this when they could be working on new features or whatever. In a better world, I'd love for bug fixing, or even just polishing workflows for existing features, to be as high of a priority for CAD companies as new feature. However, only one of those things is seen as growing the install base...
IkmoIkmo 10y ago
I remember meeting an engineer over easter brunch a few years back and I asked him what his actual day looked like, you know day to day activities. Then someone started making a toast so he quickly summarised 'I make calculations, and then I add a giant margin of error that's totally unnecessary if our model was completely correct, just to make sure' haha.
paulyg 10y ago
What all the non mechanical engineers don't realize is the bend allowance is different for types of material (steel, aluminum), grade or strength (high tensile steel will stretch less than mild carbon steel), and even the tool (mandrill vs brake bender). Software like Solidworks and Inventor come with default values. You need to know to go in and change them based on your circumstances.

by sammcf 10y ago

rtpg 10y ago

I don't get this though... Like if the software calculates bend radii that will break things, isn't that a bug? What's the explanation for that even happening? Is it just that the formula involved means that you can't always get a "right" answer, so the software will be like "well, this is the best we can do, figure out the rest"?

coldtea 10y ago
>Like if the software calculates bend radii that will break things, isn't that a bug? Well, it's not like all companies fixes their bugs or care about them. If, as the parent says, all CAD packages have such issue, what you gonna do?
- chris_wot 10y ago
  The difference in LibreOffice not rendering an EMF correctly and a bug that causes wrong dimensions on a drawing in a CAD tool is that LibreOffice gives you a bad looking document and a CAD program costs you money when the material starts to be cut up and it all had to be thrown away as scrap. Mostly I can fix the LibreOffice bug and make the document render correctly (unless it physically corrupts the document when saving), once physical material gets an error it's often impossible to fix.
chris_wot 10y ago
So what are you gonna do? Move to another CAD package? Take a look at the reasonably large number of CAD packages listed on Wikioedua here: https://en.m.wikipedia.org/wiki/Comparison_of_computer-aided_design_editors https://en.m.wikipedia.org/wiki/Comparison_of_computer-aided... Virtually all of them are proprietary. They are too expensive to move away from, and every one of them has flaws that firms workaround and build there workflows on. Retraining, proprietary file formats and the disruption of migrating an essential part of most businesses is prohibitive. Frankly, IMO the CAD marketplace is an example of how closed source really bites an industry. Funnily enough, most engineers - who would never accept inaccurate tolerances on items like screws or bolts - seem to accept this is the way it is and despite the stories we read here let the situation continue. I'm frankly amazed there aren't more failures in infrastructure from bugs in CAD software. That's more a testament to the fact that engineers are engineers though - when they realise there are problems, they dig in, figure out where the problem is and find a solution, and in the case of CAD software I really think they build or design their way around the problems. Overall, CAD software is, despite all the flaws of being coded source, still pretty amazing and vendors do fix issues eventually. We can do so many more things in society because of it, but as we evolve as a society it becomes apparent there are limitations. Eventually CAD software vendors will be disrupted, but we're a long way from that point right now!
- rtpg 10y ago
  I mean sure, things have bugs, right? I get that. Sometimes autocorrect messes up things I type. It's not really a bug per-se, more just a side-effect of how things like autocorrect work, right? But sometimes Chrome crashes, and that's a bug... The OP seems to say that the bend radius thing is more of the first case than the second, and I'm wondering how true that is. I'm not really sure of the science behind it (maybe it's an impossible constraints problem).
  - chris_wot 10y ago
    Oh, no doubt. The issue is that the tools don't appear to document their limitations and constraints, and there is often no way of knowing until you get bitten. At least with OSS you can look at the code behind the program to work out how it calculates things like the bend radius. I'm no engineer, so I had to look it up (Wikipedia was no help whatsoever on this occasion) but I located this page which gives the formula and an explanation: "Calculating the correct flat pattern layout is crucial to getting a good quality finished part from your press brake. Yet, many CAD and CNC programmers have no idea how to calculate the required values. Years ago, the real experts created cheat sheets and tacked them to the wall. They only taught the new apprentice how to apply the results shown on the cheat sheet, not how to calculate the numbers. Well, now those experts have retired and it's time for a new generation to learn the right way to do the calculate the correct flat pattern layout." http://www.sheetmetalguy.com/bend-allowance.htm http://www.sheetmetalguy.com/bend-allowance.htm Now admittedly the guy does put the responsibility on the operator, but it appears CAD packages can indeed calculate this for you. So long as the right info was input into the program, the formula should be programmed correctly and the operator should be able to have confidence in the results!
SixSigma 10y ago
It is a formula but you need to know the physical characteristics of your metal.
malka 10y ago
Or it's on purpose. This way you can sell training courses and each user is locked on it's cad implementation due to the cost of retraining on another tool

kazinator 10y ago

> "yeah, you're gonna have to take that into account." What? But the software was supposed to take that into account, which is why it it supports bend allowance calculations in it! (I have a tiny bit of experience with this; I designed some sheet metal thing. I somehow clued into bend allowances before doing it, even though it was my first and only one! I figured, you can't just draw the naive shape; the metal has a bend radius. How the hell does that work? [half a day of Googling] Oh. I used a 2D drafting program that had no clue about this, so I manually drew the pattern with the allowance in it already, based on the thickness of the aluminum.)

sparky_z 10y ago
> But the software was supposed to take that into account, which is why it it supports bend allowance calculations in it! Structural engineers aren't allowed to treat software as a black box and trust that it works correctly. They're legally responsible for choosing their tools and verifying that they work correctly.
- kazinator 10y ago
  I.e. the "that" in "that into account" means the tool being possibly wrong. There is no, "sorry the TV blew up in your face; my scope was out of calibration". :)
- chris_wot 10y ago
  Uh, I don't think there is a better example of closed source than CAD packages.
  - afuchs 10y ago
    I don't think he was talking about verifying the software itself. Even if the software was open source, would it have been formally verified to make sure it has zero bugs? It is better to check the calculations by hand to make sure that no critical errors make it into the final design.
    - 10y ago

cm2187 10y ago

I don't know how expensive is the stuff you were making but if it was just about cutting metal it can't be that expensive. Isn't testing samples the 101 of engineering or have we gone 100% trusting CAD now? It looks to me that there are so many things that can go wrong in real life like deformations because of temperature, different calibrations on different models of machine, different composition of the material used, etc...

sammcf 10y ago
Metal isn't that expensive I guess but still, several hundred dollars and a turnaround time increase of a week+ minimum isn't always on the cards for every project. Sure, if I had unlimited time and budget, I'd physically prototype and field test everything. Client demands tend to take precedence. Part of being a good engineer is balancing those concerns, learning which parts of the design and manufacturing process can be trusted implicitly and which assumptions must be rigorously tested - suffice to say I've been a lot more stringent about sheet metal design in the years since.

PhasmaFelis 10y ago

> I reported the issue to my reseller and the response was essentially "yeah, you're gonna have to take that into account." > I guess it's not really a bug per se, just hidden behaviour that you have to learn about and take into consideration. I'm not that kind of engineer, so I may be missing something, but it sounds like the reseller snowed you into blaming yourself for their faulty software. I get what others have said, that a structural engineer is responsible for the things they produce and for doing whatever is necessary to verify correctness, but that doesn't mean that software engineers get to pass off responsibility for the things they produce.

malka 10y ago
Usually there is a lire in the eula for this purpose. I have no idea how this would plan in front of à judge though
sammcf 10y ago
I mean, it's a really easy issue to work around (if occasionally time consuming), most people who use the software are familiar with the issue, it's not enough to make me want to change CAD package given the massive sunk costs one rapidly accumulates using a particular package...I guess there's no real incentive for the software guys to devote time to fixing this when they could be working on new features or whatever. In a better world, I'd love for bug fixing, or even just polishing workflows for existing features, to be as high of a priority for CAD companies as new feature. However, only one of those things is seen as growing the install base...

IkmoIkmo 10y ago

I remember meeting an engineer over easter brunch a few years back and I asked him what his actual day looked like, you know day to day activities. Then someone started making a toast so he quickly summarised 'I make calculations, and then I add a giant margin of error that's totally unnecessary if our model was completely correct, just to make sure' haha.

paulyg 10y ago

What all the non mechanical engineers don't realize is the bend allowance is different for types of material (steel, aluminum), grade or strength (high tensile steel will stretch less than mild carbon steel), and even the tool (mandrill vs brake bender). Software like Solidworks and Inventor come with default values. You need to know to go in and change them based on your circumstances.