CAD World vs. Real World - Engineering Process
HTML-код
- Опубликовано: 30 мар 2022
- CAD World vs Real World
•••
“Couldn’t you just simulate it in CAD” is a question I get asked quite often when I show a video of an error that I’ve made in CAD. The answer is like 75% yes?
•••
Things don’t always translate directly from CAD to real life. Sometimes you need to rely on experience gained from past successes and failures, and a willingness to once again fail, in order to move on to the next step in the design.
•••
#engineering #engineeringprocess #designprocess #learnthroughplay #3dmodeling #cadmodeling #solidworks #fusion360 
Наука
I see an engineering job at bmw in your future. 'well I got it in there, but good luck for the next guy taking it out.'
😂😂
You mean every new car?
@@R_o_b_e_r_t Sometimes, but BMW and other expensive luxury brands seem to be more notorious for making it extra hard rather than somewhat annoying
@@RaisinBarXZ550 I mean there aren't any German luxury car brands anymore
They're just regular cars now
@@falcy2889 like the brand's themselves have become normalized, but compared to a Corolla the prices are higher and the luxury is too, and they still make proper luxury and expensive models
You can always cut up the model to just the area where you need to test to save on materials and time :)
Very true!
Though with a small model like this you can also bet on it working, or at least well enough that you can dremel it to fit, and use it as a prototype.
I do this allot, especially when it comes to screw holes. Just print 3 layers even to see if they line up and are the correct size
Agreed. For fit tests, I usually only print the walls that are absolutely necessary for the test, and print at a much higher feedrate.
For example, a fit check for a motor shaft would just be a small cylinder, rather than the entire rotor.
@@user2C47 true, but you have to be careful. Several times I have printed a section quickly only to discover that when printed with normal print settings fits are tighter than on the tests.
Ok so you have 6 screw-pillars printed on the lower part of the case which means you have 6 holes on the top part.
You could print 1 of the 2 screw-pillars on the top part and the screw hole on the bottom part.
This
Exactly what I was thinking the whole time.
Or... move the f-ing post over a couple mm!
There is a bunch of solutions for this dunno why he thinks there is no solution
@@IsaacAlcocer he never said there was no solution? I think he's well aware that he could change the design but made a teaching moment out of it
In my opinion this is design flaw, because force shouldn't be necessary to fit component inside housing and also what about repair possiblity, then you need to pull faulty component from the housing to replace it?
I don’t even know wtf he made but its something that has to be a tight fit so it needs force to fit it in
@@beybladesinmyheels1267 No, the motor screws in, there’s no need for it to be forced into place. This is just a bad design.
@@beybladesinmyheels1267 wrong.
Nothing wrong with a little force if it get the job done 😉
@@Engineezy Yea but when you have a faculty motor, its "welp looks like this $200 piec is junked bc of the motor" or " well i can replace the motor but its hard as hell and its on my own time, not company time so I'll just use the boss's card to get a new motor". Either or both lead to anew item, not a new part.
Also a great example of when 3D printing is a very useful engineering tool for prototyping even if the final product won’t be made of pva or resin
So true!
Perfect prior planning prevents piss poor performance
I tried to come up with an alliteration of equal epic status but all I came up with were worthless words without walue…
@@Engineezy Hey guy I'm in the same car with you. I got it from a circuit dude talking about 555 timer setups. I think the 7 P's of success have been passed down from technician to technician over the years and the originator forever remains at large.
Nice designs as well bro. 💪
I would agree with you for most things, but…
_Primarily pursuing primordial perfection prevents progress when prototyping_
I just made that up and am a little proud of myself, lol.
When prototyping, it is critical to understand what is a value-added activity. Spending tons of time trying to make the first iteration perfect slows things down unnecessarily. Because a lot of these don't go beyond a one-off though, they often stop at the iteration that is sufficient for the purpose.
You could make gaps in the screw flanges for a better fit
Well in the real world you always have a little flex in the materials
I wish freecad had an assembly process solver where you could tell it were everything should be and it would tell you if it was assemblable, and how much bending would have to happen where.
Make the screw pillar 1mm thinner on that side…
Seimens NX has a module that can determine the best way to get something in or out. I dont have access to it, but the demonstration showed the best way to get a radiator out of a car. It said through the bottom on that model.
That sounds like a very useful tool!
That's why I do a lot of test fitting prints. In this case, I'd put a negative zone in my slicer so only the clearance test gets printed, that way instead of taking 6 hours and a bunch of filament to test your fit, it might only take half an hour and use a fraction of it.
great situation to bring attention to and one that ive run into personally having not considered actually inserting the part into the printed enclosure
"if it doesn't fit, make it fit".... 👌
👌👌
Damn dude that was extremely satisfying to see it pop in place, great work!
Thanks Dylan!
You keep doing shit like that you’ll be on the mechanics most wanted list.
This is where pausing a print to insert other components will be really cool in manufacturing.
So true!
You can also put it in there while its printing, wont be able to remove if not enough room though
Very true!
Nice Job! Bit for that: you can calculate the max width of the Part with a^2 + b^2 = c^2. It is Tricky but doable.
So… you made the whole print for a test instead of just the part that could fail? Why didn’t you make a test print to find out if it fits?
Low key I knew it would fit already- RUclips loves a little drama
A quick adjustment to the thickness on that one side.
Yup you can put my idea in the bin with all the others.
You might consider creating a rough draft print profile to print that thing way faster as a tester
Couldn't you have make a small divot in the screw pillars for the pointed bit of the motor to slide it in better?
Great idea!
I cut my part down to just what I’m unsure about, print it, go make a snack for a half hour, and then come back to a finished test model where I can check if it fits
When I'm in a situation like this, what I do is cut the part in CAD and print it to make sure everything fits the way it should, so I don't print the whole part.
Definitely a better strategy than mine 😂
I think where the disconnect comes in is we forget that in the real world you can use force that causes certain parts to flex and bend. Meaning you could fit the motor into the piece. The concept doesn't really translate or is often forgotten in CAD
Well put!
CAD kinematics/assy mannequin type tools could tell you if it fits.
Test print. You print a U shape of the moter and a cup shape of the corner it goes in. Less material, faster print, and you can measure the amount of tolerance available.
Mount the servo to an access plate that bolts to the side. Several easy edits to change the viability of the part.
Did he just pronounce CAD “COD”
"Ossisted"
Canadian accent! 😂
I would have just taken the part along the curved edges off further and made them part of the lid or whatever, so the opening is totally square without "hidden space" underneath
Cad doesn't account for the slight flex in the motor mounts. If it was made of any more rigid of a material, some point of contact would likely bend or break
Possibly! Luckily we got a little flex in this one
Sometimes the only way to know is to be willing to fail. Nice!
👊👊👊
Could you have just paused the print before it started covering the area the motor was going in, placed the motor, and started it back up? Maybe double sided tape to keep it in place if you’re worried about it moving during printing.
Definitely an option!
@@Engineezy Unless you plan on removing the part in the future. I typically only use this method for inserting nuts or when I need to fill a part with ballast.
You could always just Dremel that little bit of the outer housing of the switch off that way it would actually be removable without using a bunch of force
Pythagoras also made a way of working that out 😆
Did u know you can use collision test in SolidWorks?
A small notch in the one pillar, rounded if necessary to avoid stress risers, could have been included, to overcome forcing the servo in.
That would definitely help!
This is basically the reason to 3D Print a Prototyp.
Exactly!
You could just pause the print and put the servo in before the top roof part is finished.
Apparently printing in 0.6 nozzle is the best way now, especially for speed, I do it with 0.3 layer height and the prints are much faster.
Gonna be honest, you could tell it could fit from the start
Cool work! Would you consider altering the shape of the flange in the corner just a little bit, or maybe moving it a mm further from the motor?
Idk the application, but if 3d printing isnt the final production process for the housing i would be worried about production tolerances of the housing.
I take it you're not familiar with his channel. He creates mechanical sculptures. It's not likely that there will ever be more than a few iterations of his current piece.
@@danielstewart3507 ah, youre right i am not familiar with this channel. If its low production run then that tight tolerance is ok, even a good thing in a way:)
@Daniel Stewart got it right, design for production is a whole other animal!
If this was pla, I would just use a soldering iron to reshape the flanges a tiny bit to get the motor in
Could've gotten away with a single screw on both sides in the center
You could have used a collision detection to know whether the part is in contact or not
Maybe you could try to stop the print in a layer that let's you incert the motor and keep printing over it afterwards?
That would definitely work!
That would be so easy to work around
Design challenge. Every issue you restart with a new design that uses the foresite gained from the "failed run" to start from scratch. I don't know. Seems like it would bring out interesting design choices.
Big F hammer fixes everything.
Too true 😂
You can definitely fit it easier with bigger openings
If this wasn't printed on a resin printer or something, just pause the print before the upper brim is printed, lay in your part, then let the print finish. I do that all the time to embed parts for projects that dont need to be taken apart.
Great idea!
In some situations, you may need to print the outer piece AROUND the inner one (in that case the motor). Did you try this with any design?
Thats always a fun way to go, definitely have tried it in the past!
I don't 3d print but could you stop the print near the end install the motor and continue to print your piece?
I sometimes forget to check whether or not something can be assembled before making it and honestly it's really interesting to move the part around in cad to check.
Unfortunately there are no plugins to remind us to check 😅
I mean there is a way, it’s called accurately measuring and accounting for the expanded print volume, but alright
You can do this on real CAD packages, like Catia. Big corporations won’t spend millions on tooling just to find out things don’t work.
Im working as a Assembling specialist at R&Ds. And yeah usally its the other case.
Next time I recommend watching the print and right when the layer is taller then the servo you can put the servo in and let the 3d printer print around it
Very true!
@@Engineezy lol I can tell you just started reading your comments like 4 mins ago
Hey I have a question regarding your job. I am currently in the process of deciding what I want to study in university. So I therefore want to ask you for inspiration: what do you work as?
I work as an engineer/artist in no particular order! Good luck
You could also pause the print
The key to success is failure and learning from your mistakes.
Wise words!
Mm plastic deformation as an installation method this is a very reliable and long term solution
As long as we stay in the elastic region of the tensile curve we should be good to go!
You could stop the print halfway, place the motor and continue the print.
Definitely true!
Could you cut out the desired location and print a smaller portion to test fit?
Yeah that would def be the smartest way to go!
Pfft, yea that is the reason why stuff blows, now we know it 😁
😂😂
Just calculate the size tolerances.
How did you get such a clean looking print?!
Those prusas sheeeeeesh
You can add it to the piece while it's printing
Cool idea!
As a Civil Engineer, we solve this with BIM
If you dont need to swap it you could insert it part way through the print.
Very true!
Wait a second… does that mean engineering includes designing AND testing?! What a surprise 🤓😉
Definitely a real surprise for some haha
Like a gloooove!
🙌🙌
This is a problem with engineering practice mentality. If it doesn't fit in the model, you are guaranteed screwing you build/repair team once they have to deal with it in the field. I am a design engineer, and got to work a lot with our mfg and field service teams. If you absolutely have to do this, give it to both parties first, and make sure you aren't going to have problems. As an example with this case, try getting that motor back out when it is attached to something else.
I mean you could've guessed it'd fit if it was near 1 to 1 fit cause alot of things can stretch a little
All the engineering that goes into cars and sometimes ya just gotta whack it with a hammer to get it started
Haha so true!
“The caad world” lol
Canadian accent eh
If you wanted to test if it'd fit, you could have just printed either that one section of the model, or printed a simpler geometry and faster to print mock version.
Very true!
@@Engineezy Although on second thought, it'd probably barely cut down the print time, so you might as well just do the full print anyway haha
How are you going to to put the screw that’s gonna hold the bracket on the side where you can’t get to the holes on the SERVO
There’s a solution for that! Stay tuned for the larger video
@@Engineezy 👍🏼
it was obvious that you could put it in by rotating it in like a clip.
I can't believe this is the best way. I thought i did this like this cos i was a noob but... Is that it? It cant be
Might not be the best way, but it is a way!
You only need to print the area that you're test fitting.
Very true
What os the cad did you use?
Did u model the servo your self or is there a place to get that file?
Modeled it myself with a pair of calipers!
I wish I wasnt talked outta going to college for engineering
You can mess around with free software & resources and maybe check out if your library has 3d printing and give things a go! If you find that you really like it, then by all means dive deeper. Don't feel limited by college
Definitely doesn’t mean you can play around! Exactly what BrownMInc said!
@@BrownMInc I didn’t know that. But also engineering stuff is very complex to learn. I have massive add and terrible short term memory. I literally forget what I’m saying sometimes while saying a sentence to someone lol. Idk how much math is involved but I am a near literal genius with math my whole life so I got that going for me. And I’m 24 so college for 4+ years is outta the question and learning software would take even longer by myself
@@Engineezy you mean “can’t” play? And yeah I just saw what he typed. Thank you good sirs
@@Mike-we3rb Give yourself some credit man, my memory is shit too and I'm mediocre-average at math at best. I assumed from the first comment that getting a degree wouldn't be feasible so I meant as a hobby / something to explore.
The way I look at it is that it's a freeform way of learning / doing. I'm currently in my 3rd year of university and I can reliably say that the degree gives us the tools to analyze things deeper, if we need to. There's no class or magic moment that'll teach us how to come up with creative ideas though. That part is on us and you definitely don't need math nor a degree.
The way I approach it is I get the idea I wanna explore, try some things or research similar projects and only if I wanna refine to a high degree would I then look up whatever relevant theory. "how to x" and "what causes y" type of questions I'd search.
All's this to say I understand if life has you occupied in other areas, trust me, but don't think the doors of tinkering and making is closed to you. There are many amazing projects done by ppl who don't know a lick of theory. I assume you like cars and I'd challenge you to question how many automotive RUclipsrs are actual engineers/take a scientific approach to building some of their projects vs just loving cars to the point they just keep at it until they build something cool
For how long have you been 3d printing?
7 years 😮
Do as I do if you are unsure about if it will fit, just print the sections you want to test, instead of the whole print.
Like a glove
😂😂
Would it have been better to print just that corner so you could test fit it?
Idk much about 3d printing or CADing custom shapes, so idk if doing that would be worth the time and effort
Honestly yes, but I knew the thing was gonna fit haha just trying to illustrate an insight
@@Engineezy Ah ok. That makes sense. Thanks!
Why you dont move the screwholes a little bit down
The part it’s going in to needs the holes to be where they are if possible!
Just move the flanges?
Rhino+Grasshopper+Galapagos evolutionary solver should give you an answer in CAD. (SolidWorks is not a cure-all)
No le heches la culpa al CAD. El programa hace lo q el usuario le indica. Esto es un problema de diseño, no del CAD!!
You could have paused the print right before the bridging
What CAD software do you use?
I use Solidworks!
As a repair tech, you are the infamous engineer we hate lol
Glad it fit. Fact is though there are deft smarter wears of engineering that. Including 2 pieces. If it’s just a housing or project box for something … it’s structural integrity would have been fine. Engineering allows for pro typing obviously but don’t need to game or waste plastic lol.
How’s you get that motor model, grabcad?
Reverse modeled it using calipers!
I repair appliances. This happens way too often
I can imagine 😂
If only there was a better way to view collision