I need a crash course in industrial design
September 17, 2018 7:40 AM Subscribe
I have a nice home shop and I know how to use it. My problem is that when I'm designing things, I feel like I'm always reinventing the wheel, and usually doing a worse job. Googling stuff is always really hard because I don't know the right term for what I'm looking for.
It's a little hard to explain exactly what I need. But for example recently I have needed to design
* bearing blocks
* a tilting table mechanism that is spring loaded to stay level
* a table that can be raised/lowered with a crank
I guess basically I need to know some design principles and recipes for mechanical systems. Gears, pulleys, belts, bearings, and so forth
I know that these problems were basically solved more than a 100 years ago.
Are there good online resources for this kind of thing? Books? Videos?
It's a little hard to explain exactly what I need. But for example recently I have needed to design
* bearing blocks
* a tilting table mechanism that is spring loaded to stay level
* a table that can be raised/lowered with a crank
I guess basically I need to know some design principles and recipes for mechanical systems. Gears, pulleys, belts, bearings, and so forth
I know that these problems were basically solved more than a 100 years ago.
Are there good online resources for this kind of thing? Books? Videos?
FWIW you're looking for mechanical design/engineering info. Industrial design is something else entirely.
posted by jon1270 at 9:43 AM on September 17, 2018 [4 favorites]
posted by jon1270 at 9:43 AM on September 17, 2018 [4 favorites]
Response by poster: OK, fair enough, this is exhibit A in "I don't know what to google for"
posted by RustyBrooks at 9:47 AM on September 17, 2018
posted by RustyBrooks at 9:47 AM on September 17, 2018
As a kid I scoured 9780000913159 - The Way Things Work: an Illustrated Encyclopedia of Technology for insight. Its about finished products so the tricks you're after are buried in the illustrations. There is also a Vol 2 but it is more system level.
Misumi is a great source to buy hardware like you are looking to make... if you have CAD you can download their models and modify to suit.
posted by tinker at 9:53 AM on September 17, 2018 [3 favorites]
Misumi is a great source to buy hardware like you are looking to make... if you have CAD you can download their models and modify to suit.
posted by tinker at 9:53 AM on September 17, 2018 [3 favorites]
Youtube has some great resources of builders that design. Ox Tool co has some great ones showing them building their designs. Check here. If you are looking for very specific advice on building a specific thing, the machinist forums can be useful. Practical Machinist and Hobby Machinist are both pretty good. There are most likely more specific forums too. Get and learn to use a free 2D CAD program like Draftsight. Maybe Google Sketchup as well, so you can move things around in 2D/3D without having to have physical access to them. AvE videos are good for learning how commercial items come together, but that guy's shtick makes me ill so maybe turn the sound off or something.
Rothbart's Mechanical Design Handbook has the dope on things like gears, springs, transmissions. I have a copy and I feel it was worth the 8$ I spent on a used copy. It sounds like you don't need the numerical nitty-gritty anyway.
posted by Dmenet at 10:42 AM on September 17, 2018
Rothbart's Mechanical Design Handbook has the dope on things like gears, springs, transmissions. I have a copy and I feel it was worth the 8$ I spent on a used copy. It sounds like you don't need the numerical nitty-gritty anyway.
posted by Dmenet at 10:42 AM on September 17, 2018
*Homer_Simpson_gargle*
tinker, that book is SO GOOD. Another one that I read endlessly is "Moving Heavy Things" by Jan Adkins. His books are lovely, but also so full of instructions if you keep your eyes open.
This weekend I was trying to come up with a way to make some things, and I realized that the real solution was very simple designs -- almost Shaker/Amish in their simplicity. For example: I needed a half dozen things to sit on a table and hold up a letter sized sheet. I had a bunch of thoughts about clips and wire-bending and routing and whatnot, but in the end I drilled two holes into each board and inserted a length of dowel in each hole -- which can be removed easily for storage in a box until next time! So when I have down time I read books and browse web sites that show how things get put together, which helps me have those "simple machines" at front of mind when I am Solving.
posted by wenestvedt at 12:16 PM on September 17, 2018
tinker, that book is SO GOOD. Another one that I read endlessly is "Moving Heavy Things" by Jan Adkins. His books are lovely, but also so full of instructions if you keep your eyes open.
This weekend I was trying to come up with a way to make some things, and I realized that the real solution was very simple designs -- almost Shaker/Amish in their simplicity. For example: I needed a half dozen things to sit on a table and hold up a letter sized sheet. I had a bunch of thoughts about clips and wire-bending and routing and whatnot, but in the end I drilled two holes into each board and inserted a length of dowel in each hole -- which can be removed easily for storage in a box until next time! So when I have down time I read books and browse web sites that show how things get put together, which helps me have those "simple machines" at front of mind when I am Solving.
posted by wenestvedt at 12:16 PM on September 17, 2018
core77, especially the process tab is somewhere I go for inspiration to avoid reinvention.
Also I've developed a friendship with a manufacturing engineer, someone who's been making makings for 50 years and is now moving into CADCAM etc, this is well worth doing and you make friends too. The more complex my designs get the more I realise I can save a lot of time this way, and simplify my designs.
posted by unearthed at 12:37 PM on September 17, 2018
Also I've developed a friendship with a manufacturing engineer, someone who's been making makings for 50 years and is now moving into CADCAM etc, this is well worth doing and you make friends too. The more complex my designs get the more I realise I can save a lot of time this way, and simplify my designs.
posted by unearthed at 12:37 PM on September 17, 2018
Response by poster: So here's an example from today.
I am making a prototype of something. The part I'm working on is a horizontal surface that has a hinge down the center, so it can tilt left and right. The hinge was yesterday's project and what made me ask this question but I eventually figured out something that worked.
So it tilts back and forth but I want it to be level, and only tilt if someone pushes on it. If you stop putting weight on it, I want it to go back to level.
I can think of a lot of ways to do this. They all suffer in one way or another. I'd like to see some options and the pros and cons of them. Most of the options I can think of involve spring force in one way or another.
Like, you could attach springs on the left and right to the immobile bottom, so if you push one side of the table down the other side pulls. I think you'd probably want the springs to have no tension when it's horizontal so that the table doesn't rest lopsided due to one spring being stronger.
Another option is torsion springs but I'm not sure how easy that would be to mount, and same thing, they'd need to be under no load when the table is level. I would have to redo my hinge to accomadate those but it's an option.
Then today I thought that I could use something that is stiff but pliable, and mount it vertically, attached to both the tilting top and the immobile bottom. When you tilt the table, it would bend the pliable material but when you let go it would want to go back to rest. I have no idea what material to use or how well it would work or how much I could adjust the tension.
I know how to *make* stuff given specification, and I know how to do CAD and all that. I have a reasonably complete metal working shop, a very complete wood shop, a router-based CNC, a CNC mill, a 3d printer and a laser cutter. I can fabricate almost everything at home, I just need to know how to achieve my goals.
posted by RustyBrooks at 6:08 PM on September 17, 2018
I am making a prototype of something. The part I'm working on is a horizontal surface that has a hinge down the center, so it can tilt left and right. The hinge was yesterday's project and what made me ask this question but I eventually figured out something that worked.
So it tilts back and forth but I want it to be level, and only tilt if someone pushes on it. If you stop putting weight on it, I want it to go back to level.
I can think of a lot of ways to do this. They all suffer in one way or another. I'd like to see some options and the pros and cons of them. Most of the options I can think of involve spring force in one way or another.
Like, you could attach springs on the left and right to the immobile bottom, so if you push one side of the table down the other side pulls. I think you'd probably want the springs to have no tension when it's horizontal so that the table doesn't rest lopsided due to one spring being stronger.
Another option is torsion springs but I'm not sure how easy that would be to mount, and same thing, they'd need to be under no load when the table is level. I would have to redo my hinge to accomadate those but it's an option.
Then today I thought that I could use something that is stiff but pliable, and mount it vertically, attached to both the tilting top and the immobile bottom. When you tilt the table, it would bend the pliable material but when you let go it would want to go back to rest. I have no idea what material to use or how well it would work or how much I could adjust the tension.
I know how to *make* stuff given specification, and I know how to do CAD and all that. I have a reasonably complete metal working shop, a very complete wood shop, a router-based CNC, a CNC mill, a 3d printer and a laser cutter. I can fabricate almost everything at home, I just need to know how to achieve my goals.
posted by RustyBrooks at 6:08 PM on September 17, 2018
I'm no master builder, but am an enthusiastic learner/prototyper.
This sounds like a fun problem! All of those sound like potential solutions. Have you made prototypes or tried some of them out? I'm a little confused because I would start making a simple/crappy version of what you're talking about with cardboard or scraps of wood, and see how it works out.
The biggest realization I've had when making physical things is that: prototypes are really important -- that carefully planning out how to make something all at once is rarely helpful, and "just starting" and expecting that things won't start to be resolved until version 5 or so.
From that perspective, CAD/CAM and digital fabrication is actually not very helpful to start out with, because it presumes that you already know what you want to make. I find that working with power/hand tools crucial to make a version that actually works, then I'll sometimes make it again.
I have no idea what material to use or how well it would work or how much I could adjust the tension.
What I would do:
- Make a prototype of the hinged surface so that it tilts left to right.
- Try a piece of thin wood as the pliable material
- See if it tilts left to right and returns back.
- If it doesn't, try to understand why. Is it not plastic enough? Not returning to its original shape? Too rigid?
- Try to think of something that might work based on what I've noticed, or something that solves the problems that I've noticed. (A teeter totter toy? A playground spring toy?)
- Isolate the part that works (a heavy counterweight that eventually returns the surface to flat? a rigid spring mounted vertically?)
- Make a prototype that incorporates that
- See what happens, etc.
--
I think you'll also enjoy Matthias Wandel greatly - he's a tinkerer/inventor, and won't just show the successful version of his projects, but will show where things went wrong, etc. Some fun examples of his process: Maple Syrup Boil-o-matic, Baby rocking machine, Wood geared baby soothing machine, Slinky machine
posted by suedehead at 6:32 PM on September 17, 2018 [1 favorite]
This sounds like a fun problem! All of those sound like potential solutions. Have you made prototypes or tried some of them out? I'm a little confused because I would start making a simple/crappy version of what you're talking about with cardboard or scraps of wood, and see how it works out.
The biggest realization I've had when making physical things is that: prototypes are really important -- that carefully planning out how to make something all at once is rarely helpful, and "just starting" and expecting that things won't start to be resolved until version 5 or so.
From that perspective, CAD/CAM and digital fabrication is actually not very helpful to start out with, because it presumes that you already know what you want to make. I find that working with power/hand tools crucial to make a version that actually works, then I'll sometimes make it again.
I have no idea what material to use or how well it would work or how much I could adjust the tension.
What I would do:
- Make a prototype of the hinged surface so that it tilts left to right.
- Try a piece of thin wood as the pliable material
- See if it tilts left to right and returns back.
- If it doesn't, try to understand why. Is it not plastic enough? Not returning to its original shape? Too rigid?
- Try to think of something that might work based on what I've noticed, or something that solves the problems that I've noticed. (A teeter totter toy? A playground spring toy?)
- Isolate the part that works (a heavy counterweight that eventually returns the surface to flat? a rigid spring mounted vertically?)
- Make a prototype that incorporates that
- See what happens, etc.
--
I think you'll also enjoy Matthias Wandel greatly - he's a tinkerer/inventor, and won't just show the successful version of his projects, but will show where things went wrong, etc. Some fun examples of his process: Maple Syrup Boil-o-matic, Baby rocking machine, Wood geared baby soothing machine, Slinky machine
posted by suedehead at 6:32 PM on September 17, 2018 [1 favorite]
Response by poster: I've been following Matthias since before there was youtube - back in the woodgears.ca days. I found him because, like him, I was trying to find an efficient way to make gears at home.
I have a prototype, and I'm going to try some of the methods above. My prototype is half-inch plywood because I have a literal ton of scrap to work with. (well, half inch plywood and some 3d printed parts, I like 3d printing because although it takes forever I don't have to *be there*, I can start a print before I leave for work or before I go to bed and it'll be done when I get back. The parts it makes are very precise)
I have ordered a bunch of material to make springs with. I like my last method because it's easy to experiment with. Making springs is frustrating unless you know what you want already and commercial springs are available but again, pretty much only if you know exactly what you want.
posted by RustyBrooks at 6:49 PM on September 17, 2018
I have a prototype, and I'm going to try some of the methods above. My prototype is half-inch plywood because I have a literal ton of scrap to work with. (well, half inch plywood and some 3d printed parts, I like 3d printing because although it takes forever I don't have to *be there*, I can start a print before I leave for work or before I go to bed and it'll be done when I get back. The parts it makes are very precise)
I have ordered a bunch of material to make springs with. I like my last method because it's easy to experiment with. Making springs is frustrating unless you know what you want already and commercial springs are available but again, pretty much only if you know exactly what you want.
posted by RustyBrooks at 6:49 PM on September 17, 2018
Response by poster: Whatever I do needs to withstand many thousands of cycles. It will get used daily, probably dozens to hundreds of times.
posted by RustyBrooks at 6:50 PM on September 17, 2018
posted by RustyBrooks at 6:50 PM on September 17, 2018
Would gas springs be appropriate for the application? Would probably be less "sproingy" than metal springs (take less time to settle back to equilibrium if that's something you're concerned with) and also maybe be tunable to a certain extent by bleeding out some of the gas.
(My thought process to get to that solution: "Huh, that sounds a lot like a simplified version a self-leveling optical table. Don't those use compressed air? But hauling in a compressor might be a bit too much work, I think you can get self-contained units called.. Air springs? (no, those are in cars) Pneumatic springs? (not quite) Oh! Gas springs! *google image search to verify that that's what I'm thinking of* Yep!". Like 90% of design concepts start with "hmm, Problem X sounds kind of like Problem Y, which I dealt with using Solution Z...")
posted by btfreek at 8:15 PM on September 17, 2018
(My thought process to get to that solution: "Huh, that sounds a lot like a simplified version a self-leveling optical table. Don't those use compressed air? But hauling in a compressor might be a bit too much work, I think you can get self-contained units called.. Air springs? (no, those are in cars) Pneumatic springs? (not quite) Oh! Gas springs! *google image search to verify that that's what I'm thinking of* Yep!". Like 90% of design concepts start with "hmm, Problem X sounds kind of like Problem Y, which I dealt with using Solution Z...")
posted by btfreek at 8:15 PM on September 17, 2018
I am not a mechanical engineer, but my work involves a lot of similar problem solving. On the tilting table I think you're wrong about springs needing to be under zero tension when it's level. If the springs are completely relaxed at the neutral position then they'll be exerting very little force as the table approaches that neutral position, and those tiny spring forces won't be enough to overcome friction or slight imbalances in the mechanism or in the weight of the table surface so the table will always be off-kilter. I think you'll need to have the springs tensioned (or compressed) at neutral position, and provide some fine adjustment mechanism to allow you to balance them against each other. A detent that the table clicks into at the neutral position might make the arrangement less fussy so there'd be less need to adjust those springs if they wear or fatigue or change behavior slightly due to temperature or humidity or whatever.
posted by jon1270 at 4:28 AM on September 18, 2018
posted by jon1270 at 4:28 AM on September 18, 2018
Possibly useful: The Gates Heavy Duty V-Belt Drive Design Manual.
Also, go easier on yourself. Expecting to design mechanisms with moving parts that need to last for thousands of cycles without some phase of experimentation, error and correction is not reasonable. Big companies with large budgets and departments full of engineers mess these things up all the time. If they value their reputations then they do a lot of testing before product release, because they know they can't anticipate and plan around every sort of failure, and STILL it's easy to find 2-star Amazon reviews for products that disappoint in one way or another. Machines are hard.
posted by jon1270 at 5:09 AM on September 18, 2018
Also, go easier on yourself. Expecting to design mechanisms with moving parts that need to last for thousands of cycles without some phase of experimentation, error and correction is not reasonable. Big companies with large budgets and departments full of engineers mess these things up all the time. If they value their reputations then they do a lot of testing before product release, because they know they can't anticipate and plan around every sort of failure, and STILL it's easy to find 2-star Amazon reviews for products that disappoint in one way or another. Machines are hard.
posted by jon1270 at 5:09 AM on September 18, 2018
You are going to have some problems here because only you know what your design goals are - does it have to be flat within 1 degree and be able to withstand 1000 kg of force, or is 15 degrees fine under 1 kg? And so on an so forth. A lot of that comes from seeing inside what other people have made, past experience, at minimum a qualitative understanding of physics, and some idea of a design-->prototype iteration loop. Assigning even ballpark numbers helps me understand what materials/processes are immediately not going to work and narrows down the field. Maybe post your prototype to a hobbyist forum to get feedback.
posted by Dmenet at 11:15 AM on September 18, 2018
posted by Dmenet at 11:15 AM on September 18, 2018
And really, unless I have very specialized experience, I find the best way to design something to last thousands of cycles is to see how the prototype breaks under use and do a design iteration. Unless it's going to the moon.
posted by Dmenet at 11:16 AM on September 18, 2018
posted by Dmenet at 11:16 AM on September 18, 2018
Response by poster: I think gas springs are overkill but it gave me an idea. I was going through my assortment of springs and came across a toilet paper holder thing. You know what I mean? 2 tubes of plastic with a spring inside. I could make one of these with an eyelet at each end. As long as the springs are not compressed at rest, it should stay level and return there.
I'll probably make the springs myself so I can make them stiffer by using more loops/tighter spacing or getting higher gauge wire.
And yeah - I wasn't so much looking for a design answer here as I am trying to figure out *how* I'd find the answer. I can think of at least 5 ways to do it, I don't know which of them is reasonable, there are probably 5 ways I've never thought of. Any particular picked method I'd have to go through several iterations of. I don't have the time or inclination to make 50 prototypes.
I was sort of hoping there might be video series (dvds, online like udemy or something, youtube, whatever). I learned to operate a mill and metalworking lathe mostly from DVDs and youtube videos. I learned wood working (and a lot of the principles of furniture design) from books, dvds, tv shows and finally youtube. I'm less looking for specific answers than I am looking for the knowledge to be able to answer them myself. My example above was an example of the kind of thing I'd like to know how to do.
posted by RustyBrooks at 5:59 PM on September 18, 2018
I'll probably make the springs myself so I can make them stiffer by using more loops/tighter spacing or getting higher gauge wire.
And yeah - I wasn't so much looking for a design answer here as I am trying to figure out *how* I'd find the answer. I can think of at least 5 ways to do it, I don't know which of them is reasonable, there are probably 5 ways I've never thought of. Any particular picked method I'd have to go through several iterations of. I don't have the time or inclination to make 50 prototypes.
I was sort of hoping there might be video series (dvds, online like udemy or something, youtube, whatever). I learned to operate a mill and metalworking lathe mostly from DVDs and youtube videos. I learned wood working (and a lot of the principles of furniture design) from books, dvds, tv shows and finally youtube. I'm less looking for specific answers than I am looking for the knowledge to be able to answer them myself. My example above was an example of the kind of thing I'd like to know how to do.
posted by RustyBrooks at 5:59 PM on September 18, 2018
As long as the springs are not compressed at rest
At the risk of derailing I'd like to agree with jon1270 above about how you would probably want some sort of tension at rest, unless you are confident about your ability to make your springs/supports/etc exactly symmetric - if everything is under zero tension at rest then your hinged table will naturally flop towards the side with the slightly weaker spring. (If you apply a preloading force this will still happen, but the idea is that your applied force would be proportionally much greater, thus reducing the overall effect of the asymmetry.) Of course only you know what is/isn't acceptable in your design.
posted by btfreek at 10:28 PM on September 18, 2018
At the risk of derailing I'd like to agree with jon1270 above about how you would probably want some sort of tension at rest, unless you are confident about your ability to make your springs/supports/etc exactly symmetric - if everything is under zero tension at rest then your hinged table will naturally flop towards the side with the slightly weaker spring. (If you apply a preloading force this will still happen, but the idea is that your applied force would be proportionally much greater, thus reducing the overall effect of the asymmetry.) Of course only you know what is/isn't acceptable in your design.
posted by btfreek at 10:28 PM on September 18, 2018
Response by poster: If they are both perfectly slack - almost but not engaged - I think it'll be OK. A little preload on each is also probably OK. I just don't want one pushing a lot more than the others.
posted by RustyBrooks at 8:03 AM on September 19, 2018
posted by RustyBrooks at 8:03 AM on September 19, 2018
This thread is closed to new comments.
507 Mechanical Movements
The latter is not really instructional, but I find that the best thing is to just expose yourself to lots of different ways of solving these problems - kind of like how good writers read voraciously, you know? To that end, really being observant when out in the world and taking things apart to see how they work also help a lot. As an engineering intern I used to flip through the Mcmaster-Carr catalogue during downtime and think about how all these bits and bobs fit together (and then try to spot them out in the wild).
posted by btfreek at 8:58 AM on September 17, 2018 [5 favorites]