Filtering by Tag: modeling

STEP files

Added on October 28, 2013 by Spencer Wright.

From Step Tools, Inc.'s excellent overview of STEP purpose, applications and development (emphasis mine):

The ultimate goal is for STEP to cover the entire life cycle, from conceptual design to final disposal, for all kinds of products. However, it will be a number of years before this goal is reached. The most tangible advantage of STEP to users today is the ability to exchange design data as solid models and assemblies of solid models. Other data exchange standards, such as the newer versions of IGES, also support the exchange of solid models, but less well.

But later, in the "Future of STEP" section:

The real issue that stops faster STEP deployment is the commitment of those with the resources necessary to define the standards. The government does not like to pick solutions for industry, and industry does not like to fund the development of solutions that can also be used by their competitors. Consequently, much work only gets funded in situations of clear and desperate need such as when the high cost of manufacturing is causing excessive job losses.

The Internet and the World Wide Web broke through this cycle when "killer" applications made the benefits of the new infrastructure clear and compelling for all users. AP-203 made STEP useful by allowing solid models to be exchanged between design systems. AP-238 will make STEP compelling for some users by allowing them to machine parts more efficiently. However, like the early Internet there will be alternatives that are considered more reliable by other users. The killer application that makes STEP ubiquitous has yet to be identified.

There's a lot of good info on STEP buried in obscure corners of the web. I hope to be summarizing some of my research on the topic soon.

A good afternoon's work

Added on October 18, 2013 by Spencer Wright.

This. Was fun.

Happy Friday, everyone.

ABK1004 Barrel Offset Topper Assy from SpencerWright on Sketchfab.

Rack Ends: Current design

Added on October 17, 2013 by Spencer Wright.

I'll be writing more about how I'm going about procuring my stainless steel rack ends later, but I wanted to give an idea of the sorts of design files one needs to supply machine shops with in order to get quotes. A dimensioned (and toleranced!) PDF is always key:

N.B., my drawing template is pretty lame right now. Not having standard tolerances listed is a total no-no, but most manufacturers assume +/- .005" on any distance and +/- 1 degree angular - so I should be okay. I really just need to spend a morning setting up a new template, but that's fairly low on my priority list right now. Anyway... do as I say, not as I do.

I also usually include a STP file, but as a rule that's *only* for reference. I suppose that if you had a particularly strong relationship with a supplier and were sourcing a part with *only* standard tolerances, skipping the PDF would be okay... but that's a rare scenario.

Aaaand, thanks to my friends on Twitter, you can see the part in 3D here:

BK1008 Rack End from SpencerWright on Sketchfab.

More modeling

Added on September 17, 2013 by Spencer Wright.

For reasons I won't go into right now, I spent some time today modeling a Thomson seatpost. It's a fun little project, and one that involves some weird intersecting surfaces. I definitely got some good use out of my radius gage set, and even plunked down to get a new one that goes up to 1" (mine, a Brown & Sharpe set that I got from an old machinist, is 1/32"-1/2").

While I'm excited to check out the new 3D scanners soon (I'm hoping this weekend's Maker Faire will have a couple), manually measuring and modeling something like this is pretty fun. It's interesting to think about the way the geometries work. A Thomson seatpost consists of three main parts, which take a number of manufacturing techniques:

The post is made from an oblong extrusion. Its OD looks like a circle with a lobe on the front and back, and its ID is an ellipse with its short axis oriented front-back. After being extruded, it's turned down to its finished diameter and milled to accept the saddle cradle and hardware. I'm guessing that both of these operations happen in a cool multiaxis turning center, but they could just as well be done in separate setups.
The lower cradle is forged. I believe it's also machined, though only on one side and probably just one or two passes.
The top cradle (not shown - I still need to model it) is also forged. I believe the bolt slots are then machined, and of course the Thomson logo is laser etched after anodization.

The hardware is made by a combination of forging, cold thread forming, turning, and machining. It's also all hardened.

Thomson is a badass manufacturing company. I'm excited to finish this portion of my modeling, which amounts to describing the parts they engineered, and move onto the really cool part - screwing with all of it.

Modeling for SLA

Added on September 17, 2013 by Spencer Wright.

I spent a little while today optimizing my dummy headset for 3D printing. In subtractive manufacturing, cost can often be estimated by calculating the difference between the mass of the raw material and the mass of the finished object. The more material you need to remove, the more fabrication time and resources you'll consume producing the part, and hence the more expensive (generally) it will be.

The cost structure of 3D printing is totally different. With additive techniques, cost is a largely a function of the mass of the finished object (envelope size also has an effect in production settings, but it's less critical). The cost of a part comes down to how long it takes to make, and production time is limited by the amount of material the machine can spit out in a period of time.

As I noted the other day, my dummy headset is a bit more expensive than I'd like it to be. On the upside, though, I can remove material in a bunch of places! I took the revolved part in Inventor and made a series of revolved cuts on the ID of the part. I left some ribs along the ID to keep the perimeter somewhat intact, and left the areas right around the set screw holes thick.

Shown before I mirrored the cuts onto the top side of the part.

I was able to shave a *lot* of mass off the parts - more than 45% on the lower part - which means significant cost savings.

I still need to make sure I'm on the right track on clearances in a few locations, but cutting the extra material out should make these parts much more feasible. Keep in mind, my total development cost at this point is probably 2.5 hours of labor and $40 in parts. Were I trying to get this product to market quickly (and who knows, I may try doing so) I could be live in both Shapeways' - and my own - webstore within two weeks.

Even though I've just been playing with it for a half hour, it's pretty cool seeing my model live on the site.

Another 3D Printed thing: Dummy Chris King Headset

Added on September 16, 2013 by Spencer Wright.

I can't justify the cost on this, but with some small modifications I could print a lot of them before I hit the price of the molds I'd need to injection mold them.

Modeling. That's an M3 set screw on the top "cup;" there's another one on the backside. They'd be used to keep the dummy headset from moving on the steer tube. Once tight, the frame and fork can be treated as one unit.

Okay, the color's a little off. I'm working on it ;)

Most framebuilders, and a lot of bike shops, will have a frameset (that's frame and fork, for the uninitiated) around the shop for some time before the headset is ready to be installed. Optimally, they're able to be kept together and protected - both from things around the shop and from each other - and the natural solution is to temporarily install the fork in the head tube. For a variety of reasons, though, you don't always want to install a headset just yet, and in those cases it's useful to have a dummy that approximates the size and shape of the headset that you're eventually going to use.

When I was building bikes, I made a batch of dummy headsets out of aluminum on my lathe. It was a fun project, but it took a while and the finished thing didn't look at all like the headsets (usually Chris King) that I was installing on the bikes when they were done. Moreover, a lot of small time builders either don't have access to a lathe or don't have the time/energy/gumption to build dummy headsets themselves.

So I spent an hour or two and modeled this one. It's a damn close copy to a King 1-1/8" NoThreadSet, but SLA printed. I included two holes in the top "cup" that I'll tap out and install set screws in. When the dummy is installed on the frameset, the set screws can be tightened down to keep the whole thing together. Because the dummy is plastic, it won't mar the frame, and because it's dimensionally accurate, it could be used to mock up the steering column for use in rack building, component setup, etc.

I need to make a few small changes to reduce printing mass, but in the meantime the design files for these parts are all in a GitHub repo. If I can find a way to get the cost down a bit, I'm hoping to put them up for sale for other folks to use; drop me a line if you're interested.

Work-ish: 3D printed dropout protectors

Added on September 15, 2013 by Spencer Wright.

I think Nick Pinkston has it absolutely right: "3D printing is great, but it's only a small part of the solution." The current hardware revolution is about the workflow from development to manufacturing to distribution. Sure, I'm sure I'll have more and more 3D printed objects in my life in the next 5-10 years. But the effects of product customization (which will be significant) and kanban/just-in-time manufacturing (which I believe will be huge) will far outweigh the designer's ability to neglect draft angles when designing plastic parts. In the near future, I expect we'll be buying more stuff that hasn't been built yet than we have since the industrial revolution. In the next decade, I expect Amazon (or whomever) to be literally building the parts required to fulfill my order the night after I place the order. 3D printing will be a big part of this process, but so will distributed manufacturing and rapid delivery systems. And innovative ways of finding new products (and, on the flipside, innovative ways of finding new customers) will totally change the game.

I tend to recoil at most of the crap that's made with the current generation of FDM machines, but I've spent some time recently trying to think of objects in my life that I would accept being shat out of a MakerBot. A few traits I was looking for:

Needs to be made out of plastic
Needs to be disposable
Needs to have a rough surface quality (low layer resolution)
Needs to be something that's hard to find in a brick-and-mortar store
Relatively low part mass, to reduce print time & cost
Low dimensional accuracy to accommodate all sorts & conditions of printers
Bonus points if I wouldn't want to buy it from Amazon due to package quantity, lead time, etc.

I'm sure there are better use cases, but one thing I came up with was dropout spacers. When shipping a bike, you usually remove the front wheel and install a dropout spacer into the fork. The spacer protects the dropouts from impact from below and also protects the fork from impacts from the side. Most consumers don't keep dropout spacers around, and wouldn't necessarily think to go to a bike shop to pick some up (most shops give them away) when they're shipping their bike. When the bike is unboxed on the other end, the spacers usually go straight to the trash, and surface finish is totally inconsequential. Plus, the spacer itself isn't very massive, and the dimensional accuracy required is low.

I spent an hour or two modeling, and got Shapeways to print me the result for $13. It's a bit more than I would want to pay for a piece of plastic, but the FDM version would be basically free. The finished version is shown below. I rather like it, and think that things like it will be printed - not in the home, probably, but by brick-and-mortar third party services like Kinkos, or web shopping platforms like Amazon and Shapeways - as a matter of course in the future.

If anyone is interested in printing one of their own, I'd encourage them to grab the model on the Thingiverse. I also published the model and a bunch of other stuff in a GitHub repository.

Inventor & Shapeways. SLA parts from Shapeways tend to have a bunch of powder slag in any crevises; this one was packed full.

Back to front: Injection molded rear dropout spacer; injection molded front dropout spacer; my SLA front dropout spacer.

cycles progress

Added on July 29, 2013 by Spencer Wright.

i spent much of the past few days modeling a conical burr grinder. the project requires a number of complex sweep features, which are difficult both to measure and to model, and with which i have limited experience. the process was interesting.

the burr is ceramic (more about this in the future) and has two sets of tapered helical features. the large helices seem to funnel coffee beans down into the assembly and begin to break them down, and the smaller features do most of the work of producing the intermediate and fine grind steps.

the large features seemed - at first - relatively straightforward to design. the profile is easily measurable from the top of the part, and establishing a path (albeit one i would later realize was inaccurate) wasn't too bad. here, i'm applying fillets to all the sharp edges left behind after the sweeps were cut into the base body.

in parts like this, one tends to model a single feature and then apply a pattern on the part using the initial feature as an input. here i'm patterning the small corrugations on the perimeter of the part. i'll do this process a number of times; once you apply the pattern, you'll notice things about the feature that weren't apparent when there was only one occurrence visible. i'll roll back the pattern and modify the underlying feature, and then reapply the pattern again and check out the result.

the real result is that i shouldn't be doing this whole process on a 2008 iMac >> Boot Camp >> Windows 7 >> Inventor: a dedicated windows box would be *way* more efficient. here's to hoping that AutoDesk gets real about some WebGL modeling software?

the only way i know to model these features is by using Sweeps, which require a profile and a path as input. i would prefer to be creating a 3d helical sketch and using that as the path, but i can't for the life of me figure out how to use Inventor's helix creation toolbox on complex shapes like this. instead i've been creating planes that approximate a portion of a helical shape, and then drawing arcs on those planes. it's a pretty hokey setup, but for my purposes it's more or less adequate.

the red lines here are the remnants of one of these sketches, with the part's cut edges projected onto it. on the left, you can see the arc that i'm using to define the small corrugations' path.

honestly, most of this here is overkill. i'll probably end up getting one of these burrs 3D printed, but it's mostly academic at that point: the hard part of my project isn't the burr design, but the software that supports its function.

either way, it was a fun part to model - and a bit above my head, i'll admit. it's still far from perfect - the modeled version isn't nearly as bell-shaped as the original - but i'll let it go for now. after all, i've got the other half of the burr to model now, and it's possible i'll learn something there that'll be applicable to remodeling this half.

modeling

Added on May 28, 2013 by Spencer Wright.

recently.

my notebooks get the most minute-to-minute use when i'm modeling.  they're one of my primary tools during the process - to sketch ideas, add up figures, note relationships, etc. — my notebooks get the most minute-to-minute use when i'm modeling. they're one of my primary tools during the process - to sketch ideas, add up figures, note relationships, etc.

the Public Radio is coming along.  the pot is a little tricky to fit in the enclosure, but i think we can make it work, and the speaker and battery will be fine regardless.  i have half a mind to switch over to a Weck jar, as the lid is s… — the Public Radio is coming along. the pot is a little tricky to fit in the enclosure, but i think we can make it work, and the speaker and battery will be fine regardless. i have half a mind to switch over to a Weck jar, as the lid is slightly larger, but it lacks the brand name recognition that Ball jars are so good for.

modeling, lately.

Added on May 23, 2013 by Spencer Wright.

workday

Added on May 15, 2013 by Spencer Wright.

good workday today with zach. lotta progress, lotta dogs. my surface creation skills are pushing up. progress below, shown chronologically.

sometimes, dogs just follow the team around.

sometimes, dogs are like "what's up??!??!!"

fucked with the ring a bunch, and added the battery, but a bunch of the work happened on the "ball" logo, which has been a real challenge to nail down. it still isn't perfect, and i needed to mess with the shape of the jar to get the logo where it is now, but i did get rid of a bunch of the workarounds that i implemented yesterday, and the overall model is pretty damn good.