Lots of iterations of design

Someone asked a question regarding why the Collaborative is trying to work with lugs when it seems pretty clear that off-the-shelf options aren’t working for this project.  Or at least, preexisting lugs didn’t work for the Seven Cycles Shop Bike we just built.

Answering this question quickly becomes a circular discussion.  In the context of the Collaborative, an example might look something like this.  In fact, In order to explain one of the reasons why we’re committed to using lugs, and simultaneously frustrated by using lugs, I’ll provide a real world example.

But first, remember that one of the elements of the Collab’s mission is to work with lugs.  I’ve provided some of the compelling reasons we’re dedicated to lugs for this project—and I’ll be providing more reasons in the coming weeks.

On To the Example

I’ll oversimplify this example, a bit, for the sake of brevity but this situation actually occurred—it was a bit more convoluted than I’m presenting here.

The three Collab designers responsible for the Shop Bike design got together and laid out the bike.  One of the design parameters was a bike that anyone at Seven could ride—it is a Shop Bike after all.  This is no simple feat but certainly a good requirement for any community based Shop Bike.

Within about ten minutes of evaluating this design, it became clear that frame geometry and performance compromises would be needed, if we were going to stick with the goal of using three stock lugs out of the four required to build a frame.  Remember, we only wanted to make one lug from scratch—in order to expedite the build process.

So, the first directive defenestrated—it’s my favorite word—was the idea of a bike that fit everyone, and was safe for everyone to ride.  The second directive defenestrated was the idea of a tubeset that would ensure decent and handling with 30 lbs. of beer, donuts, hardware store supplies, babies, etc, in the front basket.

The Collab was about to begin production on that watered-down design when we realized that we were suddenly building a bike that didn’t come close to our design directive.  How’d that happen? So, we backed up and thought about how we ended up with a bike that no one really wanted.  The reason we ended up here was simply because we let preexisting standards, or options, dictate the design.  I know that this kind of tradeoff happens every day; maybe it doesn’t seem like a big deal.  Regardless, I was still amazed that we fell into this mode so easily.  In part because we design bikes everyday in which we have very few design parameter limitations—all custom, no lugs, lots of component options.  So, I kinda thought that we’d naturally push back against whatever lug limitations arose.  But, that didn’t happen naturally; we had to force ourselves to stop and reexamine the situation.

It’s so easy to lose sight of the learning road when there’s a very clearly defined, and attractive, well traveled road.  Many of us grow up thinking design rules exist for good reasons, without every asking the question why.

So, that’s one of the reasons we’re working with lugs.  To push ourselves and learn stuff that we wouldn’t otherwise learn.  It would be easy to bail on lugs and design a bike like we always do—but then we wouldn’t be learning as much.  Where’s the fun in that?  For better or worse, my Collaborators know that I’m not interested in working within pre-existing methods.

Essentially, we had to decide between being true to the design intent, or making our lives easier.  Guess what, in my world view, design intent is primary; making our lives easier is about 20^th on the list.  So, we go where the design directive leads us.  In this case, it definitely led us to a learning spot.  And that spot didn’t include compromising the number of employees that could ride the bike.  Unfortunately, it did mean spending a lot more hours making lugs than we’d hoped.

We’re making the lugs fit the design rather than the design fit the lugs.

Not to get too, artsy, but I can’t help but quote Robert Frost:

“Two roads diverged in a wood, and I —

I took the one less traveled by,

And that has made all the difference.” – Robert Frost

Before anyone thinksI’m getting heavy or righteous, I’ll be the first person to guarantee that the Collab has—and I have—I’m sure, taken some easy and fast roads along the way.  But, lugs aren’t yet one of them.  And, the Collab isn’t trying to take the fastest road, the simplest road, or the easy road.  The entire point of the project is to learn.  Part of the learning is doing all we can to take the road that’s true to our directive.  For Frost that was the one less traveled—and that often seems to be the learning road, in my skewed world.

Which leads to one of the Collab Fundamentals, so stay tuned.

Share on Facebook

Evaluating a few lug methods

Early in the Seven Cycles Collaborative we discussed the ways in which we could go about sourcing or making lugs.  Here are a few of the ways to make bicycle frame lugs:

• Investment casting:  This is the most common method of making high quality lugs.  This is the oldest form of metal forming on the planet.  It’s a pretty amazing way of making a formed part.
• Stamped from sheet:  This is the lease expensive way to make lugs in large quantities.
• Bulge forming:  This has only been used a few times, relatively speaking, compared to other methods.  Basically, a steel tube is put into a lugged shaped mold and hydrostatic pressure is used to expand and shape the tube into a lugged form.
• Welding or brazing tubes together:  This is the method that Seven Cycles is using for the Collaborative Shop Bike.  Photos to follow.  This is also the method that Seven uses on our titanium and carbon mix frames.
• Machined:  This is not something I’m aware of for steel frames but some other lugged construction frames start with a molded part and then final machine the angles and details.  I include this on the list because it’s one of the ways that we are evaluating making lugs for the Collaborative.  However, it would be more appropriate for a production setting, not so much for the Colab project.
• Combination:  This is very common, particularly among artisan frame builder.  Combo lugs take an investment casting and the add material, or more commonly, remove material to end up with a different look and characteristic than a raw casting.

We discussed a few other ways to make lugs but we’re keeping those ideas under wraps for the moment.

Lots of options; all with challenges.  More later.

Share on Facebook

Bracing for the storm

An addendum to the “Lug Options” post:

Here’s one of my favorite quotes from the Seven Cycles Collaborative so far.  It’s from Bryan Hollingsworth. He’s managing the lug design portion of the Colab.  So, yesterday when he was he was sharing the lug search results for our Shop Bike test, here’s what he said about our available options:

“It’s a perfect storm of nothing.” – Bryan Hollingsworth

If you read the previous post—Lug Options—this statement sums it up perfectly.  I just had to share it.

Share on Facebook

Lugs, lugs everywhere and not a one to drink

So, we’re collecting all the parts required to build our first test-of-concept frame:  the Seven Cycles Shop Bike I mentioned last week.  Our intention was to source the four lugs and maybe build one from scratch; making a lug from scratch was to be part of our conceptual test.  Well, it turns out that we can’t find a single lug that will work on this shop bike—not even one.

What?

We’ve looked at too many lug sources—through suppliers and direct from manufacturers, domestic and overseas, road and mountain, high quality and cheap—and cannot find a single lug that will work on any of the four joints that would typically get lugs.  “What the what?” This bike design is completely within Seven’s norms.  In fact, we tried to design the geometry so that we could use some standard lugs.  Still, no dice.

I’m sad to say that this is one of the more enlightening moments of the project for me—and for many others in the Collaborative.  Here we are trying to design a bike—a fairly average bike design, by Seven’s standards—and we can’t find a single lug that will work.

What Are The Four Lugs?

For those that don’t know what I’m talking about, the four common lugs are:

1. Head tube to down tube
2. Head tube to top tube
3. Seat tube to top tube
4. Bottom bracket to seat tube and down tube

[If I had time I’d include a diagram.  Maybe later.]

Why?

Why can’t we find lugs?  The basic reasons include:

• We want to have this bike employ a “compact” design with 12 degrees of top tube slope.  Yes, this is a bit more than average—Seven’s road bike average being about 8.5 degrees.  But, over the years we’ve built thousands of bikes with 12 degrees or more of slope.  And, apparently, lugged steel bikes must have an average of about 0 degrees of slope.
• We want to use a 1.25” diameter top tube.  Yes, this is on the unusual side for a road bike.  But, not for a mountain bike—and there are some mountain bike lugs out in the world.  Two reasons we’re going with a 1.25” top tube:  1) This is a shop bike with a big rack on the front—front end torsion can be a challenge, and a stiffer top tube can help.  2) Because we’re butting in house, we can play with diameters, tube walls, and tube butts more so we want to experiment with this tube.
• We want to use a 1.25” diameter seat tube.  Again, unusual but not crazy.  We have our reasons:  explaining later.
• Read between the lines.

So, in the end, even when we tried to compromise the design fairly extensively, we still couldn’t find lugs—actually, we found one that would sort of work.  Therefore, we decided to design our ideal since we’d have to make lugs from scratch anyway.  And, maybe a fillet joint or two—that’s a whole other conversation.

Later.

Share on Facebook

That about cover it.  We’re not experts yet but we’re well on the way to making every mistake possible.

This is one of the Fundamentals of the Collaborative:  Fail Fast.  This is also one of the mantras I’m known to mumble at Seven Cycles.  Interestingly, in the Colab, we have to fail fast if we’re going to complete this project.

We’re struggling with getting the fatigue testing methodology standardized.  We’re still testing the testing procedure, yipes.  Since there is no industry standard, we’re setting our own and although we often guesstimate and calculate these types of process correctly on the first try, we didn’t in this case.

I think we’ve just about got it nailed but close is not close enough.  More on that later.

Share on Facebook

This is the first test fixture. Version 2.0 is already under way

I can’t keep up with all that’s happening.

On Tuesday—a lifetime ago—we started our fatigue testing program.  In fact, the tester is running as I type.

The tester had been running 24-hours per day since Tuesday.

Testing Ad Infinitum

We’re trying to set the benchmark failure point at about 75,000 cycles.  I’ll post more about this later.  The reason I mention it here is that 75,000 cycles takes about 21 hours to reach.  So, at best, we’ll be running one test per day.  And, we have about 100 tests we want to run for the Collaborative.  So, we’re looking at about a 20-week testing schedule—100 tests divided over a 5-day work week.  20-weeks we don’t have for the Colab.  So, we’re now in negotiations regarding which tests take priority and which tests will have to wait for Colab 2.0—or some other project.

Hiccups

We’ve had a few hiccups in getting the equipment going; regardless, we’re running now and we’ve already had some interesting results—not so much for lug testing but for patience testing.  Here are some of the challenges we had before I thought we’d be testing last week.

• Our testing equipment also has other testing commitments so we’re trying to schedule four heavy duty testing projects all at once
• Our testing computer died so we had to get that going again
• Skip—Seven’s tool maker, head machinist, and just about everything else—had some other projects and commitments to deal with before wrapping up the test fixture
• We had an internal miscommunication about joint geometry that required a fixture change—this turned into a great Colab learning moment

We’re going to start posting results soon and catalogue them here for posterity.  And, we’re making some changes to the fixture that will minimize more possible variables.

Stay tuned.

Share on Facebook

A Metaphorical Story

I’m not a story teller and I usually don’t like metaphors.  So, here’s a metaphorical story.  And it actually relates to the Collaborative.

A few days ago I watched a fire truck backing into a fire station.  The vehicle was as big a fire truck as I’ve seen; I believe it’s called a hook-and-ladder truck.  It’s about the length of an 18-wheeler truck—big enough that the truck had two steerers:  a driver in the front and a steering person for that back wheels, too.  I’ve seen this type of fire truck maneuver before and it’s interesting to watch a vehicle of that size move so effortlessly.

What was memorable about the recent parking instance was a juxtaposition.  On one hand, the driver was very confident and handled this oversized vehicle like he’d been driving school buses for many years; he moved the vehicle with certainty and no hesitation.  On the other hand, he didn’t seem to be working well with his steering partner at the back of the truck.  It took them a number of tries to get the rear end of the truck into the fire house.  It definitely appeared that the driver was not used to a rear wheel steering; it looked like he was trying to back up a school bus rather than a very agile dual steering vehicle; he kept steering away from the direction for which the rear steerer was trying to correct.  So, the dance was a series of overcorrections and apparent confusions about what the truck was doing and why.

It struck me that this scene was a perfect metaphor for how I sometimes see people work in new situations, or when they don’t work collaboratively very often.  In those types of situations a person might have a tendency to:

• Treat a dual-steering fire truck like a school bus
• Overcompensate, counteract, and essentially fight against what the seasoned steering partner is doing
• Forget that being an expert bus driver does not a fire truck expert make
• Be blind to knowing that skill without teamwork can be detrimental to getting into the firehouse
• Work really hard, but not work together.  In fact, the harder the driving, the harder the counteraction—at that rate they’ll never get in the firehouse
• Communicate through overreaction—over steering—rather than picking up the walkie-talkie and saying “WTF?”

At Seven I think sometimes I can be an adequate bus driver but not always good fire truck driver, or steerer—and this may hold true for some others at Seven sometimes.  It could be because we don’t have many fires to fight at Seven; right.

Share on Facebook

Tubing sitting in the comfy chair

I mentioned this earlier but it bears more props:  last night we performed our first in-house steel tube butting tests for main triangles.  Mike S. is managing the tube aspect of the project and he just ran with the tubular ball, worked with Jon Henig—our senior production machinist and frame builder to run some tests.  Actually, only two tests.

What is tube butting?

In short, tube butting is a process which results in a tube that has various wall thickness.  In the bike world, the most common butted tube is on that is thick on both ends and thin wall in the middle sections.  The thick sections, after the tube is coped to become part of a frame, are usually about ¼ to 1/6 of the total length of the tube.

Only Two Tests?

The first test didn’t work well.  We tried to butt in stages.  Big mistake in retrospect.  But, it told us most of what we wanted to know in order to have a successful second test.

The second test yielded a production quality tube.  In fact, we might be able to use the part in our shop bike.  The butting process was slower than we’d like to see but for this project, it worked perfectly as a proof-of-concept.

Secrecy?

Yep.  I’m not going to share how we butt tubes.  It goes against the Open Workbook of the Collaborative.  Regardless, it is one of the few proprietary elements of the Collaborative—primarily because it’s a set of technologies we developed for our titanium and carbon bikes.

Share on Facebook

Some of Seven's shop bikes and commuter bikes

In this morning’s Collaborative meeting at Seven Cycles Mike Salvatore suggested that we build a bike to test out a few of the conceptual ideas and assumptions that we’ve been discussing so far.

The outcome of that conversation is that we’ve agreed, as a mini project—headed by Matt O’Keefe—to build a Shop Bike that will be applying some of the work we’ve done far.  Someone set a deadline for completion of the frame, fork, and stem as Friday, December 11; six working days away.  The word crazy comes to mind, not just because it’s a new product but more because we’re doing a lot of test-of-concept work on this bike.

What to accomplish in the next six days

• Build the first complete lugged frame under the Seven badge
• Butt all the tubing in-house:  eight tubes.  This will be the first time we will have done all the tubing in house.  And we have a lot of tooling to figure out
• Build a steel fork that incorporates some Collaborative aspects
• Make at least one of the lugs from scratch.  We’ll probably end up making more than one, using various construction ideas.  More work.
• Build a steel stem that incorporates some Collaborative aspects
• Machine the frame and fork dropouts

This amount of work is a bit ridiculous—ridiculous to get all this done in six days.  I’m pretty confident we won’t be able to get everything done, but we’ve set our triage list and we’ll work through as much as we can by the end of the day on Friday.

Shop Bike

This test-of-concept bike will be used for what we call a “shop bike”.  We currently have three shop bikes and they are all sort of sad:  Scrounged parts, part’s we’ve repaired too many times, patches on patches on patches.  While we do love to reuse, at times, the shop bikes get into the “refuse” category for some.

Test-of-Concept

I use the term “test-of-concept” rather than “proof-of-concept” because we’re not even at the point where I can say that we’re going to prove anything.  We really are just testing some ideas.  And whoever is brave enough to ride the bike will be proving the test results.

This will be the most expensive and over engineered shop bike we will ever build.

The frame and components go to our painter on Friday and will be ready for assembly and its maiden donut run by Wednesday of the following week.

Share on Facebook

In the Collaborative meeting yesterday we reviewed the high quality brazing recommended steel frame tubing options.  That’s a mouthful.  Because it’s a long set of limiting parameters, it also limits the legitimate tubing options that we can apply for this project.

Let’s start with sources:  In the bike industry, four serious sources exist for high-end steel bicycle tubing:  Columbus, Dedacciai, Reynolds, and True Temper.  I would think that, with four sources, there’d be plenty of options to mix and match.  Well, not true—at least from Seven Cycles’ perspective.

Now, which of the four companies offer tubesets that they specifically recommend for brazing—as opposed to TIG welding?  Only two of the four.*  That’s okay, in reality, I’d bet that a lot of lugged frame builders use tubing not specifically designed for lugs.

Okay, even if we include all four suppliers’ lug specific, or closest high-end equivalent, the options are still slim and disappointing.

For research purposes we decided to focus on the frame’s down tube—probably the single most important tube in the frame.  We don’t have all day to mess with each tube at the moment.  You might expect that suppliers would provide more options for the most important tube, than offered with any other tube in a tubeset.  And you’d be correct.  Well, the results are still quite uninspiring.

Now I’ll cut to the chase.  Two of the four suppliers offer only one option in a 1.25” tube diameter.  The other two suppliers offer two tube wall options in 1.25” diameter.  Doing that math, a frame builder might think that we’d have six tubes from which to choose.  The bad news is that there are actually only three tube wall options from which to choose:  0.04 mm, 0.45 mm, and 0.5 mm.  The other three tubes are simply overlaps to these first three options.  In the end, these three options are not what we’re used to at Seven.  And the difference between 0.4 and 0.5 is only 20% different; that’s not a lot in our book.

By the way, for the tech geeks, yes, each supplier offers slightly different tube butt lengths but this has very little real world impact on the ride off the completed bike.

All this led us to the conclusion that we’re going to butt all the tubes in-house.  More on that soon.

*I’m working from memory at the moment so the number might by three.  Maybe.

Share on Facebook