Thru Axle Unicrown Fork, Vise Pedestal

This fork is on its way to Alaska after getting a coat of black powder to fit up onto a bike I did up a couple years back.  It is a long time in the works, mainly because for the longest time I didn't have a dummy thru axle, but then because the machine tooling obsession took over and I really didn't want to hand miter another unicrown fork.  It has the original Paragon thru axle dropouts that I ordered when they first came available, so that should put a bit of a timeline to the project to those of you who want to really pick this apart.  So now its done and its about the most badass rigid fork I can imagine.  Raked tapered blades, externally butted steerer, tabs for a truss rack, clearance for 29+or-"er.

Oh yeah, I welded up a sweet new rocket ship vise pedestal.  The angle of the photograph makes it look small and top heavy.  It isn't.  Its 36" high bringing the jaws up to about 42" from the floor in hopes of getting better posture when polishing brass.  Thanks to the viking for the time on the mig welder to make this happen.  Its super sturdy and I haven't even bolted it to the floor yet.  

Here's a better picture:

Turning 4130

A while back after purchasing that lathe I got the idea of turning some tapered non-tapered steerer tubes that would better match up with the offering of 1 1/4" diameter unicrown blades and give a nice and thick attachment point for the fork "crown."  Prior efforts at achieving this effect were accomplished by brazing a sleeve onto the bottom of the crown to increase the diameter of a steerer tube where the fork blades attach.  Machining this piece takes one of the heat cycles out of the equation and ultimately yields what I believe is a better product as the reduction from 1 1/4" at the bottom to 1 1/8" at the stem clamp is carried inside the head tube by very gradual 0.75 degree taper rather than abruptly at the crown race seat.  This design utilizes the tried and true straight 1 1/8" headsets of any configuration (internal/external) rather than the 1 1/8" to inch and whatever taper cups that more and more bikes are going to, suspension forks or otherwise.  It should be noted that these designs do certainly have their place, particularly in the suspension world, but the reason for doing this is to create the best brazed bike I can, in this case a dedicated rigid.  This piece gives the blade's crown miter a full wrap of the steerer tube while simultaneously beefing up a part of the bike that sees more and more force applied to it with disc brakes and increasingly fatter tires.   Fat tires SUBSTANTIALLY amplify braking forces and chattery rigid forks are discouraging at best.

Starting with some 1.25" x .125 wall 4130 cromoly, I set about designing the steerer tube from the bottom up, giving ample room for blade and crown race seat attachment.  I carried the taper 80mm into the steerer and some trig starting with the O.D. of the crown race seat area and subtracting the stem clamp diameter gave me a change in radius of 1.1mm and a taper angle of around 0.75 degrees.  Click to photos to full screen.  Calculations at this point are more accurate than my tooling can read so I rounded up the math (I'm talking math into minutes and seconds here versus a machinist's protractor that reads 5 minute intervals).

With the design laid out onto the tube I setup the lathe.  Now I have not spent much time with 4130 on the lathe.  My first cuts were looking pretty terrible with lots of galling and I was getting discouraged when the phone rang with a call from my buddy Evan at Idyllic Cycleworks.  Evan owns Moustache number 4 and has since found himself getting increasingly involved in a bicycle oriented machine shop in Fort Collins, CO.  He was able to give me some pointers that started with some new indexible tooling and inserts that allowed for deeper cuts at a higher speed.  4130 is rather counterintuitive when machining.  All the metals have their quirks though as I'm finding out.  With these pointers and a new bit I started making some practice cuts.  Turn it UP!

The roughing cut above already improved upon my initial efforts by leaps and bounds.

Messing around with tip angle for finishing cuts.  Above are samples trying out different feeds rates and RPMs 

The results were immediately better, but presented me with the learning machiner's problem of needing to take a thick cut for a finishing pass to get an accurate diameter of 28.6 mm for the stem clamp.  I went about the process in a more planned manner than I have to date:  Two roughing passes of 0.028", then a finishing pass to take off the measured remainder accounting for tool deflection.

Now with a plan on paper, the next hurdle was the taper transition.  My lathe does not have a taper attachment allowing the turning of a taper under power feed, so the taper would have to be done using the compound slide under manual feed.  This meant a smooth transition from power feed to manual feed would be imperative.  This lathe will stop itself with the power feed on if an obstruction is met, so I made up a carriage stop.  By placing this in front of the carriage where the tool bit met the start of the taper I could manually engage the compound feed without stopping.  Stopping even for a second would produce a groove in the work piece that would be a direct trip to the recycling bin.  There is certainly room for improvement on this method by milling out a taper template for the cross slide to follow under power, but that's another project for another day.  In the meantime I managed to produce three usable steerers and one decorative piece of practice metal, and also some seriously long pieces of blued out phone cord lathe chip.  The blued out coil was a happy cut in chromoly, and you still can't fake a chip.

Thanks for following along in the geek text!

Handlebar Fixturing

I've been on a bit of a handlebar bender lately (hahaha).  A calculated bender though.  I mean I've been calculating my bender.  I mean calibrating my bender.  Wow, worst opening lines ever.  So I took some of that custom drawn True Temper stock and started turning it into some bar samples in order to make repeatable bends on the home made bender which is not machined, so not totally calculable, and more meant for bending fork blades, but works just fine for 7/8" tubing.

This being the machine era has led me to start working up some fixturing for handlebar pieces, particularly bull moose stem pieces.  Handlebars, especially fixed position handlebars, are always a challenge as you are taking a position that (hopefully) works for works the rider, yet that still has some degree of adjustment (bar tilt, stem height, stem extension and angle), and trying to replicate the position permanently.  Probably the biggest challenge in this is the difficulty of measuring all of the varibles as even when isolated they are still very three dimensional.  (Frames, on the other hand are quite two dimensional in design even though the end result is three dimensional).  Different brands of handlebars place the bends at different locations along the bar giving a different extension and hand position between bars.   Its not just 9 degrees bend and 700 mm wide, its where the location of the 9 degrees sits giving the end result of 700 mm.  Placing the bends more inboard toward the centerline will result in a hand position with less extension and a longer piece of raw stock.  Then there's rotation and upsweep and the fact that this will all be altered on bikes with differing head angles.  It all comes back to hand position in whether or not the effort is a success.

I decided to start with the hardest one first--the gray one with the large single bend on the bottom of the picture above.  Single bend bull moose 17/34 degrees sweep with 80mm extension based around an existing setup that was working for the rider on a bike with a 70 degree head angle with no stem angle.  With the bend of the new bar being forward of where the stem clamp of the current setup would be, I was left to choose a point in space to establish 80mm of extension.  For me this is done by measuring in from the ends and connecting with a line to represent a regular handlebar centerline.  I used an existing bar for this measurement and adjusted as I felt necessary.  Now for the hard part, three dimensional mitre angles.  I plot work on a full scale drawing, but getting to this step stumped me for a while, and it made me wish for some tooling that doesn't exist yet, at least to my knowledge.  

Work of this type usually goes better if there is an all encompassing plan laid out to the end, but sometimes that is just not possible and the problems must be met as they arise.  The latter was definitely the case with this project.  I started with making the fixture above out of some scrap solar energy parts without really knowing if it would fully accomplish what I was hoping.  I figured that if nothing else it would allow me to miter the steerer clamp, but was hoping to do all mitering without removing the pieces from the fixture which could create a whole new set of variables.  I made the miter for the clamp and was still puzzled by the bar angle needing to accommodate the 4 degrees of upsweep I had designed.  I left the pieces in the fixture alone for a few days and what finally led me to the answer was the realization that I needed to think of these cuts around the bar and not around the bike.  4 degrees of upsweep at the end translated to a measurable 1.5 degrees looking at the side profile which was then added to the head tube angle to clamp the bar.  The cut angle was matching that of the bar at 17 degrees AND I was able to mirror the cut so as to not change any settings or re-fixture between sides!  Confused yet?

Now onto the next fixturing problem.  Thus far I've fixed these projects for tacking with whatever is laying around that'll pass as a flat surface.  Fortunately I was passed along an old Anvil seat stay mitering fixture with the directive to cut it up and make it work.  Awesome words right there.  The possibilities of perpendicular tee slots in machined metal are pretty endless.  For now I am just using the fixture as a flat surface for clamping as you can see here.  I did make some 4 degree wedges for fixturing the upsweep.  The steerer clamp is held on the fixture I made for the original bull moose project some years back.  Everything is lined up using squares off the flat and is then measured/marked for center.  All told there is much left to be desired in future tooling, but having a few more of these done will benefit that design greatly.  

The finished product atop another manufacturing side project.   I'm really stoked with how the machine era is furthering the process.  Its been a complete re-calibration on my part, but a quite enjoyable one that I feel will continue to produce better results.  All told THIS bar probably took longer than if I had just cut it up with files, but that time investment will pay off in the next one, and the third one is just hours in the bank.  I read an argument recently that machines eliminate the artisan (artistic) element to bicycle fab, but I find little merit in this argument.  I'd venture to say that its an argument made by someone who has not used a machine.  You can drive them at your will, just like the bike in the end.  Sure you can use the machine to make cookie cutter product. Sometimes this is their place and rightfully so.  You can also use them to create plenty of things that you just couldn't with your hack saws and files.  Its all tools for the process.  Sometimes you need tools to make tools.  I will say that I'm glad to have the last six or seven years of doing things the way I was to serve as a background for knowing what I wanted to do in the machine era.  Turns out that isolating learning curves is a good idea too.

The M.M.F.F.H.U

I will now proceed to sum up the most missed spectacle of the Sedona Mountain Bike Festival: the Moustache March Friday Fun Hater Unvitational.....

In the 6th promised hour and after much triumph and tribulation in the field of bonk management, the third attempt at crossing Oak Creek proved to be the divisive move in group dynamics.  The bonus miles, hours, trails and roads that followed this maneuver were left to individual devices.  ALL in attendance survived and were reportedly satisfied with the experience.

***The rest of the event was also top notch***

Attention Sedona:

The Descent of the Fun Haters is Imminent. 

Donatello Does Machines: The Barker PM Mill

Adventures in Craigslist volume 231:  A horizontal mill in Williams, the next town over, and given the fascination and the rarity of affordable heavy (or otherwise) machinery here in AZ, I figured it was worth a trip over to check out.  It followed me home from another old guy's shop, him sighting that it wasn't up to the task of planing metal that he had hoped it would accomplish.  In addition to the mill came 10 or so slotting cutters, a new (to me) drill press and two metal tables, or rather table parts shoddily cobbled together to look like tables.  The price seemed right and without driving upwards of 700 miles or paying freight, this stuff is hard to come by in these parts, so I forked over some bills and loaded up the truck.  That was at some point last summer and upon getting it unloaded and onto a bench to inspect I decided that a full overhaul was in order and there it set until I found the time to mess with it last month.

Now I'm no machinist, more of a machin-er at this point, definitely a tool junkie, but I am not exactly versed in what to look for in a good machine tool purchase.  What I saw that day in Williams looked like it had potential so I brought it home.  A little research found that this is the PMO model Barker horizontal mill, with overarm support for slotting operations.  It has a micrometer feed on the z-axis and lever feeds on the x and y axes.  The action in the feeds was functional with a little slop, but seemed serviceable so I proceeded with a mechanical overhaul of the gibbs and slides.  The 4x12" table appeared to have had a tool or two run into it, not showing true mill grade flatness, but seemed adequate enough to bolt up a vise; and a little bit of galling showed on the y axis slide, but cleaned up with a file to smooth and tight action.  Little bits of Phil Tenacious oil and grease, a bushing here, and a new bolt there and within a few hours this thing was feeling mechanically sound.  Now to make it run.  

So the information listed above was exactly all of the information I could find on this mill, save for the fact that it is still in production relatively unchanged over the last 60 years and that the current purchase comes with a five speed belt drive and low or high gear set.  This feature seemed to be absent on this unit.  The 1750 RPM motor was geared up with a 1x3 pulley system making it suitable to cut plastic.....maybe.  No wonder my man in Willville couldn't do anything with this machine.  To boot, when I took the bracket holding the reversible drum switch apart to get a better look at the wires in the rotting power cord, the plastic drum promptly crumbled in my hand as did the rubber once containing three strands of copper wire.  Add those to the parts list.  

As my reading turned up that these are intended to be setup for single task production (PM=production mill), my main hope for this unit was if nothing else dedicated chainstay mitering.  But looking at the over arm support and the slotting cutters that came with this machine, it seemed that slotting would also be among the tasks performed, as well as main tube mitering.  There really is not much to this thing so it seemed as though changing setups between tasks would not be too terribly time consuming.  

The most productive source of picture information I could find on this mill was eBay.  That said photos were not very lucrative here or anywhere.  I was hoping to get a look at the gearing system, particularly one with the rumored low gear drives.  Nothing surfaced so I went about designing a new belt pulley gear system for this mill.  Working backward from what I perceived to be this mill's future work loads, I came up with a range of usable speeds I thought would be adequate.  It seemed prudent to use the slotted clamp piece dangling on the "rocker spindle," that solid 3/4" bar that balances the mounted motor with the mill head, as the basis a tensioning spindle for an idler pulley on a two belt system.  Now the question would be one of what would fit this relatively small space AND allow me the most usable gear range.  Several evenings of calculations with an online Belt Pulley Calculator (linked) proved invaluable.  

Envisioning a usable gear range somewhere between 100-150 RPMs at the low end and 1000 RPMs at the high, the next factor to consider would be the block width of the pulleys and the load on the spindle and bearings at the mill head.  Looking through some general information on belt pulley dimensions it seemed that a 1/2" belt (type A) would be easy enough to find and a type A pulley would be easy enough to make on the lathe.  The overall width of one pulley is 3/4" so that would give me room for a 3 pulley stack on the 1" diameter head spindle.  Pulley diameters were chosen using the distance between centers feature in the pulley calculator and by matching belt lengths.  Changing the input/output values for RPMs, and working forward starting with the 1750 RPM motor, then backwards from my desired working RPMs allowed me to determine that a fully overlapping transmission (such as one found on say a 12 speed drill press) would not be feasible.  The gear diameters required within such a small space would not allow for this.  With some room to grow back toward the machine casting I determined that a 2 speed reduction system off the motor would drive an idler/tensioner pulley stack, giving a six speed transmission.  There really is a lot going on here spatially, so playing around with the calculator both forward and backward was what made this all possible.  Low gear on the motor was determined to be 2":8" and low gear off the idler was determined to be 2":6" giving an overall reduction of 1:12 and a calculated low spindle RPM of about 145.  From here, belt lengths were held constant to determine a higher set of gears for both systems allowing for equal throw in the tensioner that had yet to be designed.  A leap of good faith here that with these rough paper dimensions in belt pulley calculations, something could later be created to accommodate belt tensioning in the remaining space.

With pulley diameters determined, it was time to shop for material.  Aluminum seemed like the best choice for shaping on the lathe for speed sake, and cutoffs were substantially cheaper as they got shorter, so it seemed smart to bolt together the pulley stacks.  Doing so would also reduce time on the lathe.  I filled a 35 gallon trash can with the stringers of four pieces of aluminum.

Below are the finished (mostly) parts.  From left to right: head spindle drive pulley set, tensioner/idler pulley stack with pressed in bearings, tensioner pulley spindle, motor drive pulley, and tensioner mounting bracket, coffee, tooth brush.  Hope my mostly made up names make sense for those still following.

As assembly commenced, some tweaking and re-working of certain pieces was needed, mainly the tensioner mounting bracket and spindle underwent several modifications to allow for clearance in the now tight space between the main casting and the motor.  The motor also had to be moved back an inch or so and was accomplished with a plate with offset holes.  One of the belts was miscalculated an inch too short.  Otherwise the whole process was void of defeatist notions and tool throwing, always a plus when fabricating on a "leap of faith."  Below is how the setup came out under tension and operational as of last night.  

Below is a closer look at the back side of the tensioning setup.  The square plate is bolted to the spindle with the two button head screws at a tolerance that allows the spindle to slip freely in the slot of the gray mounting bracket.  Tension is provided by the screw on top passing through the mounting bracket and threaded in the spindle.  What can't be seen here is the face plate (think four bolt bike stem) that now holds the mounting bracket to the rocker spindle as the original pinch bolt just took up too much room.  Eventually I would like to add a quick release mechanism to the tensioner screw.

Had to make some custom hole saw arbors for the 1/2" 3c collet.  I'm not sure yet if these are beefy enough to handling cuts with hole saws up to two inches, but it gave me a reason to learn to thread on the lathe.  Next move may be to try and make some arbors that bypass the collet and tighten up directly with the draw bar.

Finally the mitering setup at first go.  The vise pictured here leaves a little to be desired as clearances are tight when turned to angle and the jaws are fairly short.  Successful miters were made, but some tweaking in the set up of the z-axis along with some centering jaws in the vise would be very helpful.

So 2014 was the year of machine acquisition, making 2015 the year of machine production.  In the year that promises hover boards and flying cars my resolution is to program myself as human robot from 100 years ago.  Thanks for reading!

The Green Plastic Bin of the Dells

This project started off as many do, with a ride.  Matt has been riding mountain bikes for as long as I have, maybe longer, and mostly here in AZ.  When the call came for a new bike, Matt had gotten to steal a ride on a friend's custom 29er.  He was thrilled and raved about it.  I figured it was time to take a trip down to Prescott to sample some of their geology.  Prescott riding is great, some of the best in Arizona.  Its super accessible, but offers most of what can be found in both Flagstaff and Sedona, with an added element of backcountry remoteness that gets a little harder to find in my part of the state.  I asked Matt to take me on two rides that would exhibit the opposite ends of the Prescott riding spectrum.  I brought Shrek down to see how 26" wheels and Flagstaff steep build would fair and met my riding buddies at the Granite Dells north of Prescott for a day of technical mayhem.  

Commonly the response of the people we encountered that day (hikers) was one of shock and disgust.  "You people ride your BIKES here?"  The dells offer the best in Arizona trials, straight out of the '90s. No doubt Hans Rey hisself has spent some time here.  Where Sedona gets off on big lines and picturesque landscapes to find itself pictured in nearly every magazine in print today, the Dells are just a little bit too far to the left to be mainstream.  What I mean is that even the most advanced riders at some point will be walking, or at least trying moves again, and always looking off line to see what else can be ridden.  The cool part is that as long as one is willing to try moves without getting discouraged, the area is actually quite friendly for group rides of multiple skill levels.  As slower riders make their way through challenges its never as though anyone is waiting with the surrounding playground.  At several of these points we stopped to make the bike switches.  What Matt was capable of aboard the 29er was instantly obvious to those of us watching him.  He had the confidence of a cat in native habitat instead of what more looked more like an uncoordinated ungulate taking its first steps.  The difference watching a truly seasoned rider was shocking.  We all saw it.  Aboard my bike I told him exactly what to expect, it is not as balanced and favors the downhill.  He jumped on for a move that transitions from a steep down into a steep up with a wheelie move in the transition.  He spit himself right off the back of my short stay 26er, just as I told him he would.

Our second day riding started with a familiar climb to the top of Mount Union in the Bradshaw range to descend the Yankee Doodle trail.  We and some others had done the first part of this trail on the last Intro to Bike Packing trip.  The first part would be dissuasive let us say.  Chunderry loose big rock rubble.  And while this fairly continues the whole way back down, the rubble does find its interruptions with fast sections of forest trail and near unnavigable rock drops.  This is the backcountry of the Prescott National Forest, usually only accessed by motos; it is in this steep terrain that my own bike shines.  Matt on his twelve year old full suspension Yeti, couldn't find the comfort zone of his bike in this rubble, and if it wasn't clear from the day before, it certainly was now.  He is riding beyond his bike's ability.  He didn't want to go faster down trails like Yankee Doodle, he just wanted to be able to make the moves with confidence.  Bearing these trails and rides in mind I went to work designing. 

Shorter stay 29ers are among the harder bikes to build in my opinion.  The large 2.4" rear tire eats up a lot of room so accommodations need to be made.  The actual stays (bottom bracket to rear axle) measured 429mm in the design leaving the horizontal measurement (what's left when bb drop is taken out of the equasion) at 425mm.  This was long enough to use my regular chain stays instead of a plate yoke and still have clearance for an XT double crank.  In selecting that double, however, a direct mount front derailleur was mated up for maximum tire clearance.  A curved seat tube was rolled out at my neighboring builder's (and sometimes my employer's) shop.  I turned down a piece of 1 3/8" sleeve tubing to reduce weight where the material isn't needed leaving the top to fit a standard size seat collar.  These pieces are fitted into the front triangle, thought figuring for and jigging an effective seat tube angle and an actual seat tube angle provide a bit of extra challenge.  Actual seat tube angle with the curved tube on this bike is around 69 degrees with the effective at the rider's saddle position being around 72 degrees.  The bike was fitted with a Thomson 27.2mm dropper post with under the saddle lever and the front end was designed around an Xfusion 100mm travel tapered steerer fork with 34mm stanctions for stability.  Interestingly I am finding that the benefits of larger stanction forks are much more noticable aboard hard tails than they are on full suspension bikes, as the rider tends to ride more "in the fork."  With the fork being the only suspension available a more forward standing position is assumed to push the front wheel through turns and obstacles (where prudent) when braking through these moves is less needed.  This in turn affects the design in terms of wheelbase and head tube angle.  The trick here is to figure out to what degree the intended rider will find any and all of this beneficial and balance all the elements of the frame's geometry.  In this case, Matt wanted a stable and predictable ride down the hairy stuff, not to go faster, so a shorter travel fork was selected to maintain the efficiency on the uphills and undulating technical trails.  

Above are the first specimens of the new custom drawn True Temper chain stay stock.  The stay on the right has the crush for tire clearance while the stay on the left just shows the raw bends laid out on the full scale overhead drawing.  These tubes will get and additional  bend clocked at 45 degrees at the rear to accommodate the disc brake on the chain stay shown in the photo below.  Dropouts are joined and brake tab attached to isolate the variables.  Then chain stays to BB is tacked.  Then seat stays bent up and mitered.

All is tacked in the jig, then removed for final brazing.  Then the real fun begins with hours and hours of polishing brass.  Finished bottom bracket assembly below.

Finished seat tube cluster

finished head tube assembly

Somewhere in the process comes the time to think about paint.  I have really had the best luck picking for people with guidance, as I know what comes out of the powder shop at this point and a good description often gets better results than looking at digital color swatches over the internet.  Matt had something specific in mind for this one, however, that warranted a trip up to Flagstaff to talk with my powder shop.  He wanted a flat green.  Extra polishing on my behalf.  Flat shows every flaw.  His description of the color was:  "There's this green plastic bin."

We got the bike built up in time for Matt to come up to Flagstaff for a test ride on our equivalent to the Dells, right before taking it out to the Sierra Nevada for a true test.  He was quickly out-riding me in my own back yard.  As for his extended field testing, well the words that follow are those of the rider:

  

Hello! So, headed back to Az, rode our last ride in Santa Cruz today. The bike has been awesome. Probably not by accident, it really came alive in my favorite ride of the trip, in South Lake Tahoe, which was quite possibly the best ride of my life. 5 hours of climbing on superbly laid out trail, fairly steep, periodic hard but rideable tech, insane views, huge trees, to a nice alpine pass. The descent was mostly through boulder strewn landscapes, but with a line through everything, some steep bouldery tech, back into the woods...
It was ideal for that bike. It turns really well, very light on its feet. And during the ride, I figured out the correct body position, which was back, of course, straighter legs and arms. In fact, especially when the seat was lowered, I found myself in positions I've never been in. So fun and fast and reactive, great on the steep steep. Its possible I e never felt more relaxed and tuned in at speed. Cool huh?
We rode three rides in Tahoe area, and they were all fantastic. The remaining two rides were south of SF and in Santa Cruz, and they were fun ripping in the forest, but nothing close to Tahoe.
So, I'm stoked. I can't wait to ride it in the dells, and the Bradshaws. It's best terrain is what it was designed for, go figure.

Thanks for the cool project, Matt.  Goats shred Trollville soon!