Friday, June 29, 2012

One Trick Pony?

Although I have not a chance to do more bike work I have been busy.    Sometimes I feel a bit like a one trick pony:  start with a billet of aluminum, end up with lots of chips and a few parts.  See below.  At least these parts were for a true racebike.  Maybe I need to do some welding to break out of the funk?

I hope to get back to some bike parts in the next week ot 2.


Saturday, April 28, 2012

One more in the 'Done' column

I was supposed to go to Lime Rock Park today to watch the Porsche Owners Club flog their cars around the track.  While I'm usually up for anything to do around a racetrack when I walked out to the bike this morning at 8am and saw a frozen puddle in the street the 2 hr ride to the track suddenly seemed much less attractive.  I made the obligatory 'I gotta back out last minute' call, took the heat, then went back to bed!

After a couple hours more nice warm sleep I decided to make some use of the day and finish up the swingarm pivot plates.  The shop was deserted and quiet, nice conditions to make some progress.

I had left the parts setup in the mill so that realignment for the bearing bore operations would be easy.  I can use the same boring bars used for the engine bearings.  The bores went quickly, so fast I missed them with the camera!  Strike that, I had dropped them in the wrong folder.

Bore out one bearing hole...
Then flip it and bore the other:
Machining the reinforcing rib tapers and surfacing the outside of the bearing tunnel:

With all the fixtures and programs done and verified, making a second part only took a couple of hours:

The parts look great and I'm one step closer to getting Scott down for some sheet forming and welding sessions.

Next up is the upper shock mount crossbar, a much simpler part that should go quickly.

Free t-shirt to first person that spots something that's amiss in the pictures.  Post it in the comments and google will be the time-stamp arbiter.

UPDATE:  here's the problem that can be seen with a close examination of the second to last picture:

That slot in the bearing pocket shouldn't be there!  It was a problem with a Z retract value that was defined incrementally instead of absolutely so when the tool went to start the finish pass the Z retract did not retract enough and instead of clearing the part plowed through it.

I'll have Scott Kolb weld this mistake up and then remachine the bore to size.  Normally there would be a concern about the welded material in a stressed area losing its temper and strength but this part will be heat treated after the entire swingarm is welded so it will then regain its full T6 temper strength.


Friday, April 27, 2012

Progress on the Swingarm Pivot Plate

Over the past few days I was able to get more machine time in on the swingarm pivot plate.  I made a lot of progress but didn't quite finish up.   For the first time in a while I had a couple of broken end mills slow progress down.  The order of some roughing operations was incorrect and when you have a 1/2 end mill sticking out of the collet 2 1/2" it doesn't take much of an unmachined area to increase the chip load make the end mill snap like a twig.
The fracture surface is something I find interesting.  These cutters are made from cemented tungsten carbide, an extremely hard material made by sintering extremely fine metal powders under high temp and pressure.  The resulting material will quickly dull a common shop file and still retain an unblemished mirror finish.  The only drawback, and is shared by many ultra high strength materials, is that it is brittle.  You can load it agressively but not quickly.  Unplanned high loads, say like plowing into an unmachined area, make the material behave more like glass than metal.  You can see the fracture surface is extremely smooth and none of the tearing appears that is exhibited by most lower strength materials.

This clean fracture is an indication of high quality material.  There is no bending or elongation with carbide.  It's all or nothing, and in this case, nothing.  The good thing about mill tool breakages is that if you catch it before a toolchange then usually it's nothing more than resetting the milling chuck with a new end mill.  that's what happened this time, no part damage, just a slight snap and then silence.

To rewind a bit to where we were at in the previous post, I had the part blank in its usual place on a subplate on the trunnion:
I then used that wonderful Ripper insert mill that was used on the TZ cases to do some serious roughing and break through to the shock pocket machined in the previous operation.
This left the part in a semi-recognizable state.
Then used a 1/2 ball end mill to surface machine the outside of the pivot bearing tunnel:
Then index the table 90 degrees and rough machined the bearing bores and finish machined some of the interlocking features that locate the aluminum sheet parts:
Which left the part in this state where I called it quits for the day:
The next step will be to break out the boring bars and machine the dual bearing bores on this side.  Then I'll spin the table again and machine the opposing side and the bearing bore for the dual needle bearings.

The pivot bearing arrangement is one that I thought was designed quite cleverly, then upon further research seemed to be the way most modern Japanese sportbikes are done!  One major design constraint of the swingarm pivot is to keep the bearing diameter as small as possible in order to keep the sprocket to pivot distance as small as possible.  This is desirable because it both reduces the variation in chain tension and allows for a longer swingarm for a given wheelbase, a rare win-win situation.
This pivot axis uses a dual needle roller bearing on the chain side because needle bearings are very low profile and can handle high radial loads.  This allows me to keep the pivot close to the sprocket and withstand the considerable chain pull forces.  However needle roller bearings no not provide any axial load capacity so something else needs to be done to keep the swingarm from flopping from side to side when the bike leans.  On the side opposite to the sprocket there is some more room for a larger diameter bearing so that's what I do.  The double row ball bearing provides excellent axial load capacity when installed with a retaining ring and just to be safe on the radial load rating I tuck another needle roller bearing in next to it.  There is a long hardened steel inner race that the needle bearings ride on and this inner race is clamped against the inner race of the ball bearing in order to provide a continuous inner stackup that can be torqued down against.  This arrangment leads to a very rigid pivot joint, a desirable trait in any axis of motion.  Of course, both sides are sealed in with low friction shaft seals.  A dirty bearing is a horrible sight on a racebike.

Hopefully these features will be machined this weekend.


Wednesday, April 25, 2012

Thanks Guys

Here's a little update on the fundraising front.  T-shirts have been going out to the twenty or so people who generously contributed at that level.

I've also had one taker of the $250 level for fairing display and that was Chris Hodgson from San Jose BMW.  I've known of Chris since his days with CC products, having been exposed to the rare BMW motorcycle racing bug through Gregor's Boxer project.  The engine was stuffed full of CC goodies.  I also met Chris a couple of years ago when he pitted next to Scott's LSR bike at Bonneville.  I had some of the V4 engine parts with me to show at the AHRMA Miller round and he was like 'Hey, I follow that project online!'.  It was a rare moment of the student becoming the master, or something like that.  Anyway, Chris dug deep and helped out.  Thanks!

Over the weekend I had the first $500 donor to have his logo engraved on the engine cases.  The person was Lawrence Smith, an Aussie that not only loves performance bikes, and not only knows machining and engineering, but is a FFE (funny front end, an industry term...) advocate!  Amazing!  How do you know he is a FFE lover you ask?  Well, this is his first project bike:
Pretty damn cool looking IMO.  There's a nice writeup with some pics and video over at TheKneeSlider.  Lawrence is very happy with how the bike performs and in his own words 'after all FFE is the future'

He sent me some pictures of his next project that just got on the road:

Lawrence took nearly the opposite approach to me in his effort:  he wanted to maximize reuse of an existing bike and do something that would get him riding quickly.  His use of the girder with its steered arms is one of the decisions that made a retrofit easier as he could still use a headstock style frame, although heavily modified in that area.  His results are impressive and allow easy back to back comparisons with stock bikes.  I'm sure that has he gets the geometry optimized it will be a very positive comparison for the FFE.

Anyway, that was really cool of him to contribute to my project when he has his own going on too.  You rock.

Some more swingarm pivot plate machining will be coming soon.


Saturday, April 21, 2012

From Front to Back

It feels food to have gotten some of the front suspension linkages completed.  For the next part to make I'll switch from the front of the bike to the back and do some work on the swingarm.  I know Scott has had plenty of spare time lately and would like nothing more than to start forming the sheet and welding it up.

The swingarm is a welded assembly of a lot of folded sheet and billet CNC parts.

The sheet parts were cut out by North East Water Jet from 6061-T0 material because of the various forming and bending operations that need to be done.
T0 temper is full soft and can easily be bent without tearing or other material disruptions.  T4 or T6 temper are a lot stronger but have issues with formability.  Since T0 temper is so soft, you may ask how will it be strong enough to withstand the loads of a roadrace motorcycle.  The answer is simple: after welding, the entire assembly will be heat treated to a T6 temper.

This is as of yet an unresolved area.  I need to make a welding/heat treat fixture that is light enough to manipulate to gain welding access, strong enough to keep the parts in proper alignment, and of a material that can withstand the T6 processing temperature of aluminum without weakening.  Good candidates are cast iron surface plates or old machine tool tables or beds.  As I have neither on hand this needs to be sourced at minimal expense.  If any readers have something suitable please drop a line.  12x24 is a minimum size.

The billet parts are made from 6061-T6511 material as getting material in bar form in tempers other than T6xxx is difficult and largely unnecessary as bar material is rarely bent or formed, it is usually machined, for which a T6 temper is desirable.  T4 and especially T0 tempers tend to be gummy when machined and leave an unsatisfactory surface finish.  Some areas of the billet parts will lose their T6 temper during welding but will be restored to full strength by the subsequent heat treat process.
Scott machined the axle mount/chain tension adjuster parts and they are damn good looking.  The remaining billet parts are the linkage mount/pivot bearing support and the upper shock mount.  Today's post will be on the first machining steps of the linkage mount/pivot bearing support part.

This is a decently large part that will withstand considerable loading from chain tension, shock force, and tire traction.  The part has several features to locate and interlock the various sheet parts in proper alignment.  today's post will be some of the relatively simple bottom features: the shock clearance hole, the linkage mount pocket, and some fixturing holes and clearance operations.

I will be able to make 2 parts totaling 5.5 lbs from this 73lb block of billet material.  It may seem wasteful but is really the most efficient way to make a couple of untested parts.  The waste chips will be recycled into delicious Italian heroes.
Most woodworkers cringe when they see me machining aluminum on a table saw.  In reality it is a great way to trim stock accurately to size.  Bandsaws are nice but leave a ragged edge.  With the proper carbide tipped blade the resulting finishes are excellent.  In my infancy as a machinist I was as reluctant as anyone to cut metal on a table saw but years ago after witnessing Eric Moore cut stock for a project to size and square within +/-.005, I was a convert.  My technique is not as smooth as Eric's but it gets the job done.

For anyone that doubts that I actually cut 2.5" thick aluminum plate (in 2 passes) with the table saw....
Smooth as butter.  A little WD-40 goes a long way in saw cutting aluminum.

With these two plates I needed to machine trunnion mounting holes then will flip the part and machine the underside.
Here it is clamped to the trunnion:
And here are the 4-16mm reamed/M12-1.75 threaded holes i use to bolt parts down:
 After machining both blanks I bolt one ot the table and start machining:
The 2" ripper mill used in the TZ250 case project worked like a dream on this part.

Two blanks, two parts!

Now I need to do a bunch of CAM programming as the 'inside' of the part is much more complex than the bottom size.  Hopefully will be able to post more this week.


Tuesday, April 17, 2012

Finishing the Upper Control Arm

Good weather and motorcycle parts, almost a perfect combination.  It would be perfect if all those motorcycle parts fit together to make a complete bike you can ride around in this summer preview.  One day soon!

I was able to get more time on the project and finished up the machining of the upper control arm.  I was able to make a second use of the baseplate for the crankcase girdle as one of the needed holes intersected with any of the existing holes.  After a little bandsaw prep to cut off large excess stock areas I was ready for more machining.

Those chunks went into the recycling bucket that gets emptied every few weeks and results in a nice Lisa's Deli sandwich for everyone lucky enough to be in the shop that day.  They make great fresh mozzarella.

Now back to the mill with the part bolted onto the trunnion with a baseplate and a few precision spacers to lift the part a bit off the table surface.
The first operation results in a recognizable part and a lot of chips.
Then I turn the table and rough out the bearing bores and give the arm its tapering profile.
Now fine boring the bearing pocket to a M6 tolerance grade.  That's Peter's boring head and bar.  Its a very nice setup he got from Mari Tool.

Then I spin the baseplate and drill and counterbore the trail adjustment screw hole.
This operation is semi-manual, the tool is a 2 piece assembly: a keyed shaft and removable counterbore cutting head.  It is used a lot in the aircraft industry where it is not possible to get a spindle on the correct side of a hole to make a counterbore.  So at the price of a little manual intervention you can get a clean spotface for a screw head where you otherwise would not be able to.  It is a good technique which I've used on both V4 crankcase castings, some main case bolts on the recent billet TZ250 crankcase project, and now on this suspension link.  There are automated tools that accomplish the same thing at a greatly increased price.  When my production volumes make the manual technique unwieldy than an automatic tool will be purchased.

There are 4 screws that hold this adjustable arm design together.  To maximize clamping force over the largest possible area instead of using washers I machined some 6Al-4V Titanium cap plates.  These cute little parts prevent marring and ensure a tightly clamped joint.  Since the trunnion was still on Peter's mill I used my Bridgeport CNC knee mill to make these.  While this machine is in no way capable of making 90% of the parts in this project it is perfectly suitable for a lot of the smaller less dimensionally critical parts.  There was even a piece of aluminum already clamped in the vice that I could use as a baseplate.
I had a sheet of Ti from another project that was the perfect size.  I machined 2 holes on the manual mill, bolted the part down, than ran the program that automatically separates the part from the extra stock upon completion.
Drill new holes, repeat a few times and there are enough parts for 2 complete arms.
One more assembly to cross off the To-Do list.  Yea!


Friday, April 13, 2012

One More Part Checked Off the List

The CAD software I use, Pro-Engineer, has an assembly mode where you can create simplified representations of the assembly you are working on.  One simplified rep I have defined is the 'Build' state where I blank parts that are finished and today I went into it and checked off 'arm-upper-bearing.prt'.  The current build status is:

Some days the state of this assembly is depressing and some days it cheers me up.  Today I was in good cheer.  Lots to do, but lots done.  The reason for my cheer is a pair of nice looking multi-axis parts:
As you can guess, this is the result of the second operation on some of the parts in the last post.  In the previous operation I had machined all of the critical functional features:  bearing bores and locating tongue.  this second operation was merely to remove material along the outer perimeter of the part.  Although it needs to be done carefully to leave a balanced amount of material, the tolerances are much looser than the +/-.0002" the bearing bores needed.  For comparison that is about 1/10th the thickness of a human hair.

The fixture was simple, a machined boss that would locate in a bearing bore and a threaded hole to clamp the part down.  I also had to machine a step to keep the part from spinning due to machining torque.
Here's the part bolted in and being machined.
After completing the outer profile I indexed the trunnion and drilled the hole for the trail adjustment screw thread:
Here's the part assembled to the steering stem:
Its starting to look really nice and I guess that's reason to be happy about the build image.  I'll do the same for the chassis pivot part of the arm assembly tomorrow and keep the ball rolling.


Progress on the Upper Control Arms

Ahhhhhhhhh.  Machining and fabricating sweet parts is nice but doing it for your own bike on a nice sunny day just seems to be that much better.  Something about the sun, sky, and the self determination of being on your own that makes the air taste a bit sweeter.   Or maybe its just some new type of pollution.

The upper control arm is going to be a 2 step type of part.  Because it is the main adjustment point for fast turning/front weighted bike to stable/rearward weight bias I needed to be able to quickly make and test changes while keeping a rigid part that will not introduce any vagueness into the system.  The solution I came up with is a tongue-in-groove sliding joint with a screw adjustment to keep the setting under heavy load.

This 2 piece design will allow quick and accurate trail changes in the hot pits with no disassembly or swapping of parts.  For our first round of bikes and testing this is what we will use.  The plan is that testing will reveal several preferred settings which we will then machine in a fixed length one piece arm:
This one piece arm will be lighter and, more importantly, more rigid laterally.  The overall stiffness will be similar to the 2 piece design but the introduction of a bolted joint into an assembly inevitably results in the introduction of tiny amounts of hysteresis which can possibly have a negative effect on performance and feel.  As a tuner it is nice to have a chassis with adjustments at every pivot point but the cumulative affect of so many bolted joints can compromise the overall performance of the chassis.  This one joint in the assembly should not have significant adverse effects and will let us quickly dial in to optimal trail settings for a variety of conditions.

Now, on to machining........

This is a process we've all seen before:  billet, vice machining, flip, subplate machining, then repeat.  The suspense is gone but the cool pics are all new.

Here's the pile of billet that will soon become two complete control arms:
Here's the mill with the trunnion and vice mounted on it with stock clamped:
 Machining in progress:
Done with the first step of upright end machining:
Now on to the chassis pivot end:
Repeat both a couple of times and you get the resulting picture:
Now I need to machine a couple of fixture plates to hold these parts and then run the second and final operation.  Will post again real soon.