Author Topic: Team Go Dog, Go! Modified Partial Streamliners  (Read 1438628 times)

0 Members and 7 Guests are viewing this topic.

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 5503
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3480 on: December 27, 2019, 01:03:01 AM »
The concern Mike mentioned in reply 3469 is valid.  It is hard or impossible to check the sprocket alignment with a straightedge on most bikes.  A string strung tight against the sprocket faces was what I used for years.  These pages are from the owner's manual for my original BSA.  A version of the wheel alignment tool shown can be made for the sprockets in many instances.

Imagine planes across each sprocket's face.  The sprockets are in line if both faces are in the same plane.  The sprockets can be adjusted to be in line if the planes of both faces are parallel to each other.  A dished sprocket of modified sprocket mounts can be used to do this.

Life gets worse if the planes cross at an angle.  That can be a symptom of a bowed frame and is most likely the result of a wreck.  It is always a good idea to keep this in mind when trying to align the wheels and sprockets.  Sometimes a new frame, or straightening the old one, is the only cure.   

Offline Old Scrambler

  • Sr. Member
  • ****
  • Posts: 721
  • Going Fast - Slowly
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3481 on: December 27, 2019, 09:35:52 PM »
Bo......Be sure the bike and table are level and the wheels have been centered to the frame.....the laser is a simple table-top unit that must be leveled.  The beam can be raised or lowered.......so just set it on the bench behind and beneath the center of the rear wheel and aim it along the bottom of the rear wheel at its center and at the center of the front wheel......mark the rim centers on the bench....then move a little to the right and do the same, then to the left of center and make the marks..........all should be parallel.....check it with a straight-edge.....then set the laser behind and beneath the rear sprocket....center it along the bottom edge of the sprocket and aim to the center of the output sprocket.........make your marks and check for parallel lines with the wheel marks.

Dave Murre showed me how to do this and uses a lathe to machine the necessary parts. 
2011 AMA Record - 250cc M-PG TRIUMPH Tiger Cub - 82.5 mph
2013 AMA Record - 250cc MPS-PG TRIUMPH Tiger Cub - 88.7 mph
2018 AMA Record - 750cc M-CG HONDA CB750 sohc - 136.6 mph
2018 AMA Record - 750cc MPS-CG HONDA CB750 sohc - 143.005 mph
2018 AMA Record - 750cc M-CF HONDA CB750 sohc - 139.85 mph
2018 AMA Record - 750cc MPS-CF HONDA CB750 sohc - 144.2025 mph

Chassis Builder / Tuner: Dave Murre

Offline Bookfla

  • Full Member
  • ***
  • Posts: 57
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3482 on: December 30, 2019, 04:38:18 PM »
Lasers are not too bad but it is hard to get finite definition under say .060". If you have a clear pathway a newer style Florescent T5 lamp at 5/8" diameter can get you a cheap straight edge to align from. It also works great for wheel and swing arm alignment.

Offline charlie101

  • Full Member
  • ***
  • Posts: 150
  • Indian 101 buff
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3483 on: December 30, 2019, 05:56:16 PM »
Have you tried to stress the mounts for the swingarm and then measure the alignment? There's a tremendous pulling force at the chain when going full speed on the salt. A loop wire with a turnbuckle, between the sprockets, would somewhat indicate how flexible the frame is. 
« Last Edit: December 30, 2019, 06:01:22 PM by charlie101 »

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 5503
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3484 on: December 30, 2019, 11:52:54 PM »
Figuring out the flexure at the swingarm to frame junction is mental.  That disqualfies me from taking that approach.  Some bracing will be designed so I can be confident that there will be minimal flex.

Spoked wheels were used with inner tubes.  One piece wheels will be used to eliminate the inner tube and to get a more desirable rim width at the back wheel.  The bike shop that helps me, Cascade Moto Classics, had this pair of rims.  They were dented where someone ran over a curb.  They gave them to me and they are for a 2014 Triumph Tiger. 

The rims were sent to AccuTru in Kettering, Ohio.  They were recommended to me.  They were told about what I plan to do with the bike.  They said the repaired rims should be good for over 200 mph.  The bike will never go anywhere near this fast so the fixed rims should be OK.  They did a good job.  I cannot tell that they were dented.  Fixing up these old rims saved about $500 compared to buying new ones.   


Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 5503
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3485 on: January 01, 2020, 12:06:06 PM »
Folks that design a lot of this new stuff never worked as mechanics.  The inner spacer between the wheel bearings has close to same inside diameter as the bearings.  Expandable bearing removers cannot get a good grip on the inside face of the inner race for removal.  The wheel was heated so the aluminum bearing housing would expand more than the steel bearing.  This reduced the grip from the interference fit between the wheel and the bearing.  The bearing puller is this thing I call the "foot."  It is wedged in the spacer and pushed sideways so there is friction between the foot and the inner spacer.  Then the slide hammer is used with the foot to shift the bearing slightly outwards.  Now there is a small radial gap between the spacer end and the bearing.  Conventional expanding pullers can be used to finish the bearing removal. 

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 5503
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3486 on: January 01, 2020, 12:29:42 PM »
Ceramic wheel bearings were investigated.  Literature from companies that specialized in ceramic bearings was ignored.  They have a big financial stake in what they say.  This could make the recommendations suspect.  Engineering data was reviewed from big established bearing companies such as Timkin, Toyo, GMN, NSK, NTN, and others.  They often sold ceramic and many other bearing types.  Their guidance would be more impartial, is my hope.

Wheel bearings are sealed low speed deep groove radial ball type.  Lifespan is not an issue with a seldom used race bike.  There was very little, if any, advantage for ceramic bearings in that use based on the literature.

Then, some investigation was done on the top of the line performance bikes from major manufacturers.  The four cylinder BMW1000SS would be a example.  All appear to use conventional steel bearings.  So, I decided to use steel bearings, too.

The Tiger 800 bearings removed are Japanese NSK 6204DU and they are replaced with Polish NSK 63030DU with the same width and outer diameter.  The inner diameter is smaller to match the Triumph Bonneville axle.   

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 5503
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3487 on: January 03, 2020, 08:50:44 PM »
Quite a lot of fabrication is needed to make this wheel swap and the problems are common to the typical job.  Lots of lathe work will be done on a 1947 belt drive Logan lathe.  It has a hard time machining the common 304 stainless steel alloy.  Very light cuts are needed so the belt will not slip.  Unfortunately, the shallow cut depths are within the work hardening zone and tool life suffers.  This problem is bypassed by using 303 annealed stainless.  Parts in direct contact with salty water will continue to use 304 alloy 'cause of its superior corrosion resistance.   https://www.clintonaluminum.com/stainless-steel-303-verses-304/

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 5503
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3488 on: January 03, 2020, 09:35:03 PM »
A spacer spreads apart the inner races of the wheel bearings.  Flanges in the hub separate the outer races.  It is important that the inner races are the same distance apart as the outer races so the balls in the bearings will be centered in their grooves.  The Tiger 800 spacer is shown between the bearings in the picture.  It cannot be used.  The Bonneville axle and bearings have the same widths but their inner diameters are smaller, so new spacers are needed.  A new rear wheel spacer is made from 304 stainless and the last operation is to cut it off at the proper length. 
 

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 5503
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3489 on: January 03, 2020, 09:43:51 PM »
The cutoff bit is easy to select based on the info on the back of the pack with the bits.  There is an "X" in the "Cut-off" box covered by my thumb.  Some precise measurements are needed so the tools and parts are put in a room with a temperature around 68 to 70 degrees.  They are there for an hour to warm up. 

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 5503
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3490 on: January 05, 2020, 01:09:53 PM »
The new stainless steel spacer should keep the wheel bearings the same distance apart as the aluminum Tiger 800 spacer.  The spacers are made from different materials and they fit axles of different diameters and bolt torques.  This is a typical problem during a wheel swap and the next few posts will have the calculations.

Stainless steel is a lot heavier than aluminum.  A new spacer made to the same dimensions as the aluminum one would weigh more than necessary.  Fortunately stainless steel is much more rigid than aluminum.  The two spacers can be considered to be springs under axial loading.  The modulus of elasticity is a measure of the stiffness.  It is 28,000,000 pounds per inch for annealed 303 and 10,000,000 psi for the aluminum.  The steel is a stiffer spring.

The desire is to have both spacers deform the same amount under identical axial loads.  The spacer lengths will be close to the same so the moduli of elasticity and cross-sectional areas are the only variables to be considered.  Cross-sectional area of steel spacer x modulus of elasticity of steel should equal cross-sectional area of aluminum spacer x modulus of elasticity of aluminum.  The aluminum spacer has 0.560 square inch area.

Area steel spacer = 0.560 x 10,000,000 / 28,000,000 = 0.200 square inch
 
A hole is drilled to 0.692 inches diameter in the steel piece to fit the smaller diameter Bonneville axle.  Some basic math says the outer diameter of the steel spacer needs to be 0.857 inches.  This is considerably less than the 1.170 inch aluminum spacer diameter.  Some weight savings can be had by considering the higher strength of steel in the design calculations.

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 5503
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3491 on: January 05, 2020, 03:27:30 PM »
Now the spacer design has progressed to this.  Both spacers are 5 inches long.  One is annealed 303 stainless steel with an outside diameter of .857 inches with an inner diameter of .692 inches.  The other is aluminum of indeterminate alloy with an outer diameter of 1.170 inches and an inner diameter of .810 inches.  Both spacers have equal resistance to compression deflection.  This assumes the compression pressures are below the yield stresses of the materials.

Normally I do a yield stress and column buckling design of the original part and the designed replacement.  The attached shows the horrendous differences in yield strengths of different aluminum alloys and tempers.  No design or analysis can be done on an unknown piece of aluminum.
https://www.amesweb.info/Materials/Aluminum-Yield-Tensile-Strength.aspx 

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 5503
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3492 on: January 07, 2020, 01:03:39 AM »
Disregard posts about stainless steel spacer.  I made a big design error.

The axle bolt is always tightened with a torque wrench having a fairly recent calibration certificate.  The torque is 65 Newton meters or in imperial measurements, 752 pound inches.  The nut is supposed to have clean and dry threads and a typical 0.20 thread friction coefficient would apply.  This would give 5,970 pounds tension for the 16 mm diameter axle bolt threads.  The axle is in a greasy environment and it is very possible there could be some chain lube on the threads.  A lower 0.15 thread friction coefficient would apply.  This produces a much greater 7,960 pounds bolt tension.  The greater bolt tension will be used in the spacer redesign.

This was the step that I forgot to do when designing and machining the stainless steel spacer.  It was designed for the 5,970 pound bolt tension and it would have collapsed and buckled under the 7,960 pound lubed thread tension. 

   

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 5503
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3493 on: January 07, 2020, 01:41:34 AM »
The spacer in the Tiger 800 wheel is shown.  It is indented where it contacted the bearing inner race.  This is what I want to avoid.

The old Tiger 800 wheel bearing is on the right and the new Bonneville one is on the left.  The thrust face on the inner race is what contacts the spacer and the pointer shows it on the Tiger bearing.  The thrust face area is 0.352 square inches on the Bonneville bearing.

The 7,960 pound bolt tension is on this 0.352 square inch face.  7,960 / 0.352 = 22,600 pounds per square inch.  A rod of 6164 aluminum alloy with T6511 temper is in the scrap pile.  It has a 34,800 psi yield strength and it will be more than adequate.  The factor of safety is 34,800 / 22,600 = 1.5.  As seen from the reference a few posts previous, many aluminum alloys would not have yield strengths above 22,600 psi.  Alloy 6061 is relatively easy to machine and corrosion resistant when compared to other aluminum alloys.  This is an added bonus.       

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 5503
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3494 on: January 07, 2020, 03:03:44 PM »
This is a link to the buckling calculator. https://www.amesweb.info/Beam/Column-Buckling-Calculator.aspx

We did hand calculations for column design when I worked for the highway department.  This was done so someone could see how the answer was calculated.  This program is what I used to check my hand calcs.

Cut and paste versions of the section properties table and the buckling output are attached.  The 0.58 safety factor is less than 1.0.   The spacer would have crumpled when an oily axle nut was tightened.