(Note: Donations are not tax deductible)




Forum's been "upgraded".
Things will look a bit different.
A *few* posts might have gotten lost in the crossover.
PM bobc with problems, or post in "Website Suggestions".

Author Topic: Team Go Dog, Go! Modified Partial Streamliners  (Read 774453 times)

0 Members and 3 Guests are viewing this topic.

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 4939
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3300 on: February 08, 2019, 10:51:40 AM »
Desired clearances are as recommended by Bill Mc Knight at Mahle.  He looked at the shells and I gave him all sorts of technical info to help him make good recommendations.  This was a few years ago.  Recommended clearances are .00075 to .0010 per inch of shaft diameter + .0005 extra clearance for racing.

This is .0017 to .0021 for the rods and .0016 to .0020 for the mains.  The mains were installed at .0015 and the rods at .0017.  The rods were originally at .0015 and I used hand polishing to increase the clearance to .0017.

Marine Crankshaft has been recommended to me as a resource with experience helping the flat track team.  That is where I will start.  If that does not work out I will try the folks in Denver.   

Offline charlie101

  • Full Member
  • ***
  • Posts: 148
  • Indian 101 buff
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3301 on: February 08, 2019, 01:11:26 PM »
A good idea is to take a photo of every part before shipping, AND the parcel on site at the post office or where you send it from... Just saying it saved me a lot of BS from...

Offline ggl205

  • Sr. Member
  • ****
  • Posts: 584
  • G/FL 218.282 since 1995. G/FL record since 1993.
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3302 on: February 08, 2019, 01:54:39 PM »
Bill Mc Knight was unquestionably the man to talk to about bearings. Before he retired, Bill was intimately involved in all genre of race and performance engine building and very approachable. But Bill is now retired and doing other things so we are left with his legacy and other highly qualified people still active in our industry.

After looking at your clearances, WW, you were most definitely on the tight side for rods and mains.   

John

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 4939
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3303 on: February 12, 2019, 10:23:51 PM »
This build diary shows how to do basic things.  This is ordering flow bench testing.  The standard suction for flow bench data in most applications is 28 inches of water.  It is possible to use a bench with as little as 10 inches of suck and to convert it to 28 inches using equations.  This is adequate for most situations with non complex air flow.  The plenums may have complex internal flow patterns with turbulence losses.  A flow bench that can pull a full 28 inches is chosen to do the work.

The flows are measured at various valve lifts.  There is little chance a lift over 0.45 inches will be used on this engine due to geometric limitations like the cam lobes being to big for the lifters.  Tests will be done up to 0.45 lift with 0.025 increments.

The exhaust port is often fitted with a little stub to resemble  an exhaust system like shown in the first foto.  Port flow data for the head, only, is used in many computer programs.  The additional losses in the intake system are calculated by other methods.  Flow data for the entire intake system is what I am currently using in the programs.  Intake losses in the manifold, carb, filter, etc are being "double counted" at present.  Flow data for the cylinder head, alone, will be ordered.  The second pix shows the intake side of the head.  The valves will be in place during the testing.

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 4939
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3304 on: February 12, 2019, 10:33:52 PM »
The first picture shows the carb and manifold.  The throttle is wide open during flow testing in most applications.  There needs to be a way to make this happen.  The intake system used last year is shown in the second photo.  It is a filter attached to the carb end.  Tests will be done with this to make a baseline for comparison.  The engine put out over 100 horsepower which is not bad for one of these dogs.  More is needed.       

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 4939
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3305 on: February 12, 2019, 11:36:25 PM »
Potential horsepower from a well developed engine can be estimated based on flow bench data using an empirically derived equation.  This is from Page 50 of Harold Bettes' "Engine Airflow."  hp/cyl = Cpwr x test flow, where hp/cyl = horsepower per cylinder, Cpwr = coefficient of power, and test flow = cfm flow at the same test pressure that Cpwr is applied.  This is a two cylinder engine, Cpwr = .26 at 28 inches test pressure, and test flow at maximum lift = 230 cfm.  It is assumed that 92 percent of the power is reaching the rear wheel.

Rear wheel hp = 2 x .26 x 230 x .92 = 110  This is a realistic estimate.  It shows that any increase in intake flow can result in more power.  A higher lift cam is worth investigation.  Flow data shows that it is not an option.  Flow at .425 and .450 lift is only 231 cfm.  This is intentional.  The bench data for flow through the entire intake system shows that the flow curve flattens at around 231 cfm above .4 lift.  A cam with .4 lift was chosen.  More lift would not do much good.

The air is drawn into the filter across a sharp edge where the screen meets the black plastic thing that is clamped onto the carb end.  That is an inefficient situation and it leads to high inlet flow energy loss.  Much less energy is lost if the intake end has an elliptical shape.  This is hidden inside the plenum shown in the pictures.  It is described in recent posts to this build diary.

The dominant wave effect characteristic at full throttle and high rpm is the resonant harmonics between the bell mouth end and the inlet valve inner face.  Chrysler did a lot of research on this and I call it the "Chrysler effect."  Much guidance is available on the tuned length for optimum performance.  Much less research has been done on the shape of the bell mouth and the tapers leading to it.  There is a research report on this I read very late at night a few months ago.  I could not find it the next day and I should have saved it on the computer.  This shape is more dominant than tuned length for a four valve engine with substantial cam overlap, in my experience.  The elliptical bell with the rounded edge worked best for me in the past.  It will be used for this plenum inlet system.

The first flow test with the plenum will be with an unobstructed entry from the filter.  See the first pix.  The test will be on the complete system shown in the second picture.

   

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 4939
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3306 on: February 12, 2019, 11:49:14 PM »
The average demand for air into the bell mouth, not including fluctuations based on valve opening and closing, is lower than the peak demand.  The peak demand when the inlet port is filling can be up to 6.25 times as high as the average demand.  This is well explained on pages 50 and 51 in Don Terrill's "Horsepower Chain."  This factor is for a single cylinder like the intake on one side of this two cylinder engine.  This peak pull is trying to jerk the flow through the filter screen without the intake plenum.  The plenum provides a reservoir of air upstream from the bell mouth that can be pulled in during the peak demand.  This reduces peak flow demands on the filter screen.  This is called plenum effect.  The plenum volume is 1.25 times the 500 cc swept volume.  This extra allows for the volume of the combustion chamber and 110 percent increase in volumetric efficiency.  The plenum is sized to provide one gulp of air.   

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 4939
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3307 on: February 13, 2019, 12:13:14 AM »
The acoustic vibrations within the plenum can enhance or screw up the intake harmonics.  This is often called the Helmholtz effect.  There is a lot of variance in opinions in the literature about this phenomenon.  My feeling is that it is a weak force compared to the Chrysler effect when the intake tract harmonic resonant length is near optimal.  This would be at low midrange and at high midrange and top end of the rpm band of this engine.  The harmonic resonance is off key at midrange and the power suffers.  There is a dip in the power curve.  The Helmholz waves can be a benefit in this midrange dip area, is my feeling.

The plenum is acoustically isolated from the filter using the walrus valve.  Without the valve the volume is the inside of the filter and the plenum.  The valve reduces the plenum volume and it increases the Helmholtz resonance frequency.  It also increases the intensity.  It is hoped this will reduce the dip in the power band at midrange.  This will make the bike easier to ride and faster on loose surfaces like the sand at Pendine.  This last picture shows the valve in place and the complete system.  Flow tests will be done with the valve in place, too.

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 4939
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3308 on: February 17, 2019, 11:37:19 PM »
The loading ramp for the trailer is a 6-inch wide aluminum channel.  It is too narrow for the rear tire.  It was sawed right down the middle.  The two halves were spread 1.75 inches apart and four 1/4 inch thick gusset plates were lap welded onto the bottom side to hold the rails in position.  A 1/4 inch thick plate was sawed out to fit in the gap.  The channel web is thinner than the plate.  It is .180 thick.  One inch long welds were made on each side of the plate 4 inches apart.  The welds overlap each other so the top side welds are directly over the bottom side welds.  Some of the middle welds cracked completely through while the ramp cooled after welding.  Any help is appreciated.  I have no clue about what caused them or the best fix. 

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 4939
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3309 on: February 17, 2019, 11:38:36 PM »
More crack fotos.

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 4939
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3310 on: February 18, 2019, 09:01:23 AM »
The extrusion (channel) is 6061 -T6.  The wire is ER 4043.  These appear to be compatible.  The plate can be either 6061 or 5356.  The supplier stocks both alloys.  I will check with them today.  The plate finish is very shiny.  I have never seen this on 6061.

Weldability and hot cracking are a concern.  The lap joints seem OK and both materials are welded together there.  All problems are in butt joints.

Decades ago I watched a guy make something like this out of aluminum.  He welded a 1/4 inch length of each of the many welds on the thing.  The welds cooled from the earlier addition before he made the next one.  He said this reduced the stresses in the structure due to heat related distortion.  I did not do this.  Maybe I should have.

Offline Peter Jack

  • Hero Member
  • *****
  • Posts: 3621
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3311 on: February 18, 2019, 10:04:44 AM »
I would use a prep where you're making the welds and then use a good preheat so the whole area is well warmed. Short welds might certainly help. You say the material is shiny, is it anodized? 6061 material will definitely work better than a 5 thousand series alloy. PM me or email me if you want to discuss it further.

Pete

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 4939
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3312 on: February 18, 2019, 11:07:01 AM »
Thanks, Pete.  The welding shed is cold and often I can see my breath when I am working.  The heater is turned on in the shop.  The ramp can be set on the wood stove in the parlor so it can preheat.  Then I can grab it and run out to the shed and weld it while it is still hot.

The steel yard says the plate is 5052 or 6061.  It is shiny and I will sand it to remove any anodizing.  The welding shop said to switch to ER 5356 wire if there is a chance the plate is 5052.  So, I bought a spool and will try it.

There are a few pieces of 6061 plate in the scrap pile.  They will be used for the more critical parts of the trailer rebuild.  The welding shop said to use similar materials if I can.

Go ahead and send me a PM.  My experience is mostly with steel so I do not know much about welding aluminum.     

Offline Peter Jack

  • Hero Member
  • *****
  • Posts: 3621
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3313 on: February 18, 2019, 12:00:06 PM »
Bo, for preheat I'd be tempted to get a propane torch similar to Harbor Freight #91033 and a 20# propane cylinder. This will allow you to add more heat if needed and you won't risk doing acrobatics carrying around hot metal. The advise you've been given sounds solid. I'd stay away from the 5356 wire unless you have to use it. It's much more brittle and subject to cracking. 5356 is much nicer to use if you have to do any machining as it has much less tendency to clog the tooling. It's much easier to grind with a carbide burr.

When you're not using the 20# cylinder store it outside or under a well ventilated lean to. Propane is nasty stuff if it leaks indoors.

Good luck.

Pete
« Last Edit: February 18, 2019, 12:02:23 PM by Peter Jack »

Offline wobblywalrus

  • Hero Member
  • *****
  • Posts: 4939
Re: Team Go Dog, Go! Modified Partial Streamliners
« Reply #3314 on: February 18, 2019, 06:39:49 PM »
The torch would need me to have four eyes and three hands to hold it and to weld.  A camp stove was deployed to provide heat and a piece of train track to hold down the ramp.  5356 wire was used to tack.  Note the cracks in the tacks.