Landracing Forum Home
November 25, 2017, 12:25:47 AM *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
News:
BACK TO LANDRACING.COM HOMEPAGE
 
   Home   Help Search Calendar Login Register  


(Note: Donations are not tax deductible)







Live Audio Streaming and Archives of Past Events
Next Live Event: TBD
Pages: 1 ... 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 [177] 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198   Go Down
  Print  
Author Topic: Team Go Dog, Go! Modified Partial Streamliners  (Read 521500 times)
saltwheels262 and 2 Guests are viewing this topic.
Koncretekid
Sr. Member
****
Offline Offline

Age: 70
Location: Yarmouth, Nova Scotia & Lafayette, Co.
Posts: 991





Ignore
« Reply #2640 on: November 25, 2016, 03:51:11 PM »

Pete,
The end caps are just freeze plugs.  They're only 2" OD, but the resultant muffler (96 - 1/8" holes in the central 1" diameter tube, with a washer welded inside the 1" tube and wrapped with fiberglass matt) does a nice job on the old Ossa trials bike.
Tom
« Last Edit: November 25, 2016, 03:54:44 PM by Koncretekid » Logged

We get too soon oldt, and too late schmart!
Life's uncertain - eat dessert first!
wobblywalrus
Hero Member
*****
Offline Offline

Age: 64
Location: backwoods Oregon
Posts: 4451





Ignore
« Reply #2641 on: November 26, 2016, 09:25:05 PM »

Those welds look like they will do the job.  Actually, for the typical bikes of the Ossa's vintage they look better than most factory welds.

The folks at Kibblewhite digitized my old #813 cams and they sent me the file.  Then, the digital models were input into the virtual build and the exhaust and intake systems were redesigned.  The digital models give more accurate results than the ten point cam lobe descriptions.  This made quite a bit of difference.  The power curve is much flatter when I use the actual profiles, for some unknown reason.

Vizard, on Page 107 in "How to Build Horsepower" has a graph for lobe centerline angle.  It looks like 116.8 degrees is about right for an engine with 11.68 CI per inch valve diameter.  Then, 0.75 more spread is added to account for the 11.5 to 1 compression ratio.  It is 1 point higher than the 10.5 to 1 ratio for the graph.  Now the spread is 116.8 + 0.75 = 117.5 degrees.

He says "To compensate for the valve angle, practice indicates anything from zero to 2 degrees, depending on the port downdraft.  The Triumph head has angled valves and well shaped side draft ports.  I was thinking of adding a degree to the LCA to account for this, but Vizard does not say to add or subtract.  Does anyone have recommendations?       
Logged
Koncretekid
Sr. Member
****
Offline Offline

Age: 70
Location: Yarmouth, Nova Scotia & Lafayette, Co.
Posts: 991





Ignore
« Reply #2642 on: November 28, 2016, 08:28:40 AM »

Valve timing is still a mystery to me in spite of reading several articles.  This year I tried (my basic premise seems to be trial and error) increasing the duration and overlap by using tappets with a larger radius than recommended by Megacycle.  The effect was about 20 more duration and maybe 15 increase in overlap.  No dyno time but the results at Loring and Bonneville produced almost exactly the same speed as last year (actually down .5mph at both events), so no help there.

In your post of November 18 you stated the following: "The engine simulation program has an option to optimize lobe center angles using a simplified emptying-filling model.  The exhaust system types are input and the model assumes they are optimized.  It gives realistic lobe center angles for the application.  Usually this is 100 to 105 for the intake and 110 to 115 for the exhaust."  That would indicate a lobe separation angle of 105 to 110 by my calculations.  Now you're talking lobe separation angles as high as 117.5.  If you set these cams to give 105 lobe center on the intake (early opening and closing), you're going to get very early closing of the exhaust valve (130 exhaust lobe center, if I've done the math correctly, which seems to indicate valve closing at TDC with 260 duration.)

You won't get much (any) overlap at TDC on the exhaust stroke.  Am I doing the math correctly?

Valve timing-ly challenged Tom
Logged

We get too soon oldt, and too late schmart!
Life's uncertain - eat dessert first!
wobblywalrus
Hero Member
*****
Offline Offline

Age: 64
Location: backwoods Oregon
Posts: 4451





Ignore
« Reply #2643 on: November 28, 2016, 11:59:18 PM »

Tom, maybe this will help.  The engine is two 500 cc singles.  This is virtual power.

The lobe centerline angle based on Vizard's predictions is 118.5.  This gives 3.5 degrees overlap at .050.  Not enough and the virtual power was bad.  This is what works best and the engine is not especially sensitive to LCL.  The cams are all-purpose grinds and numbers are based on digitized profiles and include the effects of tappet clearances.  Intake and exhaust lift is around 0.380, intake duration at .050 = 241.0, exhaust duration at .050 = 239.8, intake centerline = 104.5, exhaust centerline = 113.5, lobe center angle = 109, overlap at .050 = 22.5.  Power is 90 at the back wheel, at 8,000 through 8,500 rpm, at 6,025 feet altitude, 72.1 degrees F, and 34% humidity.  All figuring was done for this environment.

Headers from valve seats to merge points are 28.7 inches long.  Taper is in steps from 1.335 to 1.750 inside diameter.  HP was more sensitive to taper than most anything else.  Collector is 2.5 inches in dia and 15.8 inches long.  The 2.5 inch dia is the biggest I can get from Burns Stainless and a 3-inch woulda worked better.  Megga is 20 inches long with 4 inch end dia.  Intake length from valve seat to first major expansion is 10.7 inches.
   
These are the backup cams in case I have problems with the bigger ones.  Flywheel power at SAE conditions is 115.  The engine can make up to 117 HP using the Harold Bettes/Superflow equation based on 225 cfm intake flow at 28 inches.  These moderate cams do a good job.
Logged
Koncretekid
Sr. Member
****
Offline Offline

Age: 70
Location: Yarmouth, Nova Scotia & Lafayette, Co.
Posts: 991





Ignore
« Reply #2644 on: November 29, 2016, 07:52:20 AM »



The lobe centerline angle based on Vizard's predictions is 118.5.  This gives 3.5 degrees overlap at .050.  Not enough and the virtual power was bad. 

This is exactly the point I was trying to make.
Logged

We get too soon oldt, and too late schmart!
Life's uncertain - eat dessert first!
wobblywalrus
Hero Member
*****
Offline Offline

Age: 64
Location: backwoods Oregon
Posts: 4451





Ignore
« Reply #2645 on: November 29, 2016, 08:48:24 PM »

The notes for the engine setup in the post-before-last are looked.  The cam centerlines were optimized for peak power at various rpm.  The lobe centerline angle (LCA) stayed about the same.  Both cams were retarded nearly the same amount to move peak power up in rpm and the same procedure in reverse to move the peak power rpm down.

The LCA on a BSA cam is fixed at grinding.  An offset woodruff key on the cam drive or driven gear might be used to change the can timing.  The offset is for 1/3 of the tooth pitch.  For example, shifting the the cam gear 1 tooth pitch moves it 15 degrees.  Using the offset key it can be advanced 5 degrees with no shift in gear teeth.  Shifting the gear teeth one tooth and reversing the offset key gives 10 degrees advance.  Shifting the teeth one tooth and using a straight woody key gives 15 degrees advance.  The opposite would be used to retard the cam.  Use of the engine simulation package with the cams at various settings should give one that works good.  Another way is to leave the key out, optimize cam timing, and to tack weld the gear on to the shaft in a location where it can be easily ground off when the motor needs to be taken apart.
Logged
wobblywalrus
Hero Member
*****
Offline Offline

Age: 64
Location: backwoods Oregon
Posts: 4451





Ignore
« Reply #2646 on: December 02, 2016, 01:14:06 AM »

Various cam lobe treatments were discussed with the cam grinder.  She saw some used cams that had super finish treatment that looked like new.  She did not remember many details beyond that.  This is one of a couple of super finishing sites on the i-net.  Does anyone have experience with this?  www.cryogenicssuperfinish.com
Logged
RansomT
Sr. Member
****
Offline Offline

Age: 58
Location: Georgetown, KY
Posts: 475





Ignore
« Reply #2647 on: December 02, 2016, 06:41:28 AM »

O.K. for what it is worth..taking a page out of the MotoGP world from a few years back; I've recently started setting cam timing based on PTV clearance and not worrying that much about numbers.  On my all motor engines, setting the quench area really tight (say ~0.030) produces more TQ/HP (power over the curve high up in rpm) than getting the cam "numbers" just right.  I just make sure that the PTV clearance is at least 0.045 on the intake and 0.075 on the exhaust PLUS I check the valve to valve clearances (but that is another subject all together).  Of course the rest of the build must be able to handle those tight clearances..
Logged
ggl205
Sr. Member
****
Offline Offline

Age: 70
Location: Wichita, Kansas
Posts: 356





Ignore
« Reply #2648 on: December 02, 2016, 10:07:36 AM »

Various cam lobe treatments were discussed with the cam grinder.  She saw some used cams that had super finish treatment that looked like new.  She did not remember many details beyond that.  This is one of a couple of super finishing sites on the i-net.  Does anyone have experience with this?  www.cryogenicssuperfinish.com

Yes.
Logged
wobblywalrus
Hero Member
*****
Offline Offline

Age: 64
Location: backwoods Oregon
Posts: 4451





Ignore
« Reply #2649 on: December 03, 2016, 10:05:34 PM »

Us old Triumph tuners know all about that procedure.  It is the "dial in as much overlap as you can without bending the valves" method.  It does work.  Unfortunately, on this bike the pockets in the pistons are made for much bigger valves and higher lift cams than I am using.  The lobe center angles can be tightened up to some pretty insane values before the clearances get to those specified or there is valve to valve contact.

The program looks at cam profiles, flow test results, intake and exhaust characteristics, and a bunch of other things in the "HybridSim" modeling method.  It optimizes the lobe center angles.  A single analysis takes about 45 minutes.  The results are typically big horsepower over the rpm range when the exhaust system sonics are optimal.  Outside of that range the power is limp.  Right now I am looking at different collector/megga combinations to widen that peak power period rpm.  There is a lot of overlap when I do this procedure.  Much more than Vizard or anyone else recommends.
Logged
wobblywalrus
Hero Member
*****
Offline Offline

Age: 64
Location: backwoods Oregon
Posts: 4451





Ignore
« Reply #2650 on: December 05, 2016, 08:46:36 PM »

This is how the cams were selected.  The grubby worksheet shows flywheel HP at Bonneville environmental conditions.  The numbers in squares are the highest readings in the columns.  All are for the #813 all purpose cams I used for years and they are in good shape.  I just need to realign the cam gears.  The hybrid-sim method was used and it considers the wave action from the exhaust system and measured cam lobe profiles.  Power was optimized for the range of 7,500 to 8,500 rpm

Option 1 are the cams with the slipped cam gears. Intake centerline is 97, exhaust centerline is 89, and lobe centerline angle is 93.  The piston to valve clearances are pretty tight.  HP is not exceptional.

Option 2 are the cams at settings Web Cam recommended a few weeks ago.  Intake centerline is 107, exhaust centerline is 103, and LCA is 105.  Good results with intake lengths of 10.2 and 10.7 inches.  Up to 91 HP.  Primo.

Option 3 are the cam settings on the cam card, 107 intake, 105 exhaust, and LCA = 106.  Good results with the same intake lengths as in Option 2.

Option 4 uses settings recommended by the Hybrid-Sim method using 10-point cam data.  Intake is 96, exhaust is 112, and LCA is 104.  Good low end and midrange power.  Not good settings for land speed hi jinx.

Option 5 uses settings recommended by the filling-emptying method using 10-point data.  104.5 intake, 115 exhaust, and 109.8 LCA.  Not exceptional.

Option 6 uses Vizard's recommendations assuming they are for 0.050 lift.  Intake is 116.8, exhaust is 116.8, and of course, LCA is 116.8.  Nothing exceptional here.

Conclusion.  None yet.  The full race cams are digitized and they will be looked at next.         


* 2017 Build 058.png (303.61 KB, 800x566 - viewed 40 times.)
Logged
wobblywalrus
Hero Member
*****
Offline Offline

Age: 64
Location: backwoods Oregon
Posts: 4451





Ignore
« Reply #2651 on: December 05, 2016, 11:12:24 PM »

These are the other two cams I had digitized.  They are all-out racing cams designed for land speed.  All eight tappet bucket bores need to be enlarged with all new buckets.  Also, the ignition modules need to be reprogrammed for a 10,000 rpm rev limit instead of the 9,000 rpm setting they have.  The fellow that digitized these said the valve seating speed for the intakes is very high and I will have durability issues.  Analysis settings are just like with the #813 cams posted previously.

Combo 1 uses the cam card timing which is 112 for intake cam centerline, 111 for exhaust cam centerline, with an LCA of 111.5.  These cams only help on the very top end with power at this setting.  Otherwise the lowly 813 cams are better.

Combo 2 uses settings determined by Hybrid-Sim optimization.  Intake is 92.6, exhaust is 109.2, and LCA is 100.9.  These settings make about 5 HP more than the 813 cams within an RPM band I can easily use.  There might be clearance problems between the valves and pistons with this tight LCA.  These cams almost certainly might benefit from the tuning method described previously by RansomT. 

Combo 3 used center angles optimized by the filling-emptying method in the computer program.  Intake is 104.2, exhaust is 112.4, and LCA is 108.8.  This is not an optimal timing for these cams.  HP is not as great as it is for Combo 3.

Combo 4 uses Vizard's 116.8 intake and exhaust angles and LCA.  No exceptional power.  Maybe VIzard's durations are for a valve lift less than 0.050? 

Conclusion.  Somewhere between these two cams is a better lobe profile that will give more HP than the 813 cams and use the tappet bucket size I have with good long term valve train durability.

 


* 2017 Build 059.png (459.43 KB, 1018x720 - viewed 41 times.)
Logged
wobblywalrus
Hero Member
*****
Offline Offline

Age: 64
Location: backwoods Oregon
Posts: 4451





Ignore
« Reply #2652 on: December 05, 2016, 11:14:52 PM »

There is a paper by Dema Elgin titled "Performance Camshafts" http://www.elgincams.com/campaper.html.
Logged
wobblywalrus
Hero Member
*****
Offline Offline

Age: 64
Location: backwoods Oregon
Posts: 4451





Ignore
« Reply #2653 on: December 05, 2016, 11:48:58 PM »

Good, the link works.  That does not always happen for me.  The paper describes lots of the issues I am facing when I look at the squiggly lines on the computer screen.  Webcam makes all sorts of profiles and they sent me a bunch of timing and lift data.  Unfortunately they do not have digital profiles.  I need to have the cams made and get them profiled.  There is quite a bit of information on rod ratios and valve sizes on Page 3.  It is also mentioned that "Exhaust flow in 4-valve engines is very high, somewhere in the 80% to 90% region."  The Dynomation program also said this would help.  The head is down at Kibblewhite now getting larger exhaust valves put in with port work.  It should give exhaust flow figures more closely matching the intakes after this is done.

Cam timing is discussed in some detail.  He says on Page 10 "Make the overlap period as short as will complete the job of scavenging."  This is something I will do after I get the new cams with their digital profiles.  That is, widen the LCA until the power drops and use the widest that makes good HP.  Dema explains on Pages 11 and 12 why moderate cams work best in these engines.  The new cams will have 258 degrees duration at 0.050.  This is within the range he mentions.  He also says "In a four valve engine the intake and exhaust cams can use the same duration until the intake cam gets into the 270 to 280 degree duration range.  So, two identical cam profiles were ordered being the same on the intake as the exhaust.

Somewhere in the article Dema says to make the cam lift as high as the valve lift where the head flows well.  The intake flow peters out after 0.40 lift.  A #208 grind is specified for the new cams.  It has .408 lift and after subtracting tappet clearance this should be right at .40.   Plus, it should work with the existing 28mm tappet bucket diameter.  This is as tall as the cams can be for these buckets.

That is the reasoning behind the cam selection.  Adjustable gears will be put on them so the timing is not an issue now.   

 

   
Logged
Stainless1
Administrator
Hero Member
***
Offline Offline

Age: 66
Location: Wichita KS
Posts: 6457


Robert W. P. "Stainless" Steele Wichita, Kansas



« Reply #2654 on: December 06, 2016, 01:48:36 PM »

Interesting info from Elgin
Thanks for the link
Logged

Stainless
Red Hat 228.039, 2001, 65ci, MSA Bockscar Lakester with a little N20 
MSA Bockscar Lakester #1000 my fastest mile 245 and change, 84 ci turbobusa motor... but Corey's 233 MPH H/BFL record is still 3MPH faster than mine.
Pages: 1 ... 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 [177] 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Simple Audio Video Embedder
Valid XHTML 1.0! Valid CSS!


Google visited last this page October 13, 2017, 09:08:37 AM