| Author | Message | ||
     Jadow |
O.k. I have seen it talked about on this forum and other Buell sites....But has anyone besides the company actually Destoked a Buell motor? I have been thinking about motor work this fall. I have AAC 3.750 cylinders with stock 2K flywheels. Been thinking how I like the Firebolt motor but my M2 looks sweet. so how about Firebolt Flywheels with Millinium 3.8125 cylinder/pistons. Come on you closet destrokers....come out and confess! | ||
| Smokin84 |
Jadow, The Firebolt crank has a different motor sprocket shaft bearing, and will not work in the "old" motor. | ||
| Jadow |
True smokey! But I have it on good authority, It just might be possible to pull the race sleave off the spricket side and turn down the shaft to accept a timken bearing set up! You know anything is possible with hard work and alot of money! HA! | ||
| Benm2 |
I was curious about this myself. Here's what I've got: If you assume that the issue at high rpm's with a Buell is crankpin / piston speed related (rather than valve float), then you look to de-stroking to allow higher piston speeds. The problem with destroking is that it also cuts displacement (like the 'bolt). Anyhow, I dragged out some old textbooks and found the equations for piston acceleration. I wrote an excel spreadsheet, and made myself some pretty charts. I started off with stock bore/stroke, and assumed that engine life at 7000 rpm was "acceptable". I then ran other charts for piston acceleration at different strokes, and compared peak piston accelerations (which, from my charts, appear to take place around TDC). With the CHARTS peak accelerations matching (by eye), I re-ran the chart at different RPM's until the peak acceleration matched the STOCK specs. I've got: Stock Stroke: 3.812" rod length 6.94" = 7000rpm Stroke 1: 3.125" rod length 7.25" = 7950 rpm Stroke 2: 2.812" rod length 7.5" = 8400 rpm However, note again that each of the stroke cuts reduces displacement. Stroke 1 was 984cc's, stroke 2 was 885 cc's. Note also that I added rod length because I wanted to keep the piston higher in an attempt to get bigger bores (pistons hitting each other at the bottom of the stroke when the bores get big). So, with another glass of wine at the kitchen table, I calculated out what each engine would pump (displ x rpm), figuring that engines are just air pumps anyway. I didn't divide by two to account for it being a four-stroke, but it was for comparative purposes anyway. Anyhow, I got: Stock @ 7000 = 297 cfm 3.125 @ 7950 = 276 cfm 2.812 @ 8400 = 263 cfm That ROUGHLY would mean that a bolt would have slightly less power than a 1200 IN THE SAME STATE OF TUNE. So, again with more wine, I started adding bore into the mix, to see where I could get the cfm. Anyway, I used even values in an attempt to get the engines back to 1200cc's or so: stock = 297cfm 3.125 stroke x 3.8125 bore = 1169cc's = 328cfm 2.812 stroke x 4" bore = 1158cc's = 344cfm I don't think 4" bore is possible, but who knows. If you assume (funny word, that is) a linear relationship between "pump flow" and horsepower, then a 'bolt with 3.8" bore pistons could make 10% more power than a stock-stroke 1200. That could mean an "XB12R" (in a "similar" state of tune) could make about 85 hp at the tire. Intuitively, I think I'm leaving something out here, but I'm not sure what it is. I'd expect a greater power increase based on a 200cc increase in displacement. The bolt has about 77hp/liter, so 200 cc's should add about 20% of that, or 15.4hp, which would make a bad-a** buell (92.4hp). Given that stock buell's can be made to put out 95-100 rwhp in road-racable trim fairly easily & reliably, a similar short-stroke 1200 could make 10-20hp more with similar reliability, but a wider powerband. Such an engine would be MUCH fun in a firebolt on a track full of R6's. JMHO, Ben | ||
| Darthane |
My head hurts now. | ||
| Smokin84 |
Jadow, I like your way of thinking. I too, like to do thinks that "aren't possible".  | ||
| Blake |
Ben, The stock T-Storm'd 1200's with lightning cams are putting down 85 rwhp. So 10% more would give 93 rwhp. Cool stuff to ponder. Unfortunately, I believe the stock valvetrain is indeed the limiting factor, not piston speed, and definitely not piston acceleration. | ||
| Benm2 |
Blake: Aren't piston speed "limits" a loose relational way to correlate piston accelerations? After all, velocity doesn't break things, acceleration (and subsequent forces) does. Anyhow, the fun of such a project would be to get the pushrod valvetrain to be "comfortable" at 8400rpm. I've looked around a little, but no one seems to make titanium lifters. Possibilities there, maybe. I also thought that high-ratio rockers could be used in conjunction with SHORTER lift cams to reduce the net motion of the lifter/rod combination. Titanium rockers are another possibility. But, titanium stuff is expensive, and its difficult to machine. Shorter pushrods are a possiblity, but they don't weigh that much anyway (especially in the middle). A goofy idea would be to "desmo-ize" the lifter & pushrod, setting up the cams with a "negative" to pull the lifter & rod back down. I can't think of a way to do it with hydraulic lifters, as pulling on them wouldn't work, so solids it would be. That would leave the valve springs to return the valve & rocker only, much less of a mass. Regardless, though, bolts have a valvetrain setup that seems to be comfortable at higher rpm as a streetbike. Race-worthy rpm is probably a bit higher. Thus, with bolt valve springs & titanium valves, the engine should be comfy at the 7950 rpm that corresponds to the 3.125" stroke. Since 3.8" bore pistons are regularly fit to Buell's, combining those cylinders with a de-stroked crank should be possible (assuming you can the rod length right) | ||
| Blake |
No, piston speed is highest when acceleration is lowest. And as I understand it, the concern with piston speed is not the forces imparted to rod, wrist pin, or piston, but rather the dynamic stability and integrity of the piston to rings to cylinder seal. Also the ability to feed the combustion chamber. Pushrods can work fine even at 9K rpm. NASCAR Winston cup engines handle such revs quite well. In lifters, stiffness is as important, if not more so, than strength and weight. Ti is just a bit over half as stiff as steel. Besides, I understand that the governing inertia affecting valve float is on the spring side of the assembly (valves and springs/retainers). Therefore the higher ratio rockers would not be helpful either. These engine guys know their stuff. Ti rockers... same issue, stiffness is very important. If the valvetrain gets too rubbery really bad resonances (vibrations) can result and cause the entire valvetrain to fall apart. You can certainly pull on a hydraulic piston just as well as you can push on it. I'm thinking someone clever could invent a push/pull-rod desmo valvetrain. It would be fun to try. Getting rid of the spring(s) and retainers would help a lot. Again, stiffness is vital, so Ti rockers may or may not be of benefit. How bout metal matrix composites? The pro-thunder bikes are reportedly running metal matrix connecting rods... hmmmm I agree that big springs and lightweight Ti valves would aid high rev capability. Look at Aaron running mongo lift radical grind cams hard into 7,500 rpm in a stock displacement/bore M2. You can definitely get more revs out of these engines. | ||
| Jadow |
Wow! Keep it coming guys! The thought process is flowing and I can see the XBM2 project coming together...I have heard the Buell race team has used Metal Matrix composite pushrods....infact the XB parts book list the XB pushrods are different from earlier Buells. I really beleive the Buell techs Have done their homework. Destroked flywheels with Odd length rods so stock length cyl. can be used. Question? Is there still an advantage to this detroke combo if youdon't rev to the stratisphere? I love the Desmo Idea! | ||
| Benm2 |
I think I've been misunderstood. I stand by the statement that the piston speed limit is based on the acceleration of the piston & the loads it produces. The average piston speed formula used (the 5000 fpm one) is just a way to relate stroke length to rpm. Once a piston in moving & the the rings are on the oil film & sliding, things are fine. Piston seal issues at higher rpm are related to the acceleration away from TDC, when the rod is pulling the piston down. If the acceleration is high enough, the ring gets pulled against the top of the ring groove, and subsequent combustion pressure "blows by" the ring before it can push the ring back down. But I don't think that peak velocity of the piston has alot to do with things. Also, while titanium is not as stiff as steel, the amount of flex that a lifter undergoes is based on a straight compressive load. In addition, stiffness goes up quickly with diameter. The ID of a Ti lifter could be adjusted to match stiffness. Same holds true for the rocker. Taller beam sections & a larger OD on the "torque tube" portion of the rocker could make a Ti lifter at least as stiff as the stocker. Without verifying against dusty old textbooks ( it is 6am on saturday, afterall) I seem to remember stiffness being related to the fourth power of the radius. I think the reason no one has made a Ti rocker is simply market related. The development cost would be high, and the finished product would be expensive. Is there really enough of a market to warrant the development of it? Also, I can't beleive that the rocker, lifter & rod don't affect allowable rpm. In any given spring/mass system, mass is mass. The spring's got to push it all home, not just the valve, spring & keeper. NASCAR engines are slightly oversquare, aren't they? Since the piston "speed limit" is higher, they've found ways to match the valvetrain to it. V8 rockers are very small in comparison to HD rockers (and their aluminum, less stiff than Ti) and the pushrods are shorter since the cam is in the V. These are the same sorts of tricks that V&H are using on their top fuel dragbike (you know, the one that killed the VR program). Nonetheless, I think we agree that all that fanciness might not be required anyway. Since the bolt-stroke engine's limiting rpm isn't that high, standard stuff like ti valves & stiffer springs will probably work just fine. Which leaves the work down on the bottom end. So, back to the original question: How do you make the crank, and how long do the rods need to be? I would guess the options are: (1) weld up the stock holes for the crankpins & re-bore them inward or (2) see if S&S or someone like that will make them for you. Rods, though, might be full custom, unless Bolt rods would work. I really like the idea. 3.8" bore x 3.125" stroke in any old tuber racebike would allow it to possibly run with current 600's. Agreeing with all that souped up buells sound better than anything else on the track, I'd LOVE to see one pulling R6's & CBR's on the straights. The prothunder bikes have alot of this tech in place already. If those bikes produced "over 130 hp" with 1350 ccs, a short-stroked M2 or X1 or S1 (and so on) could make competitive power. Now thats some racing I'd like to see. | ||
| Jadow |
Vibration? would Destroking decrease?Increase? ornot effect it at all. The mounting/rubber system on my 2K M2 is alot different from a XB. It would be terrible to build a Destroked motor and find MORE Vibes everywhere! Both motors are V-45s so primary imbalance should be the same in both motors..Right? Or could this be the answer to a smoother, less fatiguing ride. Comments? | ||
| Benm2 |
Should vibrate less. | ||
| Blake |
Ben, What you said about the rings not sealing is interesting. You can use titanium and/or aluminum alloys for pushrods and rockers. The lightweight alloys just don't gain you much if stiffness is a primary concern. If an alloy weighs half as much but it takes twice as much to achieve the required stiffness... With regards to stiffness, the type of loading, whether bending, torsion, compression, tension, or shear, is inconsequential. As far as the axially loaded pushrods are concerned, axial stiffness is dependent only the cross sectional area, the length, and the elastic modulus (material property) of the tubes. Diameter of the pushrod tubes is very important however, to prevent buckling. The optimum material might be a metal matrix composite or possibly berylium copper. Effects of the pushrod side of valvetrain on valve float are not zero. They are simply second order when compared to the valve/retainer/spring assembly. Valve float is governed by inertial forces where energy is proportional to mass and its velocity squared. So if the pushrod side through the rocker arm ratio is moving slower than the valve side you can see how it could become less of an influence on the valve float issue even if its mass is greater. There is a speed at which the cylinder to piston-ring oil film breaks down. Oil can only handle so much shear stress. Shear in a viscous fluid is proportional directly to the speed of the lubricated surfaces relative to each other, the viscosity of the lubricant, and inversely to the thickness of the oil film. So to achieve higher allowable piston speeds we would need either a thicker film or a less viscous lubricant. Just some things to ponder.  | ||
| Benm2 |
Oh, this is just too much fun! My wife’s eyes glaze over when I start talking about motorcycles, much less piston speed and valve float. My work is likewise devoid of true motorheads. Plenty of harley guys, but not so much interest in the Fundamentals of Combustion Engines. So, if you’ll all forgive me, I’ll continue with the discussion. The “average” piston speed at 7000 rpm (two-stroke lengths per crank revolution ) for the stock motor is 4447 fpm. The 3.12 stroker at 7950 is 4140 fpm. The other one is too short to give an achievable displacement. Thus, the shorter stroke motor has a lower average piston speed. (maybe I’ll chart out the actual piston velocities later, just not now) so it should work nicely. Column buckling loads are related to the inverse square of length, and directly to the moment of inertia, which is a function of the fourth power of the diameter. Bigger diameters gain resistance faster than shortening. Also directly related to material stiffness. So, MMC stuff of same diameter would have a higher critical load if it’s a stiffer material. But, stiffness is a direct correlation, the others are power functions so gains can be made faster there. The paragraph about the reduced motion of the pushrod end of things having less of an effect on valve float was precisely why I suggested using longer-ratio rockers. Since that portion of the valvetrain travels some fixed distance in a fixed amount of time (at a given rpm), reducing the distance it travels also reduces the amount of energy required to get it there & back. So, it makes me wonder why the XB9 has such high-lift cams instead of high-ratio roller tip rockers. Another thing to think about is what the natural frequency is of the valvetrain, as that can cause float as well. In a perfect world, I’d have a shop full of engines & an eddy-current dyno. I’d fab up an engine without pistons or rods, and run it up in rpm until something broke. And, I’ve have a lot of titanium & spare time. Alas, no such luck for me. However, I do plan to put lightning cams in my M2, and the next step after that might be high-ratio rockers. If the bank can stand it, I might try titanium inlet valves as well. Curious, though, about the titanium lifter & rocker. If you built it, would they come? We can theorize till we turn purple, but until someone cuts metal, its mostly just conjecture. Fun, though. Overall, for those who might want to attempt such a change, my (maybe) final thoughts: 1. Low lift cams with high-ratio rockers 2. Dual valve springs with damper coil 3. Hydraulic lifters, aluminum pushrods 4. Titanium inlet valves (at least) & stainless steel exhaust 5. Get the rod length such that the closest the piston skirt gets to the crank matches what it would be if it were stock stroke at the same bore (3 13/16). It might actually be a little bit shorter than stock. Manley & Carillo will make any rod you want, in a variety of materials. Just send money. 6. Get the lightest pistons you can find. 7. Have the cylinders cut & machined (top end, my guess) till the pistons are at the top of the bore, like stock. Or, buy a XB9R motor & put big pistons in it. Ben | ||
| Aaron |
I've run high ratio rockers with Lightning cams (Jim's 1.745:1 if memory serves). Hurt the power. The geometry was really bad. The roller would've come right off the edge of the valve tip if not for the lash cap. May have even ended up with less lift. Correcting the geometry is not simple on these things. Not like a Chevy where you can just change pushrod length. Different length valves, different rocker boxes, etc. | ||
| Madduck |
Blake et al, Current generation XR-750's routinely turn between 10,500 & 12,000 rpm when set up for the mile tracks. Paul | ||
| Aaron |
BTW ... the XB9 motor has a longer rod. Cylinder height, wrist pin height, bore, etc are the same, they just made the rod longer by half of how much the stroke was shortened. | ||
| Benm2 |
I wonder why they didn't make shorter cylinders? A shorter rod would have been lighter. XR750's at 12,000 rpm? Really? What sort of things are in THAT valvetrain? Aaron, thanks for the info. That's a forehead smacker, rocker geometry. Hmm, how about a rocker that envelops the roller a little bit, with early lift on a slider tip and high lift on the roller? Still trying to get my head around a 12,000 rpm XR750. I think the piston stroke is around 2.5", so piston speeds would be lower, but how they get the valvetrain to cooperate is an interesting question. Of course, the bike only needs to go a mile... Ben | ||
| Blake |
Ben, "Low lift cams with high ratio rockers"... I'm confused, you are proposing higher ratio rockers to allow higher revs even though the valve float problem is governed by the valve side assy? You are going in the opposite direction needed. You've lowered the dynamic energy of the pushrod side, but done nothing for the valve side. Right?Therefore little to no effect on the valve float threshold. Or am I the one who is confused? I think if you investigate, you will find that the geometry of the cams and rocker arms is pretty much optimized. Like Aaron alluded to, I don't think it's quite as simple as it seems. Too much ratio either way, and the transverse components of motion get out of hand. I guess that problem could be ameliorated via shims and guides and so forth. Hmmm... Pushrods... Yes, increasing diameter is effective in improving resistance to column buckling. But without a commensurate increase in cross sectional area, it does nothing for axial stiffness. I offered the advanced materials as both stiffer, and lighter weight alternatives. I still like the idea of a hydro-desmo system. Sounds like you and I aspire to the same dream life. Madduck/Paul, Are the XR750's running 4-valve heads? Don't those engines get torn down after every race? | ||
| Aaron |
It gives the spring more mechanical advantage on the rocker/pushrod/tappet. I think that's what he's getting at. | ||
| Benm2 |
I'm not convinced that valve float is governed by just the valve/spring/retainer assembly. If it were, no one would have developed overhead cams. In vibration analysis, mass is mass, and its effect on limiting rpm is fixed. I think the limits of DOHC shim/bucket seems to be around 16,000 rpm crankshaft speed (with lots of small, titanium valves), beyond that it looks like desmo or pnuematic closers are required. Four-valve heads have potential for higher rpm too. In the vein of reducing reciprocating mass at the valve end, two intake valves of equivalent opening area have more circumferencial area, so they don't need to open as far to have the same net valve opening area (lift x exposed valve circumference). However, four-valve heads would make intake tracts difficult. Once you start looking at four-valve heads, though, you might as well throw in the towel on the standard design. At that point I'd be tempted to move the cams, and put an SOHC setup in the head, and replace the cam geartrain with a dummy-gear setup & some drive belt pulleys. Dual-finger rocker arms, four-valve heads, threaded lash adjusters. But, then you're competing in another ballpark. I personally like SOHC engines, setup with two-valves per cylinder, please. Still fairly compact head, fewer valves to maintain, etc. It would be (sort of) like making 900SS engines, but I'd bet the SOHC setup would be worth at least a few hp allowed by sharper valve timing. That would be a nice engine, relatively powerful, with characteristics similar to the pushrod motors. Still not competitive with the big leagues, though. Now, some REAL fun would be trying to open up the cylinder angle to 55 or 60 degrees, shortening the stroke ALOT, throwing 4" bore pistons in, custom-made finned, billet cylinders, and grafted-on TL1000 heads with (forgive me) a radiator setup for head cooling (80% of engine heat leaves through the head). At that point, though, you might as well put in a new engine. What kind of bike would a XB9R chassis be with a 500cc version of the Cannondale single? I'd bet the whole bike would be about 30 or 40 lb lighter than a stock XB! Way off topic, sorry. Does cannondale make a v-twin? Ben | ||
| Blake |
Ben, See if you can find a reliable overhead cam, spring returned, two valve head that revs to five digits. I think you will find that the higher rev limits came with four valve heads, smaller/lighter valves.
That thing you said about natural frequency? I think you are on the right track. Titanium valves and retainers anyone? | ||
| Benm2 |
You rotten bastard! I haven't bought that book yet, just perused it in Barnes & Noble. I think NASCAR has alot to offer with regards to squeaking power out of "obselete" technology. Two valve rpm - uh, GPz750? KZ1000J? Some dudes named Rainey & Lawson won some races on those dinosaurs against (a) 4-valve air cooled hondas and (b) 4-valve water cooled hondas. | ||
| Madduck |
BenM2, XR-750 per ama rules is a bore of 3.165 with a stroke of 2.983 inches of course. This is a .040 overbore and does not equal 750cc. The bikes typically run 50 miles at springfield and Duquoin depending more on tire longevity than on motor reliability. I think the last XR-750 was made by H-D in 1983. These things last pretty damn well. | ||
| Blake |
Ben, Buy the book, very interesting stuff, no real technical/engineering meat, but some very interesting commentary. Okay, you've found some two valve heads that can be revved. Now, what are the main differences between that high revving head and our Buell heads? | ||
| Vr1203 |
I believe, the limiting factor governing high RPM is the bottom end. I spun the big end of my XR1000 trying to get the HP's out promised by the cylinder head guru Jerry Branch in the 80's.(oh, is that another discussion?  ) The little bearing rollers cannot spin that fast without sliding on the big end pin. They just slide around and dig in when they lose the oil film.I'm facing the same problem I think with the turbo bike,running it to 7000rpm+ all the time.I did not see any bad valve float but now I think it's bottom is knocking. | ||
| Benm2 |
Blake, The differences? One: smaller valves, as there both 4-cylinder. Two: Bucket/shim valve actuation. Which, regardless of smokey's comments, leads me back to my suspicion that ALL valvetrain mass is important. Which, of course, makes his comments that much more interesting. VR1203: skating bearings? There's got to be a way to overcome that. The overgrown chainsaws running last-generation GP500's all use ball-bearing cranks ('course, they rebuild them every race), but also old Suzuki GS1100's had rolling-element cranks, and spun well past 7000 rpm. Aaron: I'm still wasting valuble time thinking about your comments regarding the rocker geometry with the high-ratio rockers. I'm thinking that ANY really high-lift cam would cause geometry issues, even with stock rockers, but I haven't looked at them (in person) nor have I done any geometry. I'd think a longer arm would reach farther across the stem with the valve closed, allowing for good roller contact at full lift. I'm guessing that as some lift, tip issues become a problem for any rocker ratio. What was your experience? | ||
| Vr1203 |
Ben, I think its a question of roller to crankpin size ratios and stroke.Lets take a short stroke type 4 cylinder and imagine the bigend speeds(fpm) at redline.Now think of a long stroker,the big ends in the Harley would really moving(fpm) at that rpm .They have be bigger to support bigger rods, pistons, etc and the cases and big ends of the rods cannot be too large for everything to fit in the space of a motorcycle. .So they make the rollers pretty small.When the rollers are smaller and the crank pins are bigger(its just like a gear ratio)the smaller rollers have to spin faster to keep up.Also a caged bearing;as in the two strokes cannot crowd. | ||
| Blake |
There you go. So just maybe, four valve or even three valve heads might be the simplest most direct way to prevent valve float? |
