| Author | Message | ||||||||||||||
     Pangalactic |
My bike is back together. YAY!!! I am SOOOO glad I did it my self- the lower gasket around the push rods was never installed correctly- one of the bolts actually punctured the gasket!!!! whoever installed it needs a swift kick in the arse. It's probably been leaking oil and air since I bought the thing, but its been spraying inside the rocker box, leaving the thing a mess. Got it all cleaned up and back together, and with Cometic metal gaskets, no less, and the bike never ran this smooth, or had this torque. I never would have been told what the problem was by Ray Price, and they did just do the 500 mile service, and didn't do anything about this. thanks to Sportyeric, I must now recite a Haiku: Purring like kitten, No more puking rocker box. Summer almost gone. Figures. Good thing the fall riding season lasts forever...  | ||||||||||||||
| Aaron |
Pangalactic ... I have seen that many times. In fact, damn near every time I've fixed a rocker box leak. But every time I've seen it, the gasket hole was actually elongated, not punctured by the bolt. The gasket was all wadded up and distorted around the bolt hole. Which makes me think the gasket is actually moving while the bike is in operation. I'm thinking there's not enough clamp load and the pressure from the pushrods is making the box move around, which lets the gasket walk. Steel gaskets should fix you right up, when they move enough to contact the bolt hole, they'll stop. At least that's my theory! AW | ||||||||||||||
| Blake |
Aaron, I think both modes (poor gasket placement and insufficient torque/action of lifters) of failure are common, and each is attributable to improper installation at the factory. | ||||||||||||||
| Aaron |
So you're saying that if properly installed, they won't fail? I guess I'm skeptical of that. There's just too damn many of them and it's been going on for too long. If it was a matter of installing them better at the factory (placement and/or torquing), they would've fixed it a long time ago. That would be easy to fix. I think the clamp load is underdesigned. But I have no proof, only anectdotal evidence. Personally, I think the best we can do is use a gasket that won't elongate and distort and rip when it moves over and starts hitting the bolt holes. Until they give us a better design. But I could be wrong. AW | ||||||||||||||
| Blake |
Dangit Aaron, you sure are asking for the math aren't ya! And I got stuff I'm supposed to be doing today. And I still haven't made it to Lonestar HD/Buell to see the XB9R. You know I can't resist don't ya! (I'm gonna have the wife call you if I get in trouble tonight for not having all my chores done.)  Here's my take on the rockerbox leaking... Out of three bikes and close to 40K miles I've had one rocker box leak. All miles were run using the stock paper gaskets. The one leak I did experience occurred very early in my '00 M2's life at around 2,500 miles. The root cause of that leak was a poorly installed gasket on the spark plug side at the rear of the aft cylinder (#17 at screw #14 on page 3-11 of manual). The gasket was obviously damaged during installation; it was kinked inwards right at the corner and the gasket's screw hole was torn as though punched out during insertion of the bolt. The imprint in the gasket from the rockerbox clearly showed that only a sliver of the gasket had been clamped between the faying surfaces at the damaged/misinstalled area. Given proper bolt preload (torque) and clean/dry faying surfaces, it seems very unlikely to me that a lack of gasket clamping force would be a factor. The gasket clamp load can be estimated from the torque versus preload formula for a dry (unlubricated) steel bolt as T=0.2PoD. Where "T" is the fastener installation torque, "Po" is the bolt preload (tensile load of fastener), and "D" is simply the nominal diameter of the fastener. If the bolt is lubricated the formula changes to T=0.15PoD. These are not always accurate formulas and it is very dangerous to use simple (uncalibrated wrt specific fastener/threaded connection) torque to arrive at a desired preload, but it is a good rough estimate for initial mechanical/structural design or analysis purposes and for conservatively based analysis. Maximum valve springs (inner and outer) loads (valve fully open) are listed as 112+207=319 LB for intake and 106+195=301 LB for exhaust in the emanual for the S1. Let's be REAL conservative and say that the pushrods are pushing half again (150%) as much (I can't find the ratio between pushrod and valve rocker arm radii). That would put the pushrod loads at 479 LB for intake and 452 LB for exhaust. Let's be REAL conservative again and say that these loads occur simultaneously. So we have 319+301=620 LB on the valve/sparkplug side and 479+452=931 LB on the pushrod side. The pushrod side has three bolts clamping it to the head, two 5/16" bolts and one 1/4" bolt. The specified minimum torques are the 15 FT-LB (180 IN-LB) for the 5/16" bolts and 10 FT-LB (120 IN-LB) for the 1/4" bolt. The valve side has six bolts clamping it to the head, two 5/16" bolts, two 1/4" bolts, and two 1/4" screws (nomentclature per Buell manual). Minimum torques are the same as above except for the 1/4" screws; the minimum screw torque is 90 IN-LB (7.5 FT-LB) Using the torque (T) versus preload (Po) formula above for dry bolts (again a conservative assumption) we can find the minimum estimated preload (Po) for each fastener. Rewriting to solve for Po we get Po=T/(0.2D).
The total minimum cold clamping load on the pushrod-side gasket is thus Po=2*2,880+2,400=8,160 LB. It seems very unlikely, especially given the close proximity of the 5/16" bolts to the rocker arm shaft retainers, that 931 LB of pushrod load would be able to dislodge a gasket being clamped by 8,160 LB. Then consider that in order for the gasket to slide/dislodge, the faying surfaces would need to move relative to each other. With proper torquing and installation (clean dry sufaces are very important) I just can't see any way that a gasket with a contact-adhesive bead on one side would be able to "walk" between the head and rockerbox. In my mind, it's got to be poor adhesion due to improper torquing and/or a wet/dirty faying suface and/or wet/dirty gasket. As far as the solution being simple, well, it's about impossible to inspect the integrity of the gasket installation once it's intalled, so good old human error and negligence will always be a factor, especially everytime a new person comes online in the factory. The only real solution would be rigourous employee training and constant oversight and spot checking at the rockerbox installation station. Plus with so many of us replacing our own gaskets, the factory may not understand the significance of the problem (I know, that one is hard to fathom). Or they may unwittingly be complacent/happy enough to simply let the dealerships continue to handle the problem on a case by case basis. I'm hoping they will solve the problem one way or the other, SOON! The above highly conservative analysis along with the vast experience and opinions of a mechanic who I trust very much still lead me to believe that the problem lies mostly with the quality of the factory installation. Then again, it's just my overblown opinion and the above analysis is extremely simplistic, assuming that proper bolt spacing and geometry versus the loads have been considered. I'd concede that the rockerbox to cyl head joint might benefit from a match drilling and pinning at four locations (two on each side) to ensure gasket placement and to carry any vibration induced shear between rockerbox and cyl head (would prevent relative movement between rockerbox and cyl head even if clamping failed to do so). Okay, hope you at least enjoyed that blathering; I'm outa here. Looking forward to hearing about the Blast dyno runs tonight. Good luck. Blake | ||||||||||||||
| Jasonl |
lol....I bet Blake busts out a calculator while arguing with his wife. "well you see..apprx 33.3% of the time I'm doing the dishes while 15.6% of the time you're sitting on your a$$..." | ||||||||||||||
| Aaron |
Dammit, next time I remove one with an elongated bolt hole in the gasket, I'm gonna take a picture! | ||||||||||||||
| Blake |
I certainly believe you about the elongated holes. I just question the root cause of that problem (bad install versus systemic problem) is all. Plus, how do we know what comes first, the leak, or the walking gasket? Turns out there is no HD/Buell coroporate presence at Lonestar HD/Buell today, just an open house, no XB9R, not even a V-Rod. I'm stayin home and finishing my chores. Aaron, you're off the hook... for now. | ||||||||||||||
| Tripper |
I like Blakes posts, I save time cause I can scroll down. I understand more that way. Here is your PERFECT JOB Blake. | ||||||||||||||
| Aaron |
how do we know what comes first, the leak, or the walking gasket? That's simple ... The bikes leave the factory not leaking, and then after some number of thousand miles, they develop a leak, right? Then you pop off the rocker box, you find a gasket completely out of place, moved so far that it's bolt holes elongated when they ran into the bolts. It's pretty obvious that it walked until it got so far out of place that it started leaking. The only real question in my mind is whether it's improper torque (or some other design aspect that causes the bolts to lose torque over time), or whether it's underdesigned. I lean toward the latter, just because I've removed many leaky rocker boxes that had good tight bolts. Next time I'll measure breakaway torque, I have a beam style wrench around here somewhere. AW (mad as hell because UPS hasn't shown up, and it's raining anyway, and I can't dyno tune in the rain even if the pushrods do show up. grrrr ... think I'll take it out on Blake) | ||||||||||||||
| Ron |
Can any of you fine math majors help me with a numbers problem. I am playing with the idea of going to a set of American Air Cooled cylenders. I also plan to have a Balance Masters job done on the flywheel. This means removing the engine and splitting the cases. Which in turn means I may as well consider a modest bore increase. Perhaps up to 3 3/4" as the price from AmAir is the same up to that size. Here is the problem. How much will a 1/4" bore increase raise my compression and how do I get back down to stock. I am battling a ring seal problem with oil loss out both the breather (not much) and pipes (most of it). By going to a larger bore am I going to just make the pressure in the engine that much more and loose even more oil out the breather? NOTE: I have installed a Spyke Krank Vent and "the Cure", and routed the breather hose out of the air box already. Obviously I'm not going to be able to take best advantage of the displacement increase with the stock cams so what would be a good choice? I would prefer the power increaqse concentrated in low to mid range with current power levels at the top being just fine. This is a 2000 S-3T and I like the fuel economy for long distance. I do use it mostly for touring. Often with a passenger. Perhaps this will give you number crunchers some fun. Or perhaps I should just see if the dealer can continue to keep me off the road by fooling around trying bandaid fixes that don't work for the oil loss problem? Thanks Ron | ||||||||||||||
| Hoser |
Aaron , Blake: Some of us are curious about ,................. The clamp load increasing as the engine reaches operating tempurature , or decreasing as the engine cools. I can not remember how many rocker gasket leaks I have repaired but I can say that half had elongated holes as per Aaron's discussion , the other half had cracked and dried out old gaskets in place. My own two Buell's have been leak free but the gaskets have been done several times on the S1 for other reasons. S1 has 33,967 on it , S2T has 45,898 on it. Blake , you amaze me !! JDH | ||||||||||||||
| Tonyinvabeach |
Anybody, First question. I saw this question earlier but never saw an answer. I'm trying to fix my rear rocker box..it's leaking again. The right rear allen screw is directly beneath the frame. Is there a tool I can get that is short enough to get in there and loosen it. No one else seems to have a problem so there has got to be something I'm missing. I've heard of guys grinding down there wrenches....where could I go to get this done..don't have a grinder. Second question. When unplugging the rear turn signals, the service manual calls them bullet connectors, do you just "yank" them apart or is there a better way. I know these are stupid questions..but I've got the bike apart and I would like to fix this so I can get to work on Monday. Thanks for all the help. If someone out there is willing to let me ping them off line and possibly call, I would appreciate that as well. Thanks again. Tony | ||||||||||||||
| Hoser |
Tony: That screw , ......... use a short piece of 3/16" hex key combined with a 3/16" combination wrench. You will have to sacrifice a 3/16" allen key. A six point combo wrench is best ( stronger ) The same technique applies to the intake manifold bolts on the left side but with 1/4" hex sizes being used. The wires , .......... use needle nose pliers , dont jerk the wires apart like Homer Simpson might or you will have to do a repair. JDH | ||||||||||||||
| Ron |
Tony, Find someone with a grinder or belt sander and simply reduce a plain ole Allan wrench down to about 1/4" - just enough for the wrench to go all the way into the cap bolt. It's a pain. You can buy a tool for this. It's just a flat bar with about a 1/4" chunk of a 3/16" Allan wrench on it. But I can't recall where I saw it. The bullet connectors just pull apart. Use a little silicone spray when you put them back together and pull on the connector itself. Try to avoid pulling on just the wire. Hope this helps some. Ron | ||||||||||||||
| Bluzm2 |
Tony, Welcome to the "gasket club". Everyone I know just grinds down a standard allen wrench. If you dont't have a grinder, stop into your local full service station and ask nicely. They may do it for nothing or very little. Tell them you will treat them to their favorite six pack beverage. A decent hardware store should also be able to do it for cheap. On my M2, it needed to be quite short. Just enough of the short leg left to fully seat in the bolt. This provides the necessary frame clearance. The turn signals just unplug. A firm straight pull on the wire should do the job. The bullet ends should be soldered so they are pretty tough. If not soldered, they have a decent crimp. Good luck! Brad | ||||||||||||||
| Bluzm2 |
Ahww Ron, Ya beat me to the punch! Good call on the silicon spray. I usually use dielectric grease, but the silicon should work also. BJ | ||||||||||||||
| Tonyinvabeach |
Hoser, Thanks...copy NOT a screw..heh heh. Thanks a bunch...I've got to start somewhere!! This combined with Bill's post clears it up for me...now to figure out how to cut an allen wrench in two! Thanks! Tony | ||||||||||||||
| Aaron |
Hoser, man, I wish you'd participate more often. I really appreciate your insights, you see a helluva lot more stuff than most of us. Ron, that's a question you need to ask of the piston manufacturer, they ought to be able to tell you what compression to expect. If you just made the bore bigger and kept the same dome, the answer would be roughly 11.3:1. But that's not a likely scenario, the maker of the piston is who you should talk to. BTW, the answer to your problem IMO is better ring seal. AW | ||||||||||||||
| Ron |
AW, Just emailed Nallin with about the same question. Looks like he may have a good setup for my application. One thing, perhaps I've been up too long - but please, what is IMO? Thanks Ron | ||||||||||||||
| Mikej |
IMO = In My Opinion IMHO = In My Humble Opinion, or Honest, or any other personal choice of words. YMMV your mileage may vary | ||||||||||||||
| Ron |
Mike, Shoulda known. Thanks. I agree on the ring seal. Just about every other symtom has been given attention. All that's left is the ring seal. After much reading I believe that blowby past the rings can also cause increased crank case pressure which will have more tendancy to push return oil up the push rod tubes and back out the breather. So no matter what I do, until this situation is corrected, the oil on the inside is going to find its way outside one way or the other. Ron | ||||||||||||||
| Blake |
Hoseman: I started to say something about that; note my comment refering to "... minimum cold clamping load..." but when you look at the joint and heat loading the problem becomes very complicated. The coefficient of thermal expansion for steel is half of that for aluminum, but the steel bolts thread directly into the head while the rockerbox is insulated from direct heat conduction from the head by the gaskets. So it is possible that the bolts may settle at a significantly higher average temperature than the rockerbox. If so the bolt preload may not be significantly affected. I can't imagine that the bolts would see double (or more) the average temperature increase as the rockerbox so I would expect the preload to increase during operation. If you assume the best case, that the bolts and rockerbox see the same average temperature increase you could use the following rough formula to determine a conservative worst case increase in bolt preload: Po DT=[(aAL-aST)*tAL*DT]*[ABOLT*EBOLT/LGRIP] Where... DT is the difference (delta) between the hot operating temperature of the rockerbox (~250oF?) and room temp (~70oF). Ex: DT=250-70=180oF. aAL is the coefficient of thermal expansion for aluminum (~13E-06 IN/IN/oF). aST is the coefficient of thermal expansion for steel (~6.5E-06 IN/IN/oF). tAL is the local thickness of the bolted part (the rockerbox) LGRIP is the effective grip length of the bolt (for our case this is the same as the local joint thickness of the rockerbox) EBOLT is the modulus of elasticity for the bolt (~30E6 PSI) ABOLT is the effective cross sectional area of the bolt (obtained from standard tables) Nerdy note: The above assumes that any additional compression of the aluminum rockerbox locally due to the thermally increased bolt preload is insignificant relative to the thermal expansion of the aluminum. This is only true if the stiffness of the effective volume of aluminum resisting the bolt preload is much greater than the stiffness of the bolt. Though aluminum has roughly 1/3 the stiffness of steel, the effective cross sectional area of the clamped part will for grip lengths greater than two times the bolt diameter usually be sufficient to overcome the 1/3 disadvantage of a clamped aluminum part versus a steel bolt and so we can probably safely use the above formula. The formula above is simply multiplying the delta elongation of the bolt (resulting from the thermally induced increase in thickness of the rockerbox) times it's stiffness (AE/LGRIP). Thermal expansion is simply aLDT}. The longest bolts are 5/16" dia. A rough calc gives us... Po DT=(aAL-aST)*tAL*DT*ABOLT*EBOLT/LGRIP Po DT=[(13E-06 - 6.5E-06 IN/IN/oF) * 1.0 IN * 180oF] * [0.06 IN2 * 29E6 PSI / 1.0 IN] Po DT= [0.00117 IN] * [1,740,000 LB/IN] Po DT= 2,036 LB Resulting worst case tensile stress on the bolt (due to max torque and conservative worst case thermal effects) s= (3,456+2,036 LB)/0.06 IN2 = 91,500 PSI which is well below the allowable tensile stress for a grade 8 bolt. Disclaimer: The above is extremely rough and speculative and is intended to confuse and confound, especially Tripper! Now I'm sick of this crap not really. Hope you're happy!  Blake | ||||||||||||||
| Sportyeric |
Engineering god And writes a damn fine haiku. We are so lucky. | ||||||||||||||
| Road_Thing |
Blake: Dude, get some sleep... r_t | ||||||||||||||
| Hoser |
Aaron........ Thanks , my participation is based on time available , quite often I can only take a quick look . Blake........I'm happy , .........confused and confounded but happy . Hoseman' | ||||||||||||||
| Jasonl |
Tat has a rocker cover tool to get to those allen head bolts. | ||||||||||||||
| Bigblock |
Hi guys, IMIE ( in my incompetent experiments) I have found that the rocker boxes seem to seal a lot better and last a lot longer if you are VERY CAREFUL when initially tightening the box. What I mean by this is... long before the box has contacted the gasket or top of head, it is getting an uneven preload from the pushrods so instead of tightening the screws by torque "feel" at this point, [which will cock the rocker box unevenly, and bring one side down to the gasket first, Thus possibly grabbing that side of the gasket and pushing it just enough out of line so that it initially seals, but isn't in the proper place, starting a nasty chain reaction leading to...(GASP) LEAKS!] you should actually watch the gap betwixt the head and box, making sure that it is coming down evenly onto the gasket as you are loading the valvetrain. Once you have EVEN and FULL contact between them, then you can proceed to go to the even torquing required. Now, alot of you guys have much bigger brains than I, as evidenced by a full grasp of math and dynos and all sorts of things I have absolutely no experience with whatsoever, so you all are probably taking this into consideration already. But, perhaps there is someone out there who has made the same mistake as I (woops) and Perhaps you may benefit from this overly long winded but hopefully understandable discourse ;-) Ray | ||||||||||||||
| Bigblock |
Hmm, methinks I be overstating the obvious? LOL! | ||||||||||||||
| Sportyeric |
How very timely. Another rocker box leak! Shoddy workmanship? OK. I'll stop doing that now. My dealer suggests the problem may be a result of constricted breather. Mine tees into a single hose, as does most people's, and that may be a bad idea. With a lot of miles on the engine, I expect I have a lot of blow-by to clear from the crankcase. I'll certainly take heed of the advice offered above, also. BTW. My dealer also says an improved Harley gasket is on the way, which suggests that engineering may also be a part of it. |
