Thursday, March 29, 2007

The Great Knucklehead Cylinder Controversy


Since the title of this blog is Knucklehead Theology, you are probably wondering why so little has been said about Knuckleheads so far. Not to worry, having owned and ridden Knuckleheads for over 25 years, and drag raced them for about 15, there is plenty to come.

Today I would like to talk a little bit about Knucklehead cylinders. We have all heard the horror stories of broken Knuck cylinders. And it is my guess that there are nearly as many theories as to why they break, as there are those stories. I am not going to claim to have the definitive answer, only theories of my own. This is one that I would love to have comments from others with their own thoughts.

The first time I saw a knuckle cylinder break, was while attending Motor Cycle Mechanics school at Hutchinson Area Vocational Technical Institute in about 1979. One of the other students (a Triumph guy) had brought in a Knuck chopper to work on for a friend of his. The engine that he had was a 61", but it had cylinders that were bored out to 74" plus. I don't recall exactly how much plus they were, but it would not surprise me if they were +.060 or +.070.

That is not unusual to find, but it does leave you with very low compression. There are no 74" bore pistons that are made for a 61" stroke (at least not that I have ever found), which leaves you stuck running normal 74" pistons. The only problem with that is due to the difference in wrist pin location and stroke between them, the 74" piston will be nearly a 1/4" lower at TDC than a 61" piston. (61" is 3 1/2" stroke/ 74" is 3 31/32")

Don't know if it was the owner's idea, the student's idea, or the instructor's idea, but someone's solution was to shave material off the base gasket surface to get the pistons closer to the top. I winch just thinking about it now. One wrinkle was that before the engine was assembled, another student bought the bike! The results were predictable. After the new owner finished assembling the engine, during initial start up, one backfire while kicking broke the cylinder off at the base! As a side note, the new owner bought a set of used 61" cylinders which were also at a 74" bore. He rode the bike for one season, and then sold me the knuckle engine when he bought a new "crate" Shovelhead motor. I ran that engine for many years and many, many miles with that ultra low compression.

The next time I had occasion to see an "exploded" knuckle cylinder, was a case of being in the right place at the right time. I happened to pull up to a stoplight behind another knucklehead. As the rider attempted to pull away in a quite normal manner, his front cylinder lifted. Of course I pulled over to see if I could lend a hand. This was in about the mid 1980s, so my memory is less than clear, but I do believe that the part of the piston which unfortunately was now in full view, was badly scored.

Fast forward about 10 years. While not a knuckle, the next broken cylinder that I remember seeing shares some similarities. This was on an old Pan/Shovel stroker belonging to one of the owners of the shop I worked for at the time. It had a set of +.070 stock cylinders. This one I took personally, because I had recently bored the cylinders. This one too had a badly scored piston skirt. (I believe that the scoring was due to the cylinders being powder coated after boring them which may have distorted them)

Soon after this, another acquaintance on his knucklehead had a cylinder explode while accelerating from a stop sign. As best as I can recall, this also had a badly scored piston. (I do not believe that I had anything to do with building that motor)

It should be obvious by now that all but the first incident that I related had one thing in common (at least if my memory serves me). That, of course is that in each case the broken cylinder was accompanied by a scored piston. As far as I know, they all shared one other thing in common; a large overbore on the cylinders.

Somewhere along the line a customer brought me a set of +.060 Knuckle cylinders to check for him. Looking at the bore I found something quite interesting. The bore had actually turned blue between the fins. Honestly, you could see where the fins were from the inside because the spaces between them was blue! For years after this I avoided boring Knuckle cylinders more than +.050. I just knew that if the cylinders were thin enough to turn blue like that, they had to be on the verge of exploding.

I am not so sure anymore. Over the years I have had the pistons score badly on my wife's 61" Knuckle numerous times, for numerous reasons. At least one set can be chalked up to the manufacturer. I saved the little instruction sheet that came with the pistons which called for .001 clearance. A couple more sets of identical pistons gave similar results. Then another set of identical pistons came with a similar little instruction sheet which called for .003 clearance. Oops! Another set went because we went for a weekend trip with a group of friends when the engine was fresh, the weather hot, and the speeds excessive. Another I attribute to a burned out head gasket sucking air. Well, you get the idea. But no matter how many pistons we scored with that 61", the thick cylinders held up fine.

Maybe I should point something out here. When I speak of a "scored" piston in a Harley, it is in reality a piston that has seized. The reason that in most cases the engine does not stop, is because of the extreme weight of the Harley flywheels. If pistons "scored" like this in any other motorcycle engine the motor would be stuck tight. Understanding that this piston scoring phenomena is actually a piston seizing phenomena should really shed some light on the broken knuckle cylinder issue.

Add to this, some anecdotal evidence from my 74" Knucklehead. I have run it for many years at +.050. I never was willing to bore to +.060 for the above mentioned reasons. Each time I have had it apart, I have honed it to bring it back to straight and round. I think that the last time I put it together it was with about .008 clearance on a set of high compression cast pistons. Never a bit of problem, and I run it very hard!

So here is the big question. Are the stock Knucklehead cylinders just too weak to be safe when bored much oversize? Or is the real culprit the mechanic that set them up too tight, causing it to seize? Well, before you go blaming your local mechanic, keep in mind that he probably sized your cylinders to specifications. Its not really his fault if factory specs give bad results, particularly if he does not do many Knuckleheads. Furthermore, too tight set up specs is only one possible reason for your pistons seizing. Anything that adds too much heat, or removes too much oil can give the same results. (think timing, idling in traffic, oil breakdown, etc.)

Are the "cheap" reproduction cylinders prone to breaking too? I really don't know. Some would say that they are. Part of the problem is that most of the information we have will be anecdotal (I can't believe that I am actually going to use that word twice in one day). Since a failure like this tends to be catastrophic, most who experience it once will be leery of taking any chances of it happening again. My guess is that if you had a +.070 break on you once, you would be sure to replace it with one as close to standard as possible rather than making some adjustment and using another +.070.


Well, it looks as though I am going to have to make this into two posts. In the second half I will offer my thoughts and attempt to offer some solutions. In the mean time, I would love to hear from any of you who have had Knuck cylinders break, and the circumstances.

23 comments:

Dennis said...

Good blog, Lee! I've added you to my blogroll.

Dennis

R. Sanocki said...

Just had the front cylinder explode on my 47 knuck EL. The information you have is very helpful. You guys are hard to come by these days Thanks!

Jim Franco said...

Hello Lee...Jim here again! I didn't gather from reading this article on the potential to have OVERSIZE BORED Knuckle cylinders prone to failure...that you differentiated between 61ci cylinders or 74ci cylinders other thyan the guy's failure due to the machined base area.

Is there a diffeference in the ability of 61ci cylinders vs. 74ci cylinders when it comes to how oversize they CAN SAFELY be bored out to?

I have heard the rumor that 61ci cylinders can go MORE oversize when bored out than can the 74ci! With that same rumor comes the figure "61 ci cylinders can go out BEYOND the .070's where 74ci should not go beyond .070's! That same rumor speculates, or rather STATES, the 61 ci cylinders have the wall thickness to do that whereas the 74 ci does not. Any truth to this and can you account for it if there is? Thanks for sharing your knowledge here, Lee....

Jim Franco

St. Lee said...

OK, Jim, I am about 99.5% sure of the following: Knuckle 61 and 74 cylinders were made from the same casting, with the machining being the only difference. The 74 has no fire ring, and is taller that the 61 by the height of that fire ring (measuring gasket surface to gasket surface). The bore of the 61 is 3.312 whereas the 74 is 3.438 You can bore the 61 to 3.438 (std 74)and use it on a 74 but the compression ratio will be considerably higher. The difference in height (gasket surface to gasket surface) is just like shaving the heads to raise compression. Since the casting is the same a 61 cylinder can be bored safely to the same dimension as a 74, however the fire ring will get really thin at some point, and I have seen some cylinders where the bore was no longer centered in the fire ring so that it will disappear on one side.
I personally prefer not to bore a Knuck cylinder to more than +.050 - 74" whether it is a 61 or 74 cylinder. That would be about 3.488" Obviously a 61 cylinder can be bored to +.090 of the original bore size, but that is still only 3.402" (3.312 + .090) Perhaps that is what they meant. Usually +.090 61 inch pistons are the largest available. The wrist pin location is different on a 74" piston, so the compression ratio will be lower if using a 74" piston on a 61" flywheel.

Hope this made things clearer, rather than muddying the waters.

Michael said...

Dear St. Lee,
I can hardly contain my gratitude for finding your blog. I have been searching in the wrong places for technical advise for Knucklehead engines.
Sometime ago my buddy told me that if I made it home from the war in Iraq, that he would sell me his 1947 EL bobber project. Well I'm here to tell you I am the proud owner of said project.
There are no internals with the engine.
I just learned, from reading your blog, that I have 61" cylinders ( with fire ring very thin). The bores measure 3.468 respectively, which leads me to believe this was one of those very low compression, 61" flywheels and 74" inch piston engines of which you speak. The carbon buildup at the top of the cylinder is very deep, measuring approximately .410
I realize this is a next to impossible question, however, do you have a ballpark figure of how much a flywheel, connecting rod assembly would cost from you for this 1947 EL? I would like to send the cases to you for inspection and installation of the bottom end.
Sincerely yours,
Michael

Anonymous said...

I heard one guy who worked on water cooled racing engines say that loose is good and looser is better, my guess is this would apply double to air cooled engines. Open 'em up!

Anonymous said...

Hey Lee, I have a race engine business that specialises in EFI road race engines (for auto racing) & I am new to this Knuckle stuff. I bought a Knuckle basket case the other day and noticed the tops of the cylinders have been milled into the tops of the fins. The case is stamped 47EL but after checking stroke, crank is a 61", and the bores measure 3.4975. I'm guessing they used a .060" over 74" piston and had to cut that much off the deck to get the compression up? I am also working on some heads and cylinders off my friends 47/74" knuckle & noticed that they have the fire rings on them. They measure 3.4375, so I'm assuming he has 61 cylinders on his 74" and wondered how much does that typically raise compression? I haven't cc'd the heads or piston dome, but just wondered will it be around 10 to 1 or what? Last question, I assume even the 74" big port heads all have the register in them for the fire ring? Thanks, Todd

St. Lee said...

Hi Todd, just bought a Knuckle basket case? I thought they were all gone!

It sound to me like you probably have it figured out correctly as to why your cylinders look as they do. A 74 inch Knuck cylinder would be about 5.525" gasket surface to gasket surface. The 61 would be about 5.405 gasket surface to gasket surface. At a 3.5 inch bore there would be so little of the original fire ring left on a 61" cylinder, it would probably have been removed (O.D. of the fire ring is about 3.550") Either way, those cylinders are dangerously thin at that bore.

Any Knuck cylinder with a fire ring is indeed a 61". Using it on a 74" would be like shaving the heads .120" The 74s were only 6.6:1 C.R. originally, but hard to say what it is now, not knowing the piston dome volume. Probably not 10:1, but still pretty healthy I'd bet!

All Knuck heads '40 and later are the same, whether 61" or 74" and have the counter-bore.

Good luck on your project!

Anonymous said...

Lee,
regarding the 61-74 knucklehead cylinders, and compression etc. It would seem the easiest thing to on an FL do would be to bore 61 cylinders for 74 standard pistons,remove any top ring that's left, and add thin base plates (.100-.120)to correct to 74' compression. If they use the same casting, (as I think I also saw on the s&s blog) Do you see any problem with this? thanks, Tom

St. Lee said...

Hi Tom, That should work out fine and dandy. The only drawback to it is that most guys restoring bikes don't want stroker plates showing, and of course many of those with 61" cylinders still have the 3.5" stroke 61 bottom end.

Anonymous said...

Lee,
since the plates would be only .120" or so, could they be put on the cylinder bases with a high temp epoxy. then the edges could be finished off unseen and painted with the cylinders. As far as the el's , I can't believe someone hasn't made way oversize pistons for those, since there's so much more left in the walls for the fl size bore anyway. thanks, Tom

St. Lee said...

That should work very well, Tom. The EL piston issue is probably one of money. A cast piston is way cheaper, but only when made in large quantities. Forged pistons could be made easily enough; several piston manufacturers offer custom piston services. But forged pistons are already quite expensive and if you add a custom build premium to them, most people would walk away from the price.

Richard said...

Hey Lee,
Its Richard (Cone-Knuckle)again. I had a front cylinder lift off on the freeway in 1987. Scoring was not the problem because I wedged a board I found between the frame and head and got 20 more miles out of the motor, then it died for good.
I was talking with a Brothe3r from Colorado who had the same happen on his way home from South Carolina. Mine blew on the I74 and I55.
The only thing the two motors had in common were the Cheap ZEL cylinders. I no longer have this problem because in '89 I went to Ronnie Trocks (RIP)in Hampshire, IL and bought a set of cylinders. Good move.I run a 5" Stroke and I'm just getting ready to refresh a top end.
Back to the cylinders of the middle '80s I think ZEL was the biggest problem!
hope your having a good day,
Richard

St. Lee said...

Richard, I remember about that time Custom Chrome had what amounted to a recall on the Knuck cylinders that they had been selling because of reports of some breaking. Of course they had no way of tracking the suspect cylinders, so their "recall" consisted only of a letter to their dealers warning of the possible problem, and if I remember correctly, an offer to accept returns. I always thought that it showed some integrity for a large parts supplier to even go that far. We need more of that in the motorcycle parts aftermarket.

But the fact that you and others have seen evidence of broken Knuck cylinders without accompanying scored pistons does look like it blows my theory at least part way out of the water.

Unknown said...

all im interested in right now is owning a low compression pan or knuck..ive had 2 knees surgeries & had to sell my ironhead & now own a flathead 45..i have no troubles kicking my 45 over..i went to by an fl pan,tried to kick it over & no go..whats the best combo for running a low compression kick friendly bike..also will it run like a snail my 45..i hope not

St. Lee said...

Hi Tony, that's an easy one. For years and years I rode the bike you are looking for. It was a Knuckle with the 3-1/2" stroke 61 inch flywheels and 61 inch cylinders bored out for 74 inch pistons. The different wrist pin location on the pistons meant that when they reached the top of the stroke they would still be 1/4" below the top of the cylinder, resulting in quite low compression. In fact, when I wanted to show off, I could start the bike using my arm instead of my leg.

One would suspect that such a combo might be really anemic performance-wise, but it would run with freeway traffic and was not noticeably slower than the Pans and Shovels of the day that I rode with. Notice I said "noticeably." I would guess that top speed would have been lower, but I really can't remember an incident where I found that out for sure. Nobody I ran with back then had speedometers anyway.

Two pieces of advice if you build something like this: don't run a hotter than stock cam (it would sacrifice low and mid range power), and don't gear the bike higher than stock (use all the mechanical advantage that was built in originally).

BTW, the reason for running the motor like that back then was simply a matter of running whatever combo of parts you had readily available that would not entail spending more money.

Good luck!

Anonymous said...

its clearance you fools and your animals on them antiques. Dont you know about metal fatigue

St. Lee said...

Well, sure there is metal fatigue involved with any engines which are past the age they should be drawing Social Security. And of course clearance is involved in getting them to live long into their golden years. The whole point of the post is sorting out what combination of clearance and maximum over sizes will keep them from becoming grenades.

Unknown said...

My Knuckle experience; I bought my 47 E in 1982, it had 74" wheels, .020" over 74 pistons in 61" cylinders. Ran great for 5 years until it dropped a valve. Went back with UL wheels, ZEL 74" cyl, TRW Forged std pistons, stroker plates. Ran great for 2 years until it lost a wrist pin keeper. bored the ZEL's .080" over, fitted with Harley (Mahle) +.020 80" pistons. Ran great for 12 years until it blew out the rear cylinder wall doing sidecar service. head was hanging by the motormount, cylinder, piston and rod were destroyed. Cylinder measured .020" thick at the thinnest spot, some scoring was present. Went back with Taiwan cylinders, std JCC cast pistons (fit at .003" cl), evo rods. Proceeded to blow holes in one front and two rear cast pistons, no scoring. I went back with the original Std bore TRW forged pistons (milled the dome to reduce compression) fit loose at +.007". This combo ran great for 400 miles, when the rear cylinder broke the right half of the flange under light load. It still ran didn't even misfire. My cylinder search brought me here. So I have blown a +.080 ZEL and cracked a std bore Taiwan repo, that was set up loose. Go figure.

private said...

Hi Lee. Your opinion please. I have 47 FL. Previous owner did top end years ago. At that time he found new oem 74 cylinders and installed them which have standard bore. The motor runs fine. But I'm thinking the cylinders and pistons might need some attention. It kicks over very, very easy and puffs slightly at start up. Also I see a lot of carbon deposits on piston top as well as the head. Wondering if this is a situation
Of simply getting the cylinders honed and put it back together with new rings.

St. Lee said...

Hi "private", Really a tough question on which to give a definitive answer. First, on the "easy to kick over" - ring seal is only one of the possible causes. A lot of parts have been interchanged on Knuckles in the 70 years since that last one left the factory. A more common occurrence than you might think is the substitution of 61" flywheels for 74" and vice-versa. 3-1/2" stroke flywheels with 74" pistons is a notorious combination for very easy kicking.

Other possibilities include pushrods adjusted so tight that they don't allow the valves to close completely,valves not sealing properly, and of course as you suspect the ring seal. It would be a good idea to do a compression and leakdown test before tearing it down to see if you can isolate the problem area.

A puff of smoke at start-up could very well be a bit of excess oil in the spring cups leaking down the valve guides and onto the pistons as the bike sets between start ups.

Steve said...

St. Lee, first I want to thank you for the information you've provided in this blog on the Knucklehead cylinders. This knowledge has been lost. I have an old 61" Knucklehead I've had in boxes since the early 80's and am in the process of putting it back together. It too was running with 61" flywheels, 61" cylinders and 74" +.050 pistons. While discussing this with a friend he suggested modifying the cylinder head to accept Panhead cylinders and use 80" Shovelhead pistons. Can you offer any thoughts on this and what would be required? I know it was common back in the day but with the availability of aftermarket 61" and 74" Knucklehead cylinders is this still a viable or necessary option?

Steve

St. Lee said...

Hi Steve, thanks for the question. The Pan cylinder conversion was a good choice when there were no quality Knuck cylinders readily available, and I have done it a number of times myself, but it's probably not the best choice today. Besides, there are a couple problems with your friend's suggested fix. One is that a standard 80" piston is equal to a +.060" 74 bore, so the Pan cylinders would need to be bored to within one size of their recommended maximum oversize. The other is that because of the wrist pin location on the 80" piston, your compression ratio would be even lower than with a 74" piston. Where the 74" piston would only come within about 1/4" of the top of the cylinder at TDC, the 80" piston would be 3/8" low when coupled with the 61" stroke.

Probably the best solution for you would to purchase a set of aftermarket 61" Knucklehead cylinders and pistons. The 61" cylinders have so much meat in them compared to a 74" that even a cheap repop cylinder should be plenty strong. Just be sure to allow more piston to cylinder wall clearance than a modern motor and you should be fine (if you purchase a "pre-fit" set of cylinders with pistons, don't take it for granted that it was sized correctly).