Broken crank and high compression

What are we talking about - building a very high revving short stroke motor or a higher than standard revving long stroke motor ? With the short stroke motor, you have to use it in a bike and get a better all-round package. With the long stroke, you work at getting better mid-range power, so don't rev it as high. The bike you need to house it is different.
If I was going going to make a bullet-proof short stroke crank I'd use:
C 0.2% , Mn 0.5 %, Ni 3% , Cr 1% , Si 0.2%
then quench and temper the billet prior to machining - nitride afterwards to control the wear on journals.. Where do you buy that steel and get it clean enough to get the physical properties ?
Then the next step is stronger crankcases and longer rods and lighter pistons.
When I hard chromed the journals of my Triumph crank, it had been reground and gas-nitrided two times and was still wearing too fast and losing oil pressure. With the chromed journals I was able to use morris mini cooper copper-lead bearings. I always had excellent oil pressure after that. I don't know what the steel was in that billet crank, but it used to wear like crazy at those high revs. The motor used to pull to 10,500 RPM until I fitted the 2 into 1 exhaust, then it only went to 9,500 RPM - but it became usable power, not stupid. With megaphone exhausts, riding that bike was a suicidal exercise. It almost killed the guy who had it before me and he was an excellent rider. A lot of the problem was drum brakes, but the motor was extreme and vicious - a deadly combination.
When you have a bike like that, you tend to blame your own lack of riding skill if you don't do well in races. That bike turned me into an instant dud. I still have anxiety when I think about it.
The Seeley 850 with the long stroke heavy crank is a much better deal - so much faster because I'm not continually fighting it to stay alive.

What we are talking about is Norton crankshaft durability and I redirect you to the first post of this thread. This concept of durability really transcends short stroke versus long stroke. I have seen plenty (most) of the carnage due to long stroke crank failures, not short stroke screamers.

Drop your Ni requirements a bit and double your carbon requirement and you are talking about 4340 which is the "gold standard" for crankshaft material. Why go down a strange unproven and silly rabbit hole. You can purchase this 4340 online all day long with mill specs. That goes the same with EN40B which Dave Nourish had used for years.

Glad to see you acknowledge the wisdom of nitriding for wear but seem to continue to miss the other main reason which is to increase flexural durability. This is what we were discussing all along.

I suggest a short stroke crank is much less likely to break than a long stroke crank. Why talk about durability of standard Norton cranks when there are much better options ? The main problem is the standard of racing on offer, rarely justifies the expense of building something adequate, especially in the case of Norton Commandos. I had the opportunity to make a better crank in about 1978, however the incentive to excel was not there at that time and now the situation is even worse. I don't think this forum even existed back then. I believe that nitriding a billet crank would make a rat's arse bit of difference to it's resistance to failure by except perhaps for wear resistance on the journals.

I'd like to make another suggestion - Perhaps balance factor and compression ratio have more effect than nitriding on whether the crank breaks in service ? A Triumph 650 motor rigidly mounted with 12 to 1 comp. revs safely and smoothly to 8,000 RPM using a balance factor of about 78%. My 850 Norton motor, with 9 to 1 comp. , rigidly mounted revs safely and smoothly to 7000 RPM using a balance factor of 72%. A normal 750 Commando motor has 9 to 1 comp. isolastics and uses a balance factor of about 54%. If the motor was rigidly mounted, it would probably rev smoothly to 4000 RPM. When the Commando is accelerating hard, the chains are at full stretch and the isolastics are fully compressed towards the rear of the bike. So the effect might be the same as having 54% balance factor with a rigidly mounted motor at high revs with a comp. of 9 to 1.. Vibration is an indication that damage is being done to the motor's internals ? When you balance a crank to 70%, you balance 70% of the reciprocating weight. The other 30 % is inertia which gets chucked up the hole and overcomes compression. It probably also flexes the crank.

acotrel said:

I believe that nitriding a billet crank would make a rat's arse bit of difference to it's resistance to failure by except perhaps for wear resistance on the journals.

Click to expand...

This is your belief? Well your religion is so wrong on this point. Do some reading on the matter. I recall Ken and I provided some useful links further back in this thread.

acotrel said:

When the Commando is accelerating hard, the chains are at full stretch and the isolastics are fully compressed towards the rear of the bike.

Click to expand...

This statement suggests you have no clue as to how the Norton Commando isolastics are configured or how they work. The only compression of the isolastics is from the thrust from the back wheel, the braking thrust from the front wheel, the moment due to the weight of the bike, and the vibration from the engine/trans/swingarm/rear wheel assembly. The isolastics are not "fully compressed".

I see what you mean. In a commando, the engine, gearbox, swing arm and rear wheel are one assembly and don't move independently from each other. However when the motor is pulling hard, the crank has purchase against the rear wheel contact patch through the chains and gearbox and that provides rigidity regardless of the isolastics. The 54% balance factor might then come into play and cause excessive flex in the crank. I understand that the cush rubbers in the clutch can have an effect on whether the crank breaks or not. Do Commandos have those or only in the rear wheel ?

Politely; Alan/Dances, why don't you give it a rest on this thread and let it die. Please.

Alan, check out the various threads about cush drives in belt pulleys, or just look at the AN parts book, there are no rubbers in a Commando clutch, or the majority of diaphragm clutch belt drives uses as substitutes.

Dances with Shrapnel said:

Generally, there are three classes of Norton big twin crank failures:

1.) Fracture at filet on drive side main bearing
2.) Catastrophic failure of cast iron flywheel; usually associated with unjudicious lightening of the flywheel and abuse or racing
3.) Failure of the rod journal (only seen drive side failures) which invariably go between outer filet and oil feed hole.

Note that I "go between" as I am note sure where the failure starts, but my hunch is the oil hole.

Anyone want to chime in about knowledge is timing side rod journal failures or the start of fractures on the drive side journal?

I have personally done all three and witnessed a few others.

Click to expand...

My original crank broke where the drive side shaft joins the cheek. I had had the engine rebuilt a couple thousand miles earlier. Prior to the rebuild drive side super blend was really loose. The guy who did the engine had the crank sprayed with some kind of metal, and ground back down, or something like that. It didn't work and he no longer has this done as a result.

This was with maybe 70000 to 80000 miles on the engine. I heard a weird knocking noise before it happened. I pulled over and pulled the cover off the primary trying to figure out what was going on. I couldn't see anything wrong. I got back on the highway and it failed, I thought my engine had seized until I took it apart.

ewgoforth said:

The guy who did the engine had the crank sprayed with some kind of metal, and ground back down, or something like that. It didn't work and he no longer has this done as a result.

Click to expand...

It wasn't the dreaded "flash of hard chrome", was it?

My next rebuild I am going to look into shot peening the fillets on the main journals and rod journals. I understand they can effectively mask off the areas you do not want to be treated and the peening increases the surface residual compressive stresses - a good thing. Of the broken cranks I have seen and done, there are still a handful out there that, in my not so humble opinion, started their crack failure at the drive side journal oil feed hole. Maybe those holes should also be targeted. I'll have to dig through my bin of offerings to the gods of speed and revisit the bits if I still have them.

Anyone have knowledge of a combination of nitriding and shot peening to increase flexural durability; what should be done first, or does it make sense to combine the processes?

Back when I was still racing, before good aftermarket crankshafts were available, I used stock crankshafts with some modifications. Besides radiusing the drive side edge, fitting 3/8" fasteners, and rebalancing with heavy metal slugs, I also had the cranks shot peened by Bob Gorsuch at Excello plating. Bob was the go-to guy for such stuff for the SoCal racing crowd, and was aerospace and race car certified for shot peening. He masked off the bearing surfaces, but left the big end and drive side radii open to be peened. He claimed that doing so improved crankshaft life significantly.

Ken

If you were 'rebalancing with heavy metal plugs', were you using the commando frame and what balance factor were you using ? I don't think I'd be revving a crank with a low balance factor to 7000 RPM regardless of the isolastic mounts. Better to cop vibes low down in the rev range and have it internally smooth higher up ?

We put heavy metal plugs in the crankshaft cheeks and took weight out of the flywheel. The idea was to move some of the forces out to the main bearings instead of in the middle at the flywheel. In the process we also added enough weight to bring the balance factor up to 62% (wet) for engines in both the isolastic Commando PR and the featherbed with Commando engine tilted forward. That was the stock balance factor in the PR engines from the factory, and was a number recommended to me by a several flat track racers, as well as Axtell and my balance guy, for a tilted 750 engine in a featherbed. That's the balance factor I used for years (decades?) in both bikes with 750 engines. It worked fine for me, being reasonably smooth at high rpm. Back in those days, I used a 7200 rpm absolute max rev limit for anything with the stock 89 mm stroke. It's only when I went to a larger engine in the featherbed (920 cc) that I used a higher balance factor. I did use a higher balance factor for engines I built for some other frames (Seeley, monoshock, Bonneville streamliner), particularly the larger displacement engines. There is really no "one-size-fits-all" balance factor for Commando engines.

I've also sometimes used (and still do) heavy metal inserts to rebalance for switching from stock alloy rods to steel ones, as well as the heavier pistons in some of the big bore engines.

Ken

My friend has recently updated his Triumph 650 racer to 750cc and has rebalanced the crank to it's usual 78% factor. It still vibrates too much at high revs. Nothing is ever easy ? When I rebalanced my Commando 850 crank for racing, I tried to get the factor up to the high 70s. I plugged the hole in the flywheel with a threaded steel insert, then started to drill a hole in the other side. I suddenly thought about what I was doing and stopped. The factor ended up at about 72% which with the low comp. ratio is excellent. The motor is super smooth right up to and through the top of the usable rev range. When the bike idles, it rocks backwards and forwards. I think that if I was using isolastics, I would still go to about 70% balance factor and live with the low frequency vibes. Even if the bike feels smoother with the isolastics, the forces inside the crankcases are the same. With the isolastics there might be slightly less purchase ?

acotrel said:

I see what you mean. In a commando, the engine, gearbox, swing arm and rear wheel are one assembly and don't move independently from each other. However when the motor is pulling hard, the crank has purchase against the rear wheel contact patch through the chains and gearbox and that provides rigidity regardless of the isolastics. The 54% balance factor might then come into play and cause excessive flex in the crank. I understand that the cush rubbers in the clutch can have an effect on whether the crank breaks or not. Do Commandos have those or only in the rear wheel ?

Click to expand...

Well that "the crank has purchase against the rear wheel contact patch through the chains and gearbox and that provides rigidity regardless of the isolastics." is describing a coupling of a mechanical system from the crankshaft to the asphalt through the rubber tire patch - an ahem "Isolastic" so your view is over simplistic; especially when we are talking about components spinning at thousands of rpm. Now your simplistic analogy may weigh in if we are talking about a slow steam locomotive or something like that.

and

acotrel said:

I think that if I was using isolastics, I would still go to about 70% balance factor and live with the low frequency vibes. Even if the bike feels smoother with the isolastics, the forces inside the crankcases are the same. With the isolastics there might be slightly less purchase ?

Click to expand...

From a durability standpoint, for the Norton twin crankshaft, maintaining around 50% is the best with respect to bending moments; anything higher or lower is a compromise for frame durability and/or rider comfort. Unless an engine is extremely rigidly mounted to a massive block of something, it is floating around (Isolastic frame and rigidly frame) and this bit about "purchase" is just a cartoon of sorts. Think what you want, suggest what you want, but..........

'From a durability standpoint, for the Norton twin crankshaft, maintaining around 50% is the best with respect to bending moments'.

Can you justify that comment with evidence ? If the shaft is spinning a change in the balance factor affects the inertia of the reciprocating components, which changes the stress on the shaft. A crank which is balanced at 50% of the reciprocating weight would be much smoother at low revs that one in which 70% of the reciprocating weight is balanced. But when you spin it to high revs it will vibrate because it is effectively trying to overcome the effect of more compression events. What we probably need is a set of transparent crankcases and a strobe light or a computerised laser to see how much the crank bends when various balance factors are used in motors which are running.
I quite like the way the isolastics are used in the Commando. However a few years ago we used to rubber mount the motors in TZ Yamahas to stop frames from cracking. The bikes never seemed to be as fast. If power was lost because of the rubber mounts, wouldn't you expect the mounts to generate heat ?

acotrel said:

'From a durability standpoint, for the Norton twin crankshaft, maintaining around 50% is the best with respect to bending moments'.

Can you justify that comment with evidence ?

Click to expand...

Fair question.

Conduct a free body bending moment diagram analysis along the axis of a Norton crank at 0 degrees, 180 degrees, and 90/270 degrees for say 50%, 60% and say 70% BF and see where you get the least net bending moment. You will see that 50% BF minimizes the net bending moments. When I say net, I mean bending moment reversal is minimized. There is the justification with evidence. The analysis will reproduce the same results all day long. This was done for me by a fellow road racer and well respected mechanical engineer with more than thirty years in the aerospace industry. An all around interesting good guy and a bit of a rocket scientist.