Motoscrubs.com

I recognize that!

Another idea is that the centrifugal filter is less effective at low RPMs. The radial forces increase by the square of the RPM. Comparing 2000 RPM with
5000 for example: The forces pushing heavier than oil crud into the sludge trap is greater by a factor of 6. Imagine the effect of just a tiny particle caught between the rollers and rod (or the crank pin OD). The clearances are small. I think 1000 mile oil changes are a must.

Dave

Here's the graph I mentioned in my post of yesterday that relates to the figures Harvey posted.

Jordan,

I agree empirical knowledge shouldn't be dismissed. I have come to the conclusion that some people have suffered big end failures due to:

1. Poor Big ends due to low priority of quality control at Ducati factory.

2. Owners slogging motors that have suffered detonation, owners not knowing tell tale signs of detonation and continuing the abuse. If motors are non standard, (higher compression ratio etc.), then they would be more vulnerable. Kevin Cameron's article was very interesting for its information.

Either reason of failure could have resulted in an urban myth starting. I would prefer to dig deep and get the best information we can, so that people can just enjoy riding their Ducatis, without unfounded fears.

Nick,

The other night I saw Dovizioso come 5th in the Jerez MotoGP on a Ducati, so I would say that the Ducati engineers got it pretty right; a minor problem is that Honda and Yamaha got their sums right too, and maybe Marquez is a better rider! Regarding Hailwood's shocking (sorry, couldn't help that) behaviour; I think the frames were pretty basic back then and Honda were on a steep learning curve of integrating lots of competing demands to make a successful GP machine. I think the engineering of a big end is pretty simple in comparison to the handling of a GP bike of any era. Remember even Ducati called in Seeley to build a frame for their 500 vee twin GP machine.

Bob, Dave,

There is no particular evidence that this is a cause but I am open to anything you can come up with. Plenty of British thumpers, with roller big ends, have been used in trials competition. This sport is all about slogging motors at low revs and I don't think these machines have very much different lubrication for their big ends than the Ducati. Of course I agree that regular oil changes are good.

Cheers,

Stewart D

Post by StewartD »

I have based my torque figures on assuming the Sebring’s peak output at 6250 rpm occurs at the dynamometer’s 75 mph, and that the 250 Mk 3’s redline of 8500 rpm occurs at the the dyno’s 71 mph.

I have used the same method to calculate torque in Newton.metre from the graph’s horsepower and rpm figures as I used in my post of the 18th of April.

The torque curves I have plotted both show that torque at low rpm, (say below 3000 rpm which seems to be a level that some on this thread suggest is a speed that is slogging the motor), is small compared to the peak torque that the motors develop.

For instance for the 350 Sebring gives 4 N.m of torque at 2500 rpm compared to 24 N.m at 4583 rpm (4/24 = 16% of peak torque), and the 250 Mk 3 gives 9 N.m at 2700 rpm compared to 23 N.m at 4167 rpm (9/23 = 39%)

The force the big end bearing experiences are directly proportional to these torque figures. This is simply a fact of geometry. Torque is a the result of a force multiplied by the ‘moment arm’ or the radius of action, or half the stroke.

Such comparatively low forces, (compared to what it is designed for), will not damage a bearing.

Stewart, Yes I got confused trying to find out, why you got the torque figures so wrong, in your calculations.

A 350cc cylinder at 2500 rpms producing only 4 N.m/2.9ft/lbs??
You could get that by kicking it over on the kick-start, I would expect at least 15 N.m, to 22 N.m, depending on the valve timing.
Have another look at the table that I posted, compare the torque output for the capacity. The forces on the big end, at low rpms, are not as low as your calcs show.

I think it would have been better to use a Dyno torque graph, instead of trying to work on a HP graph.

Harvey,

Why are my torque figures wrong? Any one who has stalled a motor knows that torque is near zero just off idle.

If there is something radically wrong with my assumption for the 350, (that 6250 rpm is the revs at which the Dyno reading is 75 mph), then I suggest you tell me why that assumption is so wrong. Better still tell us your what you think I should have used, and graph the results of your calculation to give the 'correct' figure.

I stand by the figures that I graphed. Your table, is for a simulation for an unknown type of motor, so I don't see why these figures are relevant to a discussion of Ducati singles.

From the first time I posted the Dyno graph in this thread I have asked if anyone could post a Torque vs RPM Dyno graph for a Ducati single, but to no avail. I ask again.

Cheers,

Stewart D

StewartD wrote: Any one who has stalled a motor knows that torque is near zero just off idle.

...with throttle closed, but what about WOT?
The crankshaft doesn't have to be moving at all, for torque to exist.

Are we going around in circles ourselves?

I noticed this on King bearings website. It has this statement about bearing wear on their website. They are referring to shell bearings but I would think the effect is the same...

3. Geometric irregularities
Fig. 5 and 6 show examples of the effect of geometric irregularities.
Both problems are seen in the picture: local wear with shiny appearance and the overlay fatigue in form of spider web like cracks on the areas of metal-to-metal contact.
 Distorted (bent or twisted connecting rod) is one of the causes of localized loading of engine bearings (Fig.5).
Overloading of an internal combustion engine due to detonation or running under high torque at low rotation speed may cause distortion of the connecting rods. The distortion results in non-parallel orientation of the bearing and journal surfaces.
The non-parallelism causes localized excessive wear of the bearing surface due to metal-to-metal contact (boundary or mixed lubrication) occurring near the bearing edge.

...any relevance to this discussion?

Kev

Stewart, looking at the graph of the torque figures that you have added to the dyno chart, the 350 torque curve is a match of the HP curve, the torque doesn’t curve like that. So I guess that may be where the calcs went wrong.

http://www.totalmotorcycle.com/downloads/dynocharts/dynocharts-Ducati-2005-Monster620.htm

Can’t find a torque graph of the 350 Ducati, but this is of a 620cc Monster twin, the cylinder size is close, and it’s a two valve engine, so the torque curve will be close. As you can see the torque curve is quite flat, it does not drop at the low end. Its maximum torque of 55.9Nm is at 6700, but at 45 Nm at 3500.
This would equate to the 310cc cylinder producing 27.9 at 6700, dropping to 22Nm at 3500, not a huge drop, like your 4.5Nm at 2500.

Jordan,

Yes good point, I did make my comment on the run. But torque just off idle at WOT will still be low,The two dyno curves do not measure below 2500rpm (350 Sebring) and 3890 rpm, (250 Mk 3). (These are my assumed rpms before anyone criticises me). I don't think the highest torque figure would be obtained, just off idle with WOT. The air speed through the venturi would be too low for the fuel to vapourize. A smaller opening would give higher air speed across the jet block.

I think the dyno test doesn't measure down to idle speed (say 1200 to 1500 rpm) because of low torque and insufficient flywheel effect makes it difficult for the dyno operator to balance the throttle opening with the Dyno's applied load. I would like to get the comments of anyone who has done dyno testing on this.

In a petrol engine we cannot have torque without crankshaft speed, in an electric motor or steam engine we can. We are going round in circles a bit, but I think its worthwhile to check all possibilities; the next post by Kev is pretty interesting.

Kev,

I would assume they're talking about multi cylinder car engines with multiple crank throws between main bearings. In your quote it is:

'due to detonation or running under high torque at low rotation speed'

If a large force, due to detonation, causes the crank to flex slightly, then the big end axis would be momentarily non parallel to the gudgeon pin axis. Check my diagram;




for a Ducati single throw crank I don't think this will occur, the load is more or less central. For a two throw crank, I think it occurs due to the non symmetric loading that will occur. The diagram, with assumed flex shown in dashed line, makes this clear.

Harvey,

You say: 'the 350 torque curve is a match of the HP curve, the torque doesn’t curve like that. So I guess that may be where the calcs went wrong.'

The calculation isn't wrong: I took the 350 Sebring's horsepower to be 2 HP at 3591 rpm

2 / 3591 X 7122 = 3.96 N.m

The 620cc Monster is a much more modern machine. Just because it has a flatter torque curve does not disprove anything I have said. The maximum primary load on its big end will be at the revs of maximum torque, and at 3500 rpm the load will be about 80% (45/55.9) of that.

I admire your determined knowledge-seeking, Stewart.

It's odd that despite all the common sense about low revs and high load being bad for an engine, real data is elusive, "lost in the mists of time" maybe.
But as a politician once quipped, If it goes without saying, it ought to be said!

I once saw a Ducati single (coincidence) pull up and the rider switched off the engine. A short time later, he switched it back on and the engine burst into life without prodding. However unlikely, it produced torque at zero rpm. Some early cars were even designed to start this way, but it can't have been too reliable compared to a starter motor.

by StewartD

Harvey,

You say: 'the 350 torque curve is a match of the HP curve, the torque doesn’t curve like that. So I guess that may be where the calcs went wrong.'

The calculation isn't wrong: I took the 350 Sebring's horsepower to be 2 HP at 3591 rpm

2 / 3591 X 7122 = 3.96 N.m

The 620cc Monster is a much more modern machine. Just because it has a flatter torque curve does not disprove anything I have said. The maximum primary load on its big end will be at the revs of maximum torque, and at 3500 rpm the load will be about 80% (45/55.9) of that.

I am sure that there is nothing wrong with your Calculations; it is the data that you used for them. I mean “2HP at 3591 rpm”. Looking at the rpm figures that you added to the bottom of the graph, it looks like it was 2Hp at 2500, but 9 HP at 3590 rpm.

There is no way to tell how that dyno graph was taken. As most people are not interested in the torque, they are taken to find the HP, so the engine speed is brought up to a rpm, that won’t damage it, then full throttle is applied for the HP figures to be taken. They are not interested in what the HP is down there, as this graph shows. Look down at the Air/Fuel ratio, it was just spluttering along till 3400rpm when it got full throttle. So any data that you used below 3400rpm is wrong. It’s the old saying “feed crap in, get crap out”.

It does not matter, how old or modern the engine, they all work the same. Torque, at low rpm, is primarily related to capacity of the cylinder.
I think there is a problem understanding what happens in the cylinder, in real terms. So maybe it would be worth going back to basics to look at that process.

When the piston is pulled down on the intake stroke, it pulls air/fuel through the inlet valve, into the cylinder. When it stops at the bottom of the stroke, the cylinder has 350cc of air/fuel in it. The piston then starts to rise on the compression stroke, but as the inlet valve is still open, the gas is pushed out the inlet, till the valve closes about 60* later, to loose about 33% of the cylinders volume.
This leaves about 231cc of air/fuel trapped in the cylinder, to be compressed for burning.
The spark ignites the fuel that burns heating the air, to force the piston down, to provide the pressure/torque equal to 66% of the cylinder’s volume.
As the engine speed increases, less air is lost out the inlet valve, so a higher volume, say 300cc is compressed and burnt to produce a higher force on the piston, equal to about 85% of the cylinder’s volume.
At high speed when the inlet has maximum inertia, and the exhaust is developing maximum negative pressure, it will fill the cylinder to more than 100% of its volume, producing 100% of the maximum torque.

So we can see that, as slow as the engine can go, it will still produce at least 66% of its maximum torque, for the big end rollers, rod and pin tracts to handle.
As you say:

350 Sebring gives 4 N.m of torque at 2500 rpm compared to 24 N.m at 4583 rpm (4/24 = 16% of peak torque),

To produce only 16% of the maximum torque, there would only have been 63cc of air/fuel trapped in the cylinder.
The Monster had 80% of its maximum torque at 3500, so it must have trapped 248cc in its 310cc cylinder.