The effect of engine load and vacuum to tensile failure in conrod 1

[azmio3]

thread71-333643

Hey guys,

I need some help here. My group is preparing an engine in a test cell to investigate the effect of engine load and vacuum to tensile failure in conrod. Most of the papers that I found do not include the engine load and vacuum to be contributing to the stress experience near the conrod small end. However, the failures that we found strongly link the engine load and vacuum to be relevant especially when we really push the components to the extreme limits.

Any links to known publications will be helpful.

 

[ivymike]

I would normally think of engine load and vacuum to be counter to each other, where highest vacuum happens at high speed, no load conditions. Tensile failures in the small end might not have anything to do with "load" as load would be lowest when tensile loading was highest...although one might bring it into the discussion w/r/t transient conditions (reducing load suddenly at high speed).

There may be other search terms you could use ... I think some people had specific names for lifting off the throttle at high revs.

Another source of tensile failure in the rod small end is (of course) the piston failing to move smoothly down the cylinder bore (like when it's too hot to fit). This can be made worse by higher engine load, especially during very cold conditions, startup, and rapid+sustained increasing load transients.

 

[azmio3]

I normally relate engine load with throttle opening. With the term no load, basically the throttle shuts off and there will surely be vacuum in the plenum and combustion chamber that further pulls the piston up. I found many papers simulating the tensile loading but none so far includes the effect of either air cushion or vacuum.

 

[ivymike]

What you're saying in the first two sentences agrees with what I said.

Regarding the third, I expect it will be a small effect compared to inertia (for small gasoline engines at high rpm)

 

[yoshimitsuspeed]

It would help a lot if you explained your question better. I'm pretty lost as to what you are trying to figure out. I agree that load and vaccum don't really go together so it would help if you go into more detail on pressure differential.
RPM is the most stressful influence on a rod and the forces are pretty nuts. Not only that but it is at the moment that the crank is trying to stop the piston and yank it back down. As the piston flies to the top of the cyl there will be very little vacuum because the piston is compressing what little air is in the cylinder into a much smaller volume. As it travels back down the cylinder the vacuum will increase as the acceleration decreases so I really doubt it would be significant at all.
More importantly even a 13 PSI pressure differential (let's say you hit redline and snap off the throttle) on an 82mm piston still only equates to 104 lbs pulling up on that piston. Even if that could be applied at the moment of maximum acceleration on the rod it would be inconsequential to the inertial forces acting on it.

If you are curious about the padding effect of the air and or combustion you need to remember that it only happens every other cycle on a 4 stroke. This means that the rod must take care of it's self on the exhaust stroke anyway.
The padding effect does counter almost every tuners assumption that running more boost or higher cylinder pressure requires stronger rods. This shouldn't be true because if anything that cylinder pressure is taking strain off the tension applied to the rod.

 

[Tmoose]

2 stroke engine or four stroke?

Do you have some failed rods whose fracture surfaces you can evaluate?

Attached is an image of the rod journal loading of a mid 60s 426 hemi engine at 7200 rpm, full throttle, with a chunky 800 gram plus piston.
I'd expect the tensile load in the rod up by the wrist pin end to be reduced somewhat since the inertia of most of the conrod is missing. Note there is a big tensile load (16000 lbs) at 0 degrees, the end of the exhaust stroke TDC. Whether the throttle is open or closed that primarily inertia load will be about the same. A bit past 360, the firing TDC, the load goes a little compressive. Closing the throttle would tend to make that load more nearly the zero degree, large tensile load.

So, in my opinion, closing the throttle at full power would about double the rate at which the rod is exposed to the large tensile loads (every stroke) compared to full power operation ( every OTHER stroke). But in the end, the rod is going to fail by propagation of fatigue cracks ( as most do), not from a few hard yanks.

 

[azmio3]

Taoiseach, What bothers me is that various dyno and vehicle durability tests could not replicate this. It happens mostly during gear shifts

 

[Lou Scannon]

The padding effect does counter almost every tuners assumption that running more boost or higher cylinder pressure requires stronger rods. This shouldn't be true because if anything that cylinder pressure is taking strain off the tension applied to the rod.

It is certainly true that combustion pressure, which is generally higher in a boosted engine at full load than a non-boosted engine, counteracts the tensile forces on the rod. But depending on the rpm, masses, and other variables, the rod may be in tension or compression during the upper half of the compression and expansion strokes. The fact that boosted engines which get pushed a little too hard can fail their rods in compressive buckling is proof enough that the compressive forces can be considerable.

"Schiefgehen will, was schiefgehen kann" - das Murphygesetz

 

[winfieldblue]

a 13 PSI pressure differential (let's say you hit redline and snap off the throttle) on an 82mm piston still only equates to 104 lbs pulling up on that piston

I have to wonder how much the crankcase pressure will add to the tensile loading, applying force to the bottom of the piston, and what about those piston cooling oil jets that shoot a load of oil into the pistons interior (more weight). There might be a Helmholtz effect created by the piston as it rises in the bore, leading to a pressure wave smacking the piston from underneath. Just thinking out aloud!

 

[Lou Scannon]

I doubt the effects from inside the crankcase on a conventional 4 stroke engine are significant. I suppose it's possible to imagine an extreme or unlucky design where under-piston fluid forces would play a minor role in conrod loading.

"Schiefgehen will, was schiefgehen kann" - das Murphygesetz

 

[TheBlacksmith]

Since a perfect vacuum is only the complete absence of atmosphere, it is on the order of -14.9 psi. Surely the vacuum is an indicator and not a causative factor. I suspect rod angularity and the G loads during reversal at TDC are a strong factor in failure.

 

[azmio3]

blacksmith, the question that I have, how come countless durability tests could not cause the conrod to be broken into two but neutral high revving and rpm overshoot during gear shift can easily broke one.

 

[TheBlacksmith]

Then, as I see it, vacuum is an indicator; the neutral revving or missed shifts probably occur with part throttle (aka high vacuum). The real culprit, as others have pointed out, is likely drastically lowered cylinder pressures, that would tend to hold the piston down, shifting more of the load through the piston reversal at TDC into the rod as tensile stresses.

 

[ivymike]

If the durability test is meant to verify durability under hs/nl conditions then it was not a well designed test cycle if it didn't show rod failures. If the test was not meant to evaluate that condition (I haven't seen one that was) then it comes as no surprise that failures would occur under conditions that weren't considered.

Usually someone will run a rod pulsator test at some point though, and that should definitely verify rod tensile performance...but these little engines sometimes use rods with finite fatigue life too.

 

[Tmoose]

Inertial loads generally vary as the square of rpm.
Do those "extreme limits" include the possibility of overspeed, or much increased temperature??
Can rpm be monitored precisely and quickly? If the rpm flashes sky high briefly when load goes to zero then it could be significant.

Are the rods made of metal? Or some composite or polymer?
I think of most any metal part in an engine failing by fatigue, not because 10% higher inertia loading exceeded its ultimate tensile strength.

Was the rod geometry and material chosen based on some expected loading?
If so, what stress levels were predicted in the design phase for normal operation, and when operating at the "extreme limits?"

I'd be working to better understand the nature of the rod failure(s) before looking for the last 5% or 10% of rod tensile loading.
The material properties are likely to vary more than that.

How many hours or cycles does it take to fail a rod?
Are all the failed rods "young" ?
Do any rods survive operating at the "extreme limits" once, or more than once?
Do you have any failed rods whose fracture surfaces can be investigated?
Can you post pictures of some failed rods?
Are the rods forged, or cast by some method, or ever repaired by welding?
What means are available to qualify the rod material is correct?
What

If all the failed rods, but only the failed rods, happen to have a voids or other material defects, or ugly stress raising features due to improper manufacturing right at the point of failure then perhaps the fact they failed at some slight over speed merely reflects the QC needs to cull out the rods with gross material and significant manufacturing defects.

How do the piston, rings, and cylinder look after a rod fails?
Pistons that seize hard in the cylinder sometimes are attributed to rod breakage.

What is the condition of the rod big end, its bearing, and the crankshaft after a rod failure? Big ends seizing hard to the crank journal are also attributed to rod breakage sometimes.