What is the real purpose of crankshaft counterweights on a flatplane crankshaft? 6

[ELS122]

I don't see the reason in having counterweights on a flatplane crankshaft, for example an inline 4.
The centrifugal forces for cylinders 2 and 3 are counteracted by cylinders 1 and 4.
And you may say well there's shear loads resulting from that, flexing the crankshaft and reducing the lifespan of the crank and the main journals. But then why do engines last in the first place? The shear loads and flexing caused by combustion would be WAY higher than those inflicted by centrifugal forces.
So what is the real reason for having counterweights? Could it possible be an oversight by every automotive engine engineer?

 

[SwinnyGG]

The loads put into the crank journals due to rod/piston inertia are huge. The logic you're following, that the loads become shear loads, only works if the crank is infinitely stiff, which is isn't.

Cranks needs counterweights for durability and NVH concerns.

 

[GregLocock]

Counterweights are sized by looking at the bearing pressures.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[ELS122]

Are the bearing pressures measured in a real engine or is the crank just spun on a bench with rod weights attached?

 

[GregLocock]

FEA model

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[ELS122]

What about 30-40 years ago? there were many reliable engines made so there mustve been some pen and paper theory

 

[SwinnyGG]

FEA models existed long before computers did.

 

[ELS122]

Well then explain it, I'm asking why manufacturers do this and you're basically saying 'reasons.'

 

[BrianPetersen]

It is possible without using FEA to work out pretty closely, the center-of-gravity of a single crankshaft journal complete with its counterweights and isolated from its neighbors, assembled with the portion of the con-rod big end excluding the beam portion of it.

If that center-of-gravity coincides with the center-of-rotation, then the component of crank-journal loading that is due to the rotating parts being off-center will be zero. Of course, the reciprocating forces, both inertial and due to combustion, are not offset, total crank-journal loading being a vector sum of all these.

Increasing the crank counterweights beyond this, so as to partially offset the vertical reciprocating forces, will lead to a side-to-side loading component in the crank journal forces, but it is entirely possible that the peak vector sum of these forces on the crank journals will be lower at some point where the reciprocating forces are partially offset at the cost of increasing side-to-side vibration.

"BuT yOu CaN't CaNcEl ThE rEcIpRoCaTiNg FoRcEs" true, you can't, but you can minimise their sum total under some particular design operating condition that is of interest.

At high RPM, the inertial forces are not negligible, and the forces trying to bend and twist the crankshaft are not negligible.

 

[TugboatEng]

If you are using the pistons in two adjacent cylinders you counterbalance eachother, then the crankshaft must transmit these forces between cylinders IN ADDITION to the more useful combustion forces. These additional forces lead to greater deflections which can be amplified by harmonics and lead to significantly higher crankshaft stresses. As engine speed increases so do these forces.

On the opposite end of the spectrum, at very low speeds, counterweights are infact unnecessary. These balance-weight less engines often have critical speeds where they cannot be operated for extended periods of time.

 

[SwinnyGG]

Well then explain it, I'm asking why manufacturers do this and you're basically saying 'reasons.'

Fun fact, no one on this forum owes you anything.

With that said - the explanation is very simple, and it was already provided to you.

Inertial forces are very large at high N. Those inertial forces do not resolve to simple shear in the crank - the crank is not infinitely stiff, so large perpendicular forces mean large bending forces. Counterweights reduce the magnitude of those forces.

 

[GregLocock]

I was only involved with crankshaft design after FEA came in, but before then we'd have split the crank up into a series of bending elements, and then run an iterative solution for the bending, and hence the reaction forces at the bearings. I'd guess that we had Fortran programs to do that it's the sort of thing I spent much of my 20s writing.

The funny thing is after all is said and done you usually end up with around 50% of the reciprocating mass, and of course 100% of the rotating mass. Which is a big clue that inertia, locally, is the issue.

For one crankshaft for an I6 we treated it as 6 individual pistons, so ended up with 12 counterweights. It weighed 2 kg more than its predecessor, but has a great reputation with those getting 500 hp out of a 4 litre engine.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[enginesrus]

Counter weights are win win, for all the above mentioned information, as well as adding extra flyweight.

 

[ELS122]

The centrifugal force on the crankshaft would be constant at x rpm, so the crank would be in a constant flexed position. So if you counterbalance those forces the crank doesn't flex at high rpm.
But the combustion forces arent constant and only happen in brief moments twice per rotation. So they won't counteract the centrifugal force adequately.
Reciprocating forces should be considered when choosing crank counterweights even in a flatplane crank. Perhaps with some way to tell how much flex is in the crankshaft would be a good way to tune the counter weight on a balancer.
Also consider how the ring friction will also add onto the "effective" reciprocating weight.

 

[3DDave]

The forces from combustion are only seen because the crank reacts against them. If not, the gas would expand at multiples of free-air mach number. Part of that resistance is accelerating the counterweight angular velocity. "Brief" is relative.

 

[GregLocock]

Typical torsional resonance of a crank is at 300 Hz. That is it responds to stuff that happens in 3 ms very easily.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[BrianGar]

Fun fact...I once cut a vw gti inline four crankshaft exactly down the centre - when viewed from the side, with 1+4 at tdc, and 2+3 at bdc. Big end journal bits weighted exactly the same as the counterweights.

I could write about 10,000 words more on this but I'll hold off. All I'll say is most aftermarket crank balancing companies don't have a clue what they are at, and I'm also told some crank designers were not too sure either.

''Its all a tradeoff'' is a safe answer in these circumstances when in doubt.

If you are in doubt...google pictures of any same cylinder count cranks....from inline 4s, to F1 V10 items, or even mass produced Ferrari engines vs whatever. They all work/won races, have much the same cylinder bore/piston mass, yet can all look vastly different in terms of counter weights - and even with PTO mass being similar.

Brian,