Crank balance and design to minimize vibration

[twillcox]

Hello,

I am designing a new crank for a motorcycle and I have some questions regarding the balance factor. I've written an excel based program to calculate main bearing forces as a function of the dynamic balance of the crankshaft and reciprocating (primary and/or secondary rod forces). I need to validate the software somehow, and would like to work from an inline 4 if possible (or parallel twin). I need some way to find the mean bearing force plots for an inline 4 crank in 'perfect' primary balance (if this is even achievable???). I am guessing that noone has this available...so the next best thing that would validate the software is an overall block force plot as a fcn of crank angle...both forces and moments. I can force the crank to be at exactly 50% (presumably what is desired for primary balance considerations alone).

engine speed (which will only affect the magnitude) and other variables, reciprocating mass, etc are all adjustable

If anyone knows of an easy way to validate this, i'd really appreciate it.

Thanks

 

[GregLocock]

I'd be a bit surprised if there were no SAE or IMechE papers on this, particularly from the 70s and 80s.


I wouldn't say that 50% is really an optimum for anything.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[twillcox]

right...agreed. I have yet to dive into the 50% reciprocating value as a target, but was just using it for my simplified case of an inline-4.

I just found out that running the inline-4 crank with a 1243 firing order (i.e., crank journals are up-down-down-up at tdc #1), gives a "ZERO" force in the vertical and horizontal direction, and ZERO moment in the vertical and horizontal planes. This is what i was looking for and couldn't get

I had been getting what appeared to be a secondary force effect in the horizontal plane (i.e., 2 cycles per rev) (note I had forced all secondary forces to zero). It was perplexing b/c i'd always heard everyone say that an inline four has perfect primary balance, but is hosed when it comes to secondary balance. turns out this was true...I was neglecting to add in the reaction of the block to piston in my horizontal force summary.

Have you ever seen a main bearing plot, or overall block force/moment summary? I'm getting numbers approaching an 8 kN-m moment and 45 kN force on the block. frequencies are 1-2 cycles per rev depending on what crank layout is used.

Any vibration analysis on an inline-4 would be useful

 

[GregLocock]

I have but I can't tell you much about it, that was from 15 years ago.

I strongly suspect that either Taylor or Heywood will have plots of bearing forces.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[strokersix]

When I did slider crank bearing force plots in college I missed the mass unit conversion in english units and totally screwed the results. Sounds like you are using metric but just in case, check to be sure you didn't make the same mistake that I did.

 

[GregLocock]

You may find

http://www.pattakon.com/educ/balance.exe

a good crosscheck on your maths

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Lou Scannon]

The text we used in my uni was "Kinematics & Dynamics of Machines", by Martin. You will definitely find the analysis you are looking for in there.

 

[Kier14]

Hello twillcox,

I did quite a bit of work on this subject at the start of the year. When I was doing the work I analyzed a Ford I-4 1,6l gasoline crankshaft(its a single counterbalance per bore type) and found that Ford don't actually take into account for the mass of the big-end part of the conrod. They balance the crankshaft with out 2/3 of the conrod in mind. This essentially means the assembly is 100% out of balance(when you add in the big-end). To work out the unbalance Force is simple;

F=mrw^2

F=force(N)
m=Total mass of a single web(with or with out 2/3 the conrod mass)
r= Distance from the centre of the main bearing to the centre of gravity
w=2*pi*rpm/60

The force generated by this unbalance will be in a circular motion following the crankshaft rotation. As the crankshaft is symmetrical with two counterbalances up and two down, the net force is zero. This is why there is no vibration felt outside the engine. The only extra force is on the bearings. Now bearing in mind that typical gas forces in a gasoline engine can reach over 40kN on the conrod, its going to take some pretty large unbalance forces to have any serious effect on the bearings.

Regards,

Kieran.

 

[twillcox]

Thanks Kier14,

I have done exactly that with the rotating masses, i've also summed acceleration forces on the piston, and converted these to rod forces. The rod forces at it's respective angle is then applied to the crank pin, which is distributed by a moment sum to the crank main's. The 'balance' lobes and any other off-center lobes are also summed via the moment summation and applied to the main bearings.

And yes, i'm getting a rod force of 44 kN due to combustion pressure. Only about 25 kN due to piston accel (at TDC exhaust), which is the worst case loading point.

I would really love to see some data if you have anything related to an inline-4 (or even a 6 cyl). My main bearing plots make sense in that they are continuous, and all sum properly for typical crank configurations, but i have no idea what their shape needs to look like.

Thanks

 

[RossABQ]

As I recall, Triumph parallel twins were balanced to something real close to 50%. Not that they are particularly smooth...

 

[Kier14]

Hello twillcox,

From what I can understand, you are looking for the total force through a main bearing of the crankshaft for every degree crank angle. I didn't actually do this but I can show you what the force on the conrod should look like so at least you can see if thats correct. Hopefully it should be attached to this post but I'm not sure as I haven't done this before.

I can also try to attach the full report files as well but I'll see how this post goes first so I know what to do.

Regards,

Kieran.

 

[twillcox]

Ki Kieran,

I'm comfortable that we're getting the same results, see pic. Good work on that btw.

So...the next step was to calculate main bearing loads which is a sum of moments/forces assuming each main is rigidly supported and each cylinder contributes only to the two nearest main bearings. Do you have any main bearing force plots?

And the final step is to sum all the main bearing loads (and piston-to-block interaction and combustion gasses-to-head) to get the 'block' loads for vibration analysis.

I have results for both of these, but will format them to match whatever you might have.

an example...

with an inline 4 cylinder, 1243 F.O. at 10,000 RPM, I get a force ocillation of 16 KN inline with the cylinder (side-to-side forces sum to zero). The frequency is 2 cycles per crank revolution. Also, there are no net moments on the block (ignoring engine torque axis). The force vibration appears to be due to the secondary effects (primary's are 'perfectly' balanced as far as the overall block is concerned). Is this something you can verify? i know our numbers will be different due to crank counterweights and recip mass numbers, but they should look similar.

Thanks

 

[Kier14]

Hi Twillcox,

Yes those look like pretty reasonable plots.

I don't have any main bearing force plots as for my work it was not necessary, all I needed was the maximum and minimum stresses for Goodman diagrams.
For vibration purposes the sum of the bearing forces will equal zero anyway as will the primary forces, the second order forces which is what you are seeing will be 4 times as much as one bearing 'couple'. i.e they all add up for the crankshaft and are in the direction of the cylinder axis and have a frequency twice engine speed.

I know what you are trying to do but I haven't gone that far, one thing about what you are saying is that you want to work out a total block loads, ignoring cylinder pressure differences between cylinders, you will only have loads from second order forces which are not possible to balance unless you use a balancing shaft system. These loads will act along the cylinder axis so you wont have any side to side as you would in a flat plane V8. The other thing is if there are differences between cylinders I don't think this would show up as a force on the block but would actually show up as a torque on the crankshaft which would cause torsional vibrations in the crankshaft(I think). You will have to check this out however so don't take my word for it.

Let me know how you get on anyway as I would be interested to see further work on this subject.

Regards,

Kieran.