Piston pin boss stress

[HarleyE30]

Hi everyone,

Attached is an image of an initial piston design (bosses are huge so I can initially see where material can be removed). Also please ignore the irregular shape of the piston, it's not for a normal application!

The FEA is simulating the loading in the upwards direction (piston is upside down in image), which would be produced by the gudgeon pin, reacting to the gas force in the cylinder (80 Bar ). The load is applied as a bearing load in the appropriate direction

As you can see, the major stress area is the edge of the pin bore. This piston is aluminium (modelled as AL 6061). Max stress at this point is 101 MPa. Would this be a concern? Is the mesh too large perhaps?

Any advice appreciated, thanks.

 

[ivymike]

the mesh looks large, but 101 MPa sounds like a high load for an aluminum pin bore.

what is the average bearing pressure? your oil film will not support as much load out at the very edge.

Your edge loading will be compounded by the bending of the piston pin (gudgeon pin) - are you able to increase the OD of the pin to reduce bending and spread the load?

 

[HarleyE30]

I have taken the 80 Bar Pmax and turned it into a force by multiplying by the cylinder bore area. This comes to 46400 N.

I have then applied HALF of this to the pin bore as a bearing load acting in the appropriate direction (so 23200 N); thus assuming each boss takes half the load.

Is this the correct way of looking at it?

There is indeed provision to enlarge the pin.

Many thanks

 

[ivymike]

that sounds fine for a first look. if you take 23200N / projected area of pin bore, what do you get? I would normally want that figure to be <55MPa.

it has been such a long time that I can't remember what a reasonable guideline for pin bending is - but I suspect you'll be able to reduce your edge load a fair bit with only a small change in pin diameter.

 

[ivymike]

rusty gears turning ... 80um pin bending rings a bell, for pistons about 100mm diameter.

 

[Lou Scannon]

Evidently either the pin or the piston is flexing to produce the localized stress at the edge. See if you can figure out where the flex is occuring, and take measures to reduce it.

"Schiefgehen will, was schiefgehen kann" - das Murphygesetz

 

[HarleyE30]

I have worked the projected area out to be 2349.9 mm^2 (bore length = 23mm and Dia = 22mm)

I then divide this by two, as the force will only be acting on one half of the bore, giving me 1174.95 mm^2.

Force / Area = 23200 N / 1174.95 mm^2 = 19.745 N/mm^2 = 19.745 MPa.

Why then does the FEA suggest values much higher than this - as can be seen in the image ?

 

[ivymike]

that should be easy - the load is not evenly distributed?
29MPa is a reasonable gross pressure value.

 

[HarleyE30]

The force applied in the FEA is a 'radial bearing load' across the top half of the surface area of the bore. I assumed this distributed the load evenly. Does the difference between the FEA and hand calculation arise from the fact it is cylindrical ?

 

[Tmoose]

How are you modeling the wrist pin, and the wrist pin to piston boss interface?

 

[HarleyE30]

The pin isn't being modelled right now. I'm simply applying a load to the upper face of the pin bore

 

[Tmoose]

Hi HarleyE30,
"I'm simply applying a load to the upper face of the pin bore"

So the entire pin boss bearing surface is restrained or grounded?

 

[HarleyE30]

The model is constrained on the outside faces of the piston (cylindrical constraint).

 

[Lou Scannon]

I'm not an FEA jockey so take this with a grain of salt, but it seems to me that, contrary to your assumption, your model is NOT applying an evenly distributed load.
In any case, that would be erroneous, due to flexing as mentioned above.
I'd recommend you model the piston pin, and apply the load in some kind of realistic distribution near the centre of the pin, in accordance with the geometry of the connecting rod.

"Schiefgehen will, was schiefgehen kann" - das Murphygesetz

 

[ivymike]

Hmm...if there is no pin in the model, then the red bits can't be from pin bending, can they? Sounds like you need to work on your loads and constraints, then come back if a reasonable model still gives troubling results,

 

[HarleyE30]

"Hmm...if there is no pin in the model, then the red bits can't be from pin bending, can they? Sounds like you need to work on your loads and constraints, then come back if a reasonable model still gives troubling results"

Yes this is what concerns and confuses me. I will look into it some more. Many thanks all for taking the time to advise me

 

[pontiacjack]

I'm curious- is the 80 bar gas pressure a real-world typical number?

 

[Lou Scannon]

80 bar is certainly a real world number; might even be a peak pressure for a normally aspirated Otto engine. Boosted engines will typically see peak pressures higher than this.

"Schiefgehen will, was schiefgehen kann" - das Murphygesetz

 

[ivymike]

It is not unheard of to have peak cylinder pressure >200 bar on a diesel, although 170-185 would be more common.

 

[ivymike]

Ignore that link-not sure what happened there (phone post)