piston expansion measurements

[acannell]

measured some 2.5 stock .030 over cast Mahles

used a water bath to get the piston temperature evenly distributed

measured at same spot on skirt

at 68F 3.4701"

at 139F 3.4720"

so a 71F difference in temperature

and a 0.0019" difference in diameter

or about 7.7 uin/in/F

2618 forged alloy has a linear expansion ratio of ~12 uin/in/F

Not sure what this cast alloy is supposed to have as far as expansion ratio. Or maybe I am sure since I just measured it lol.

Whats interesting is that the bore measurements I made put the bores between 3.4700" and 3.4750" depending on location.

At this expansion rate the piston would have to get up to 248F to reach 3.475"

Of course the block is also expanding, but less so, so the "seize" temperature is probably much higher than that.

Now it all makes sense...

[acannell]

so..where WOULD piston and bore thermal expansion end up resulting in zero clearance?

if we use 6 uin/in/F for expansion rate of cast iron, and 7.7 uin/in/F for the cast Mahles, it looks like 389F piston temperature would be zero clearance

this assumes:

bore dia @ 70F = 3.475"
bore operating temp = 240F
bore expansion rate = 6 uin/in

piston @ 70F = 3.470"
piston expansion rate = 7.7 uin/in

If we change the piston expansion rate to 12 uin/in we end up with 275F piston temperature = zero clearance.

Now throw in the following things and you can see how installing forged pistons can equal problems pretty easily:

-mfg. tolerance of forged pistons +0.001"/-0" (Venolia)
-not compensating for the bore diameter measurement temperature if you measure at a much higher or lower temperature than piston measurements
-machining tolerance of bores
-measurement error in bore measurements
-specifying too low a wall clearance by even 1 thou

Theres only a budget of say 5 thou in this whole thing..if all the above errors tip it the wrong way you can easily end up with zero clearance at normal operating temperatures or way too much clearance.

[acannell]

okay did another test, but this time with a non-cast aluminum part

the closest thing I had laying around was some nearly solid cylinder of aluminum, about 3.232" diameter

im not sure exactly what alloy it is, but its unlikely to be cast given what it is (an old injection mold that was machined)

same test as before

67F = 3.2323'
135F = 3.2354"

change in diameter = 0.0031"

change in temperature = 68F

so to get inch per inch expansion per F:

0.0031 / 3.232 / 68 = 14 uinch / in / F. Compared to the cast piece at 7.7 uinch / in / F. Thats quite a difference!

[acannell]

hmm so I suppose this can also be applied to piston rings

if we assume cast iron

and say 3.474" OD with a 20 thou gap at 70F

that means the outer edge of the ring has a circumference of 10.888"

so at what temperature will the ring length increase by 20 thou and close the gap?

.020 / .000006 / 10.888 = 306F + 70F = 376F

but the bore is also expanding..at the same ratio, but its colder and its temperature is most likely kept in check around coolant temperature somewhere

so if we assume the bore expands from 3.474" to whatever it ends up at at say 240F, which is 3.4775, now how hot does the ring need to get so its ends are touching?

3.4775 circumference is 10.919"

our ring has a length of 10.888 at 70F

so our ring needs to increase in length by about 31 thou

.031 / .000006 / 10.888 = 474F + 70F = 544F

[zin]

The cast pistons will not expand in the way solid AL would, as they typically have a steel "strut" cast into them to help them hold their shape/expand less.

Also, the temps are going to be higher, and uneven, with the center of the piston highest (longest heat path) and the lands/skirts the coolest.

It's a tough enough thing to figure mathematically that some teams will hone the block, with not only a deck plate to simulate the cylinder head's stress, but will also run coolant/oil through the block to reproduce the same distortion seen while running, thus ensuring as round a hole as could be hoped for...

Mike

[acannell]

The cast pistons will not expand in the way solid AL would, as they typically have a steel "strut" cast into them to help them hold their shape/expand less.

Also, the temps are going to be higher, and uneven, with the center of the piston highest (longest heat path) and the lands/skirts the coolest.

It's a tough enough thing to figure mathematically that some teams will hone the block, with not only a deck plate to simulate the cylinder head's stress, but will also run coolant/oil through the block to reproduce the same distortion seen while running, thus ensuring as round a hole as could be hoped for...

Mike

yeah this is all just to ballpark things..without knowing what temperatures things get who knows whats happening in there. Empirical data rules here.

but I think it does show that things are pretty tight...theres not a whole lot of clearance as far as piston to wall and ring gap, and it seems reasonable that seizing temperatures could be reached at well-beyond-stock power levels.

It also shows how important coolant temperatures can be. At 400hp how much power is lost thermally directly into the piston compared to stock? Twice as much? 3 times as much? And how well is the cooling system dealing with that?

Maybe all that matters is how well things go for 8 to 13 seconds....

[acannell]

what the heck...

look what I just stumbled upon on nasioc

has anyone else seen these before or know who this guy is? maybe he has more juicy chrysler info

"here's an orthographic view of temperatures at WOT 4800RPM in a Chrysler 2.2L engine that I happen to have lying around:"

http://forums.nasioc.com/forums/show....php?t=1061799

The same document, from which I got that, lists the thermal conductivity coefficient for various parts of the piston as a function of crank angle under the same conditions. It's pretty neat stuff

What document????

[DetroitTurbo]

So when the block expands, the bores get bigger? I would think they would get smaller if the block is swelling up.

[going4speed]

The hole does get bigger as the engine heats up. Think about all the molecules in a circle. They all get bigger as they heat up. No way for them to take up less space so the hole gets larger.

[acannell]

So when the block expands, the bores get bigger? I would think they would get smaller if the block is swelling up.

one way to imagine it is like this: imagine a ring, say 4" OD and 2" ID..so its got 1" thick walls at room temperature

now we agree the OD of the ring gets larger, as I measured right?

well now, before you heat it up, scribe concentric circles along the wall on one of the flat ends

now measure at those marks instead of the OD. obviously they too must increase in OD as it heats up.

now just imagine that the ID is the "mark" you measure at instead of the OD or the scribed marks....

[Reaper1]

Good info here....subbed!

[acannell]

fun experiment for anyone with one of our blocks sitting around and a dial bore gauge.

let the block stabilize at some temperature..say room temperature (or whatever your outdoor temperature is overnight)

make the typical 6 bore measurements per cylinder (perpendicular, parallel with crank, at 3 heights)

then submerge it in a bathtub of hot water for 30 minutes

measure again

see what kind of distortion and diameter changes there are.

also measure the temperature in both settings