Crankshaft guided rods and centering on the journal

[ptuomov]

How do crankshaft guided rods usually locate themselves on a journal? V8, so two rods on the journal. In a running engine, will the rods settle with half the gap on the crank sides and a double the gap between the rods? If you think about oil flow and pressure without the crank spinning, that's what I'd expect. However, the crank spins and flexes, while the oil gets accelerated on the crank sides by the crank fillet. So anyone with any ideas how this all plays out in with crankshaft guided rods in a V8?

[MadBill]

My SWAG is that since the oil escaping between the paired rods is colliding with its mate and then turning a sharp right angle, whereas the oil exiting the outboard edge makes a sweeping turn between the crank fillet and rod ID chamfer, the latter will see less turbulence and higher flow than would either half of the center flow. I think even without any 'fillet benefit' for the outer flow paths, the total center flow would be no more than that of one outer, i.e., 33%. Factoring in the 'fillet effect' I'd guess the center path flow would be ~ 25% of the total. ASSuming uniform pressure at the entry to each, the average gap would be proportional to the flow, so ~38% of the total gap for each cheek and 25% between the rods.

[Geoff2]

I think there are numerous considerations. I often wonder when a rod brg fails in a V8 with side by side conrods, detonation is often blamed. But is that really the cause? Rod brgs also fail on low hp grocery getters. Why? If the rod is bent, the bore is not perpendicular to the crank c/line, the wrist pin hole is crooked in the piston. All of these singly or combined could 'work' the affected rod up against it's neighbour & block the oil from exiting the rod brg.

[Kevin Johnson]

I am guessing you have a CAD model of the crank and block. IIRC(??) you now have a bespoke crank and test models for new rods.

It has been ages since I bothered to look at my block. I believe the 928 rods are offset (see manual below) and that this was also tied in to the small end not being centered in the piston -- maybe the rod offset was needed to center it or some such.

Anyways, if the force vector is offset by virtue of the big end design then that would have a tendency to edge load the rod bearings which would be combated by the inner faces of the rods stabilizing/normalizing their attitude with respect to the journal.

As Geoff2 mentions -- lots of things to consider. I like Bill's analysis.

Porsche rod.jpg

[ptuomov]

I am guessing you have a CAD model of the crank and block. IIRC(??) you now have a bespoke crank and test models for new rods.

It has been ages since I bothered to look at my block. I believe the 928 rods are offset (see manual below) and that this was also tied in to the small end not being centered in the piston -- maybe the rod offset was needed to center it or some such.

Anyways, if the force vector is offset by virtue of the big end design then that would have a tendency to edge load the rod bearings which would be combated by the inner faces of the rods stabilizing/normalizing their attitude with respect to the journal.

As Geoff2 mentions -- lots of things to consider. I like Bill's analysis.

Porsche rod.jpg

I don't think it's that practically relevant, I was just thinking about it out of curiosity.

The basic design parameters for the bottom end are 122 bore spacing, 25mm bore offset, 54mm journal length, 52mm journal diameter, 14.5mm outside rod big end cheek, and 12.5mm inside rod big end cheek. Of course, clearances and tolerances.

[Kevin Johnson]

I am guessing you have a CAD model of the crank and block. IIRC(??) you now have a bespoke crank and test models for new rods.

It has been ages since I bothered to look at my block. I believe the 928 rods are offset (see manual below) and that this was also tied in to the small end not being centered in the piston -- maybe the rod offset was needed to center it or some such.

Anyways, if the force vector is offset by virtue of the big end design then that would have a tendency to edge load the rod bearings which would be combated by the inner faces of the rods stabilizing/normalizing their attitude with respect to the journal.

As Geoff2 mentions -- lots of things to consider. I like Bill's analysis.

Porsche rod.jpg

I don't think it's that practically relevant, I was just thinking about it out of curiosity.

The basic design parameters for the bottom end are 122 bore spacing, 25mm bore offset, 54mm journal length, 52mm journal diameter, 14.5mm outside rod big end cheek, and 12.5mm inside rod big end cheek. Of course, clearances and tolerances.

I guess you're good to go.

I just remember people arguing about it on Rennlist and schlepping out the block and observing the issue in situ.

[Kevin Johnson]

I looked in the shed with a flashlight -- the rods are offset to try to center the small end in the piston but are not completely successful. The early 944 also used a GKN rod but it was centered in the piston.

I looked in my files and I had scanned an OEM 928 GKN rod. Overall dimensions 84mm x 206.8mm. I showed the offset toward the web cheek at 2.529mm (rather than 2mm).

As mentioned in other posts, the center to center distance of the big end to small end is consistent across examples but the pair is randomly translated in the planar structure of the rod (off by .096mm along the minor beam axis and 1.001mm along the major beam axis). It was because of these variations that rods were stamped with serial numbers on the beam and cap.

[jed]

Interesting question my guess is and it's only an educated guess is they are float back and forth and bumping into
Each other. This floating or moving back and forth is caused by crankshaft rotation.
Kind of like ring rotation with the piston moving up and down and piston pins moving back and forth bumping into the pin keepers.

[ptuomov]

I looked in the shed with a flashlight -- the rods are offset to try to center the small end in the piston but are not completely successful. The early 944 also used a GKN rod but it was centered in the piston.

I looked in my files and I had scanned an OEM 928 GKN rod. Overall dimensions 84mm x 206.8mm. I showed the offset toward the web cheek at 2.529mm (rather than 2mm).

As mentioned in other posts, the center to center distance of the big end to small end is consistent across examples but the pair is randomly translated in the planar structure of the rod (off by .096mm along the minor beam axis and 1.001mm along the major beam axis). It was because of these variations that rods were stamped with serial numbers on the beam and cap.

I played around with some later stock cast rods and measured the rod big end offset parameters to best of my ability.

First, I stacked and clamped two rods from the big ends with the long cheeks on the center. Then, I measured the distance between the two small end outer edges. There’s a large air gap between the two small ends. The measurement result averaged about 55.60mm. This is not a super precise measurement, but close enough to government work.

Then I flipped one rod so the two rods are now "spooning". This leaves a small air gap between the small ends. Again, I measured the span. The results averaged 53.40mm.

Then I measured the small ends and big ends again separately. The big end width averages 26.96mm and the small end width averages 26.40mm.

Then I did some math, two equations and two unknowns. Variable y is the long cheek extension from the small end plane. The result is +1.40mm. Variable z is the short cheek extension from the small end plane. The result is a negative number at -0.84mm.

With these numbers, you get long cheek big end width from the beam centerline at 14.60mm and the short cheek width from the beam centerline at 12.36mm. If we take the reported bore offset to be nominal 25mm, we get 0.28mm running center clearance between rods when they are exactly centered under the pistons, which is reasonable. The journal length for one crank is 54.10mm, with the factory maximum tolerance at 54.13mm. There’s measurement errors there for sure, but I think these numbers are at least indicative.

As far as I can tell, everything is consistent with nominal design parameters of 54mm crankshaft journal length, 52mm crankshaft journal diameter, 12.5mm short center cheek, 14.5mm long side cheek, and 25mm bore offset, plus then tolerances and clearances around that.

[ptuomov]

Interesting question my guess is and it's only an educated guess is they are float back and forth and bumping into
Each other. This floating or moving back and forth is caused by crankshaft rotation.
Kind of like ring rotation with the piston moving up and down and piston pins moving back and forth bumping into the pin keepers.

Possible. But I think that there’s some oil pressure coming from the bearings that probably pushes the parts apart when they are about to close any of the three gaps. I think, I don’t know.

[ptuomov]

I think there are numerous considerations. I often wonder when a rod brg fails in a V8 with side by side conrods, detonation is often blamed. But is that really the cause? Rod brgs also fail on low hp grocery getters. Why? If the rod is bent, the bore is not perpendicular to the crank c/line, the wrist pin hole is crooked in the piston. All of these singly or combined could 'work' the affected rod up against it's neighbour & block the oil from exiting the rod brg.

I think but don’t know that rod bearings fail because of the oil supply problems mostly.

I also think that the “guidance” of the rods that keeps them from tilting sideways, whether piston or crankshaft, is only really operative at low rpms when lugging the engine. I’d think that at high rpms the rod wants to stay straight up and not tilt sideways.

[ptuomov]

My SWAG is that since the oil escaping between the paired rods is colliding with its mate and then turning a sharp right angle, whereas the oil exiting the outboard edge makes a sweeping turn between the crank fillet and rod ID chamfer, the latter will see less turbulence and higher flow than would either half of the center flow. I think even without any 'fillet benefit' for the outer flow paths, the total center flow would be no more than that of one outer, i.e., 33%. Factoring in the 'fillet effect' I'd guess the center path flow would be ~ 25% of the total. ASSuming uniform pressure at the entry to each, the average gap would be proportional to the flow, so ~38% of the total gap for each cheek and 25% between the rods.

I’d guess that since the bearing pressure causes the oil to escape to the side that is closer and this bearings exit the same amount of oil from both sides, the sides get 25% oil each and center gets 50%. Then, crank absent rotational effects, I’d guess that the clearance is distributed in same proportions. This is just a guess.

[PackardV8]

I looked in the shed with a flashlight -- the rods are offset to try to center the small end in the piston but are not completely successful. The early 944 also used a GKN rod but it was centered in the piston.

I looked in my files and I had scanned an OEM 928 GKN rod. Overall dimensions 84mm x 206.8mm. I showed the offset toward the web cheek at 2.529mm (rather than 2mm).

As mentioned in other posts, the center to center distance of the big end to small end is consistent across examples but the pair is randomly translated in the planar structure of the rod (off by .096mm along the minor beam axis and 1.001mm along the major beam axis). It was because of these variations that rods were stamped with serial numbers on the beam and cap.

I played around with some later stock cast rods and measured the rod big end offset parameters to best of my ability.

First, I stacked and clamped two rods from the big ends with the long cheeks on the center. Then, I measured the distance between the two small end outer edges. There’s a large air gap between the two small ends. The measurement result averaged about 55.60mm. This is not a super precise measurement, but close enough to government work.

Then I flipped one rod so the two rods are now "spooning". This leaves a small air gap between the small ends. Again, I measured the span. The results averaged 53.40mm.

Then I measured the small ends and big ends again separately. The big end width averages 26.96mm and the small end width averages 26.40mm.

Then I did some math, two equations and two unknowns. Variable y is the long cheek extension from the small end plane. The result is +1.40mm. Variable z is the short cheek extension from the small end plane. The result is a negative number at -0.84mm.

With these numbers, you get long cheek big end width from the beam centerline at 14.60mm and the short cheek width from the beam centerline at 12.36mm. If we take the reported bore offset to be nominal 25mm, we get 0.28mm running center clearance between rods when they are exactly centered under the pistons, which is reasonable. The journal length for one crank is 54.10mm, with the factory maximum tolerance at 54.13mm. There’s measurement errors there for sure, but I think these numbers are at least indicative.

As far as I can tell, everything is consistent with nominal design parameters of 54mm crankshaft journal length, 52mm crankshaft journal diameter, 12.5mm short center cheek, 14.5mm long side cheek, and 25mm bore offset, plus then tolerances and clearances around that.

I was building a full-race Studebaker V8, which has OEM offset rods, but using aftermarket Eagle rods which are symmetrical, but with narrower big ends. I started to do the measurements as you did, but the math made my head hurt. Finally, we just put it together, put it on the dyno. So far, so good.

[jed]

Having torn down Hunderds of high mileage engines there are always wear patterns on boath sides of paired V8 rods.
This leads me to believe that the rods are rubbing against each other and for sure there is oil from some where lubrication the sides of the rods.

[modok]

I'd expect on a V engine the rods would tend to be stuck together more often than not.