Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Register Log in
  • Congratulations KootK on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Connecting rod design 2

Status
Not open for further replies.
Aled87

Aled87

Automotive
Mar 1, 2010
24
Hi,
I'm looking for a good book on connecting rod design, i.e. forces, fatigue stress, optimisation, etc. Does anyone know if there are any out there? some guidance would be appreciated

Many thanks,
 
Replies continue below

Recommended for you

Sort by date Sort by votes
"if they are heavy enough and made of 4340 or some other superior steel......"

Back in the 70s Chevrolet produced a semi-finished rod ( K rod) based on the big block forging for small block endurance racing purposes. It's bulk required a lot of clearancing to miss engine innards. It was made of some mundane 104X steel I think. Good Geometry can trump exotic materials. Even exotic materials can not compensate for a botched design or manufacturing details in a cool looking design.

Seems like modern rods recognize the "footprint" around the rod bolts at the parting face needs to be big to make the bolts' life less bendy, and the housing bore needs to be bulky in the right way to keep the hole round at high rpm if the bearings are to survive. Checking off I/H/X on the Design sheet does not go far accounting for that.

In street use Harley big twin forked rods were almost certain to slowly develop cracks inches away from the I-beam, where the thickened rib meets the side panels where the races are located.
Some subtle re-shaping combined with shot-peening pretty much eliminated the problem.
 
Upvote 0 Downvote
briangar, good point about the big ends on H-beam rods. The natural design of big end on an H-beam rod gets rid of the through bolts and the troublesome notch that were standard. Also, the ribs coming down from the beam tend to surround the bolt hole, another good point. I think the overall shape of the H-beam leads to the best shape for a big end and it is now the common shape for the best I-beam rods as well.
Steam engines and F1 use I-beam rods. It's the in-between racing applications where there is disagreement.

Tmoose, the forked rod looks like the hardest rod to design and avoid stress risers. I like this solution by Daimler-Benz: <You can see that rather than have a fork, they split the I-beam in half and kept them separate right up to the piston pin. What is not seen is the webbing that joins the I-beam halves, turning the cross section into an H with feet. Don't know if this DB601 rod inspired Taylor at Wright Aero to try an H-beam master rod after the war, but it might have.
 
Upvote 0 Downvote
The next problem lies in what you call an I beam, or H beam. I can think of many rods that I would call in-between - what exactly differentiates one from another in design terms anyway?

The general feeling I have about both, and I have studied them lots, is depending on how much time and money you spend on either I or H the better that rod will be. There is no superior one as such.
I feel that many of the big names have just spent more time with one design and have taken them to the limits, for them to change now would involve starting over - more money.
I think you'll also find it hard to see an I beam in any current F1 engine now, but I could be wrong.

Below are what I call in-between rods - when you have them in your hand its even harder to decide,






Brian.
 
Upvote 0 Downvote
BrianGar, I think the most basic definition of an I-beam as an engineering element is exactly the same as an H-beam. It is a beam that has two flanges separated and connected by a web. The strength of the beam against bending comes from the separation of the flanges and is asymmetrical, when the proportions are like an I, being stronger along the direction of the web. The I-beam rod aligns the web with the plane of rotation where the bending moment is likewise asymmetrical, being 4 x greater in the plane of rotation than it is in the plane of the pin axis at zero rpm. The H-beam rod has the web turned 90 degrees to align with the pin axis and the separation of the flanges is minimal. The H-beam then has the flanges elongated to gain area moment and strength in the plane of rotation.

To assign a type to a rod with characteristics of both types, look at the web and flanges. The Ferrari rod you show is an I-beam rod with its web aligned with the plane of rotation, but with small ribs along the edges of the flanges. This Cosworth rod (from Tmoose) is more ambiguous: but just look at the web of the rod. It's an I-beam with even more pronounced ribs. ALL F1 rods are I-beam rods because F1 engines run in an rpm range where they can no longer afford the extra weight the H-beam requires to match the strength of an I-beam. I elaborated on this in an earlier post. See above.
I believe that the H-beam rod excels at distributing stress throughout the rod while the I-beam requires more careful design to avoid stress concentrations.
 
Upvote 0 Downvote
I think its pulling at straws to call the cosworth/pankl rod an I beam to be honest. I know very well what these rods are like in cross-section(thats one of my pictures above).

Brian.
 
Upvote 0 Downvote
 
Upvote 0 Downvote
BrianGar, I have read these discussions before in researching this topic. I rely a lot on the papers describing connecting rod design. I have not found anything so far, including these discussions, that describes any theoretical support for the H-beam rod except the brief mention by Charles Fayette Taylor. It would help if you could point out any particular post in the discussions you reference that you feel is definitive on the point.

I characterize the Cosworth rod and any rod by the distribution of of mass. The I-beam/H-beam separates mass into flanges connected by a thin web. Ideally, the flanges contain most of the mass and the web contains the minimum amount of mass possible. It is THIN. The H-beam rod web is a thin span running straight down the middle of the rod when viewed flat. The I-beam's web is a thin span running down the middle of the rod when viewed from the edge. The Cosworth rod has a web in two parts, under the piston pin eye and above the crank pin eye, that fits the I-beam description. There are features in the Cosworth rod and some other F1 I-beams that are H-beamish. Also, the Cosworth rod is the most solid I've seen. I already mentioned that the strongest I-beam rods have big ends that were originally more typical of H-beams. Also, the Ferrari I-beam rod has ribs along the edges of its flanges (BTW, do you think it is an H-beam?). There are more pictures of recent F1 rods out there, -ALL I-beams.
In this SAE paper by Mario Illien of Ilmore-Mercedes he makes it clear that Ilmore doesn't use H-beam rods in F1 (I think they do use them in other areas): < Note that this paper usually costs money to see, but is free at this site. Illien does not describe the critical pistons and rods except to say that the pistons are the driver of the design and the rods are "...I rather than H...".

Illien's words are no surprise in view of Euler column theory (check out column theory for pinned columns). Also, column theory is for the static case, zero rpm. As rpms increase, additional inertial forces come into play that increase the bending moments in the plane of rotation and makes the I-beam even more preferred. The universal use of I-beams for connecting rods makes all the sense in the world... or maybe not.
The beam portion of short rods is less than 1/2 the rod's mass and the transitions from the I-beam to the pin eyes are areas of stress concentrations and differing opinions by designers. The FEA stress analysis I have seen all show the H-beam to be a more stress-free shape. I conclude that the justification for H-beam rods lies in the way the rod actually performs with respect to durability (in applications that can tolerate some extra weight) and not in column theory or strength-to-weight.
I am looking for more and especially differing analysis.
 
Upvote 0 Downvote
FWIW... Less of an engineering and more of a practical over abuse point of view: My dyno guy does lot of nitrous motors with high cylinder pressures. His rule of thumb recommendation based on broken parts is preferably full round skirt pistons and I-beam rod, at least one or the other. He has sideways bent H-beams used as wall decoration. Sideways meaning toward the unsupported side of slipper skirt piston. He said they failed when spraying ~500hp, which may be a little theoretical in many cases.
 
Upvote 0 Downvote
Calypso, very interesting info. I also think a full round skirt is called for in high boost or nitrous situation. Don't know if I understand about the H-beams bending sideways. I expect them to bend in the plane of rotation with the flanges wavy, if that is sideways. They look extremely strong the other way, the flat way.
 
Upvote 0 Downvote
The advantage of a full round skirt is that it dissipates hat better and spreads cylinder wall loads a bit better. Really important features in a very high cylinder pressure application.

However I really don't see it influencing rod bending and actually strains bolts and big end bore more due to increased weight.

A heavier piston shifts the favour more toward the H beam as they typically support the big end better and reduce stress concentrations near the bolt holes,

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &
for site rules
 
Upvote 0 Downvote
Interesting. I guess Doug needs to do some more work. In order to see the link, you must highlight the url and click "open link". Then it opens fine.

Rod
 
Upvote 0 Downvote
Pat, I agree that the extra weight is not good, but the concern was about out of column forces, which occurs when the piston has no stability along the pin axis. Ordinarily, the rod can take whatever side loads exist, but Taylor cited this in recommending the extra strength along the pin axis of the H-beam. A boosted engine and/or one experiencing detonation would tax the rod worst. A full-round skirt should take all out-of-column compressive loads off the rod. The question would arise, which increases reciprocating weight the most, a heavier rod or a piston with added skirt area? I notice that Top Fuel and other top drag racing classes seem to use full-round-skirted pistons. F1 does not.
 
Upvote 0 Downvote
I certainly agree that cars like blown alky or nitro or nitrous benefit from full round skirts. I always put it down to heat dissipation.

My point was that the desirable full round skirt weighs more. This extra weight comes into play over TDC and stretches bolts more and pulls big ends into an oval shape more. H beams better resist that.

Top fuel uses aluminium rods anyway so the point becomes mute as aluminium rods are almost a solid rectangle beam.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &
for site rules
 
Upvote 0 Downvote
I have done some 3D modelling and FEA comparing H, I and X rods and can say this generally.

For the same peak tensile stress under tensile load the I-Beam is a lighter more effiecint design by a noticeable amount. This mean if you want to avoid peak stress for fatigue an I-Beam is better due to the stress concentrations being lower. For high rpm repeated tensile applications not involving complete overload (yield) the I-Beam will be lighter and therefore better suited.

For the same mass both an H and X have more beam CSA which means ultimate strnegth driven by plastic crippling will be higher than an I. So the I has lower peak stress but global stresses in beam are higher.

I would say the X-beam is suited for extreme tensile loads where you wont see too many cycles

Assuming rods of same mass when you use Euler buckling of the I-Beam about its best axis the force levels seem to be high enough that the beam would have gone plastic so elastic buckling is not a big concern . For the weak axis the end conditions are not pinned and so the same applies for the rods it wont be elastic that i looked at. The same can be said for H-beam while its strong axis is less strong it seems plenty and its weak axis is stronger than the I weak axis. For the X-Beam about its strong axis is much same as H-Beam about its axis and about its weak axis is much same as I-beam about its weak axis,

When comparing rods it is quite hard to speak generally but i think what i have mentioned is a pretty good comparison. To go further one would need specific geometry, application and loading to compare.


 
Upvote 0 Downvote
inline6, This statement puzzles me.
"For the same mass both an H and X have more beam CSA which means ultimate strnegth driven by plastic crippling will be higher than an I. So the I has lower peak stress but global stresses in beam are higher."

Assumptions I have used are that when machined from solid forging, the contours of the big and little ends would be the same for all types. The only difference would be the shapes of the beam and the transition regions. I also assumed either the beam cross sectional area (CSA) must be identical in all types for the same tensile strength OR the bending strength in the rotational plane must be identical for the same buckling resistance. Then the comparison would be the strength to weight ratios.
When you say the H and X have more CSA at a given weight you imply that the parts of the rods other than the beams are heavier in the I design. If the ends are identical, then the differences would have to be in the transition regions. Is that what you mean?

Also, an unfortunate fact is that I-beam rods vary quite a bit in the transition regions with some much heavier than others in those areas. Did you model a particular I-beam rod?

Lastly, did you make any comparison in the torsional stiffness? This factor escapes my simple analysis beyond noting that the I has the highest overall area moment while it appears the X has the lowest. But, I believe details of the structures should come into play, like the length/thickness of the web in the cross section and the filet radii.
 
Upvote 0 Downvote
Open sections like Is, Hs, and Xs rely on the thickness of the webs and flanges.

If for some reason you want great torsional stiffness, "boxed rods" may be the way to go.
They were used in various hot-rodded car and bike engine decades ago.
Fuel burning harley dragsters needed some help in regards rod buckling, and boxing DID offer some relief

half way down the pager here, right hand example hard to see.


I'd consider Welding on a con rod anywhere, but especially at the transitions, a really bad idea.
 
Upvote 0 Downvote
Tmoose, these are excellent references that I had not seen before. Thanks.
Yes, boxing was used before there were so many good aftermarket pieces. And, a boxed or oval section is indeed much better in torsion than an open section. Twist is important because it appears to sometimes be a factor in the buckling process. A column with asymmetric strength can fail prematurely if twisting can partially expose the weaker direction to the bending load. The I-beam is the most asymmetric of the three we were discussing.

inline6, what you say about the X-beam being like the H-beam in the strong direction (weaker than the I-beam) and like the I-beam in the weak direction (weaker than the H-beam) is what I expected. This follows if you look at the concentration of mass in the X-beam straight down the center of the rod where, except for tensile strength, it is dead weight. It therefore is the weakest design of the three for the weight, ...except that I believe an X-beam is actually stronger ultimately than the H-beam in the strong direction because, IMHO, its thick blade is not subject to the flange buckling of the two thin H-beam blades.
Your analysis is very helpful.
 
Upvote 0 Downvote
140AirPower when i used the same big end and small end with same outer profile for the same mass the I-Beam had slightly thinner flanges and web and thus a slightly smaller cross section at the narowest point. The reason is as you say the added mass in the transition regions of the I. This is also why the stress concentrations are lower in this areas. If i adjusted the I-beam web and flange thickness to the same CSA as H-beam it worked out about ~25g heavier in a 600ish gram rod which is merely a few %


with further development i think i could design an I-Beam rod to have the same min CSA in the beam area for the same mass which would mean the I's would have it.

It is quite difficult to compare them you could do

1) Same mass what is the strength comparison
2) Same strength what is the mass comparison

but what does stength mean? fatigue, bending, axial, proof load, gross collapse, fracture........)

but the most noticeable thing that came out was the I-Beam transition was far superior than the other two which is import in tension. I did not look at doing a pin contact analysis to see whther the loads are carried closer to the edge of the bearing i used a sinusoid pressure around a suitable arc uniform through thickness

 
Upvote 0 Downvote
inline6, and how does an I beam compair if it uses the same for and aft profile that an H has?
That is for or aft installed in an engine. I beam rods are most always a smaller profile than an H. Lighter yet stronger as well.
 
Upvote 0 Downvote
Status
Not open for further replies.

Similar threads

Ibrahim M
  • Locked
  • Question
Racing turbocharger design
Replies
10
Views
84
TugboatEng
TugboatEng
VineetKujur
  • Locked
  • Question
You are designing a vehicle drive t 3
Replies
3
Views
71
enginesrus
enginesrus
Rat5
  • Locked
  • Question
Cooling system design, mass flow rate ? 1
Replies
2
Views
132
romke
romke
Rat5
  • Locked
  • Question
Turbo inline 4, exhaust design sound
Replies
6
Views
76
EdStainless
EdStainless
enginesrus
  • Locked
  • Question
Connecting rod loading 1
Replies
17
Views
151
inline6
inline6

Part and Inventory Search

Sponsor

Back
Top Bottom