Connecting rod design 2

[Aled87]

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,

 

[Tmoose]

What are you making ?

MIT Press, Taylor, "internal combustion engine" I forget whether volume 1 or 2, has a section on con rod design. It's kind of an industry survey type book, last updated a few decades ago, but has some theory and examples, and usually many bibliographical references.

If you are considering PM (pressed metal) designs the SAE transactions from the 90s and 00s have a lot to say.

 

[140Airpower]

Aled87, did you find satisfactory answers to your question? I have been seeking the same information. There are many references available on line, however I have found nothing definitive and authoritative that answers CF Taylor's recommendation found in the volume mentioned by Timoose. Taylor's opinion contradicts standard practice for the highest performance rods by OEMs and is given without support. Yet it is relevant to popular aftermarket designs.

 

[140Airpower]

Tmoose, do you know of any "authoritative" sources that can resolve questions about aftermarket rod design? To me "authoritative" means other than the ad copy put out by aftermarket rod manufacturers.
All papers I have seen describe the design and verification of standard I-beam rods. It appears no OEMs or serious design houses consider any of the common aftermarket types, like H-beam, X-beam, etc.

 

[Tmoose]

No authoritative sources known.

Some successful airplanes have been built of wood. They don't look much like airplanes made of steel tubing or aluminum or XXX

There was a time when there were articles in Machine Design and Hot Rod magazine and all over the place about connecting rods and other engine components made of epoxy/whatever fiber composite materials. Google can only find an occassional dreamy patent for one now.

I bet successful "cutting edge" connecting rods made of steel or titanium could and have been made using H and I beams, and even tubular shanks (old Offenhausers).

I'm not sure the beam choice has one answer, or that we really get to see what is currently in vogue, or even that con rod beam design is being scrutinized as a frontier limiting engine design in any class of racing.

I happened to be in a position to MagnaFlux inspect lots of car and bike connecting rods in the period 1973 thru ~1990. Some vintage designs, and some genuine race engines derived from production cars-n-bikes and also some purpose built, but not Formula 1.
Cracks (indications!) generally appeared in the transition between beam and wrist pin eye, and especially in the area around bolts at the big end transition. In the rare case when a crack developed in the beam itself, there was always a significant excuse to initiate in the form of a stress riser. Might have been a knick, scratch, stamped number (ouch), filed notch, or once in a great while an original manufacturing flaw like a forging seam or lap.

My conclusion is the beam shape with "best" bending strength should and does lose to the shape that permits the "best" (lowest stress, most manufacturable) transition to pin and big ends.

In 2000 Ferrari may have chosen an I-beam derivative for reasons of their own.

http://www.arteauto.com/images/products/display/Ferrari_Piston_dec2011_332.jpg

But here's a fairly serious engine (Cosworth TJ V10 ca 2003) with a sort-of H-beam con rod.

http://i150.photobucket.com/albums/s92/chippievw/General motorsport/cossie2.jpg

 

[140Airpower]

Tmoose except for wooden airplanes what you are saying are some things I've read before. What is the most interesting is the question of stress concentrations and reliability. These are the only factors that appear to argue against I-beam rods.
From what I've read, rod design begins with Euler column theory for pinned columns and works from the pinned column's 4 times the bending moment in the plane of rotation vs the axial plane to arrive at an I-beam with its strong axis toward the plane of rotation. To vary from that (except for hollow beams) would seem to inevitably result in a rod with inferior strength to weight. As you say, there are few racing applications where extra weight in the rod is a critical disadvantage, but F1 is one of those areas. Consequently, you see only I-beam rods in F1 (except that some I-beam rods have been made with H-beam like ribs, including Ferrari rods -later than the one you show, as well as the Cosworth rod you show).
However, in the FEA stress analysis images I have seen, the stress distribution in I-beam rods looks complicated where the big and little ends join the beam. Stress in the same areas in an H-beam rod looks a lot simpler as you might guess from the smooth shape.
I noticed that CF Taylor talks about stress risers as one reason to resort to the H-beam, but none of the aftermarket makers talk about it nor anybody else.
The other popular beam type is the X-beam, which seems to have nothing to recommend it vs the I-beam, but which is probably stronger in the plane of rotation than the H-beam.

 

[Tmoose]

I think if I asked my connecting rod bearings to restrain the con rod from bowing a little by behaving like a pinned connection with about 0.002" clearance at around 12 noon of the power stroke or either BDC the the bearings would be unhappy with the edge loading. In time they might even look look like this.

http://www.substech.com/dokuwiki/li...che&media=bent_or_twisted_connecting_rod1.png

Modern pistons don't have much skirt and length on the sides, so I wonder how much lateral rod guidance they can offer.

It's a little hard to see how spindly the I-beam is, and how stout the big-end is on stock Honda RSX rods.

http://www.kseriesparts.com/graphics/00000001/HON-13210-PRB-A00-1_665x750.jpg

 

[patprimmer]

I also have never seen anything really definitive on either design type and tend to think there is more importance in specific detail than design type. Also the type of application may determine failure mode.

I know first hand the I beams in boosted OEM Honda D series engines tend to bend in compression at around the 220ish hp area. They don't break, just bend.

The same rods when used in a short stroke high rpm NA engine tend to fatigue and break as the failure mode.

I think a lot of favour was placed with H beams with the advent of the Carrillo brand rods which became very famous for increased durability at much higher loads than OEM rods. Initially virtually all serious after market racing duty rods where Carrillo or Carrillo copies and it was widely believed that the H beam design was the reason for their superiority where in fact it was more just extra metal where required, precision of manufacture and the use of superior materials and treatments.

I was always of the unsubstantiated belief that the reason Carrillo chose H beam and OEM chose I beam is that in the day and age, I beam suited manufacture by forging and H beam suited manufacture by milling and that rather than strength or durability inherent in either design was the reasons behind the different choices.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[140Airpower]

Tmoose, what you say about the lateral stability is very instructive. The I-beam rod is by far at its weakest in the axial plane, yet stock rods are never made as thick as the space would allow. Also, piston skirts could be circumferential, but usually are not. Both these facts tell us that the axial loading is light. And, if more strength and less loading are wanted, the rod could easily be twice as thick and a full circumference band can be added to the piston. Generally, what I have seen of I-beam rods is that if they are strong enough in the plane of rotation, they are extra strong in the axial plane. In fact, though the bending moment in the plane of rotation is 4 times as great for the static condition, the custom is to make the rod only 1/3 as strong (rather than 1/4th) in the axial plane to give an extra margin of strength there since you would prefer the rod to buckle last in a way that it would contact a crank cheek. The spindly proportions of current highly optimized rods is evidence of the state of design, using much less rod for the same margin of performance.

 

[140Airpower]

patprimmer, your opinions here are what I have heard and they make a lot of sense.
I also suspect that Fred Carrillo believed the H-beam form was more trouble free as a design compared to the complexity of the I-beam. All one had to do was to look at the wide variety of I-beam designs to realize that the engineering was ambiguous. Carrillo made a rod that was much stronger than OEM rods and was of superb quality. I believe the H-beam's reputation was made by this and its inherent freedom from fatigue failure associated with stress risers. This last point, reliability, is essential.
H-beam rods are probably a safer choice for racers except when compared only to I-beam rods from the very top makers, like Pankl and Carrillo themselves or the OEMs when they offer a truly strong rod.
In fact, it is telling to see that Carrillo recommends their light I-beam rods (called A-beam, a marketing name) for high RPM applications where weight is critical. One would expect they would sell you a light weight H-beam, but by recommending the I-beam they are confirming that the I-beam is stronger by weight.

 

[dicer]

I have personally never ever seen an overhead crane use an H beam.

http://www.inamar.cl/ingles/e_levante.html

The problem I see is, you never seem to see an I beam rod, formed similar to the H beam rod, that is viewing it from front or back inline with the crankshaft centerline. The I beam rods are always much narrower than an H beam rod. If designed correctly the I beam rod is much stronger than an H beam rod.

 

[patprimmer]

Personally I never saw an overhead crane being used to connect a piston to a crank so what's the point.

Where rod choice is free, one heck of a lot of high budget endurance races have been won with H beam rods

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[140Airpower]

patprimmer, the use of H-beams in high buck racing is undeniable. It is this fact that makes this discussion interesting. I think the analysis from principles, without doubt, favors the I-beam and that this is born out by the F1 example and the OEM preference where the OEMs design expensive, fully machined rods for their racing engines (not their cheap, as-forged rods). But the aftermarket racing fraternity regards H-beams very highly.
For OEMs using H-beam rods, I know of two examples. Ford bought an off-the-shelf Crower H-beam rod for one of their turbo engines a few years back. Audi used an H-beam rod in their recent (low rpm) turbo-diesel LMP car. This rod is steel with a steel piston. Also, the piston lacks a skirt section in the axial direction.
Furthermore, H-beam rods were once common in F1 before the extreme RPM era. According to Ilmore's Mario Illian, the perfection of pneumatic valve springs around 1994 took valves out of the equation as an rpm limiter in F1 and made pistons (and rods) the the power limiting parts. According to BMW's Dr. Theissen, H-beams are "...much too heavy...", so the preference has been for "...I rather than H...", according to Illian.
But, in classes of racing where metal valve springs are used, the weight of H-beam rods is less critical and their virtues are either known or reputed. Which is it, known or only reputed? That is what I'd like to know.

 

[patprimmer]

140

I agree, however I disagree that a crane is a good comparison and that I beams are much stronger. A bit stronger sure, but much stronger as a blank statement with zero supporting data, give me a break.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[140Airpower]

patprimmer, I don't think I-beam rods are "much" stronger for the weight since the beam portion of the rod is not the majority of the weight. I don't know what extra weight of H-beam rod is needed to match the buckling strength of a particular I-beam rod. You can do a calculation of the area moments in the direction of the plane of rotation for equal total areas and the same section outline. I think the I-beam may show 50-80% better area moment. Could be less. The heavier both beams are, the more they converge toward a solid beam. What that would mean for the extra weight of the whole H-beam rod I don't know, maybe 20%? That may be a lot for F1, but not so much otherwise. Another factor is that the extra section mass the H-beam needs lowers its stretch at TDC, lessening fatigue and increasing its reliability. So, the extra weight is not a total liability, so long as the crank and block are up to it.

 

[patprimmer]

I never saw actual figures but 20% sounds a reasonable.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[evelrod]

Just my two cents...I never used Carillo until recently in our newest engine. I found them "interesting" in appearance and a bit on the heavy side compared to the Cosworth rods that I have used for many years. In fact, quite a bit heavier. As to reliability...well, come on guys, racing is rather stressful on even the best parts and after all, parts is parts. I've had Cosworths fail and I suspect there has been the occasional failure in the H beam Carillos. Matters not, since the light Cosworths are not available these days (and probably unaffordable, anyway) I have been trying other options. Currently I am using a set of Pauter "X" beam rods. One thing that has come up in two engines is the designed in "crush" of the bearing shells. On both cranks it was necessary to polish out an additional 0.001" clearance than when using Cosworth or Carillo (or stock, for that matter). This link is just to show the rod and not any engine I build.

http://www.pauter.com/acura_honda.htm

No failures in either rod type so far. The Carillo rods are in an engine developing 0.13+ hp/cc at 9500 rpm and the Pauter are in a smaller engine @ 0.10+ hp/cc and 8000 rpm. Apples and oranges...vintage race engines, not F1---

Rod

 

[BrianGar]

A short bit of light reading on different conrod designs >

http://www.faliconcranks.com/Making the Connection.pdf

Brian,

 

[140Airpower]

BrianGar, That article and marketing claims by several aftermarket manufacturers are offered with no engineering justification whatsoever. This has been the problem in trying to understand the viability of H-beam and X-beam rods in the face of column theory. The claims would be laughable were it not for the success these alternative rod designs have had in racing. That success deserves a lot of consideration. In the real world, it is often the case that a design that is inferior to first order calculation can be preferred due to secondary effects. I come to the tentative conclusion that H-beam rods excel in reliability, a secondary effect that might be the primary quality in racing.
As far as ANY alternative design compares to conventional "stock" rods, if they are heavy enough and made of 4340 or some other superior steel, they can be stronger than the stock parts. I note that Pauter X-beams tend to be the heaviest of all rods which could be taken to infer they are the weakest DESIGN. Ironically, engine builders often measure the stiffness of a rod in stretch and crush apparently not understanding that they may be measuring ONLY the average quantity of material in the rod's cross section (proportional to its weight) and not the merit of the design. They don't test the rod to buckling in the proper pinned column fixture.

 

[BrianGar]

Jeez, its not my day on here today. I stated light reading. If you want heavy, Ill give you more than you want.
What sort of environment do you want them in just so that Ill know whether to go to S for steam trains, or H for high performance.

It all depends as you know. I tend to sit near the F1 end of things.

For me, H beams will always win, one major point is the support they offer around the big end - and how that helps to keep it round under extremes. The H beam surface area does tend to be higher(oil weight) but, the over-all mass about the big end is lower than others while offering better big end bore support.

Brian.