×
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!
  • Students Click Here

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Jobs

Connecting rod shape
3

Connecting rod shape

Connecting rod shape

(OP)
I cannot find any analysis of a I-shaped connecting rod vs. a H-shaped. The H-shaped rods seemed to be used in the more expensive applications, but I cannot figure out why it should be preferable.

RE: Connecting rod shape

a very simple analysis =>

H-Beam design is used were engine expected not have any
chance of damaging-detonation , its stiffer in that plane
and would increase damage

H-Beam = NASCAR , etc

I-Beam = more flex in that plane, more forgiving , so used more in Nitrous classes where exact tuneups are harded
to get quickly, and more detonation chances

the H-Beam is better when things are perfect or close to

Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

RE: Connecting rod shape

"H" shaped are easier to machine from billet, especially before CNC mills were common, while "I" shaped is probably a bit easier to forge.

The "H" style was made popular by the excellent quality of the rods made by Carillo from about 30 or 40 years ago, and the fashion continues, probably as it is "obviously different" to stock rods.

Technically they are equal strength to "I" rods under tension, which is were virtually all rod failures occur in normally aspirated engines.

Under high compressive loads, like in high boost forced induction systems, "H" beams have a slight dissadvantage as the long thin ribs can flex and allow the rod to bend.

Regards
pat

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

RE: Connecting rod shape

In school one of my professors spoke of torsional stiffness of connecting rods as a forgotten design consideration.  He said that there are some engines where torsional excitation of the piston/rod/crankpin system can be an issue.  Unfortunately I don't recall which engines exhibited the problem but I suspect they were larger than automotive type.  A tubular connecting rod construction can be used to provide acceptable stiffness in axial, bending, as well as torsional load cases.  I suspect the real reason is the manufacturing method used for automotive "H" versus "I" section rods, as Pat stated above.  Anyone know of examples of engines that had trouble with connecting rod torsional stiffness?

RE: Connecting rod shape

I have designed a conecting rod for sport aplication, the first design was a H shape rod, but due to the hight engine speed and load supported, the material stress of this design was very close to the limit, for this reason I tryed to design a second one but with I shape.
The result was that for the same weight, the I shape was around 20% stiffer, and the weigth distribution more closer to the ideal.
I'm agree with Pat arguments.

Paco

RE: Connecting rod shape

In "The Internal Combustion Engine in Theory & Practice", Vol 2, C.F. Taylor says:
"The theoretical loading on the connecting rod is easily calculated ..... the actual stress on the rod column may be far greater than such a calculated value on account of unsymmetrical loading, vibratory stresses (including the piston in rotation), and stress concentrations. Heavy unsymmetrical loading comes from the piston axis being out of line with the rod axis, more or less common to all engines. It has been found that such loading may cause greater bending moments in the rod than "whip", and it is one argument for orienting the "H" section of the rod 90 degrees from the usual direction. This design also simplifies forging and machining in many cases".

RE: Connecting rod shape

Check out this ferrari F1 rod (picture is from Desmo at AtlasF1, so thank him for it, not me!):


RE: Connecting rod shape

Picture doesn't display, is there a link?

RE: Connecting rod shape

I agree with that English Muffin.
So all arguements aside. How many failures have you guys/gals, seen in both styles? As someone mentioned Carillo rods were (are) the best. And they have always been
the 90deg I or "H". I have no experience with H but I have seen lots of "I" rods bent and in pieces.

RE: Connecting rod shape

No one said Carillo is the best. I said they were excellent. I think they were the best, but some have closed the gap and maybe even surpassed Carillo. It depends somewhat on application. What is best in a diesel truck is not always what is best in a F1, or a top fuel dragster or a sprintcar.

Crower, Manley, Argo, Precision Automation and Robotics, and even Eagle  make some excellent rods now.

The F1 rod is an "I" beam, and I expect Ferrari could afford Carillos if they thought they were better suited to their engines.

The Ferrari rod probably takes nearly twice as long to machine as compared to a Carrilo.

Regards
pat

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

RE: Connecting rod shape

One of my only experiences with a bent rod was where a piston disintegrated in one of my engines - all that was left was the pin, but it had hit the smashed up piston pieces and the rod had become bent. What I think you are calling the H style, and Taylor is calling the 90 degree orientation, would have helped a bit, but the rod probably would still have bent. I once made some very large billet CNC machined 7075 Aluminum rods for a guy with a turbocharged pulling tractor. I made them with a simple rectangular section, but he had a detonation and one got bent. So it's not true to say that rods always fail in tension due to inertia forces, although my general impression is that racing guys do usually have tension failures. If you run the numbers, however, it is not obvious, at least to me, that this should necessarily always be the case. It also seems that the buckling strength is generally likely to be less than the tensile strength.

RE: Connecting rod shape

My last sentence should have included the qualifier "in the part of the rod we are discussing". A lot of the tensile failure pictures that I have seen seem to be related to the big end (such as bolts or upper part of the bearing housing).

RE: Connecting rod shape

I agree with English Muffin.
A decade or so magnufluxing passenger car, motorcycle, and race car rods suggests crappily designed or poorly executed transitions between I/H-beam and big-end or wrist pin end have brought WAY more rods to their demise than the orientation of the beam section.
I think Taylor also points out the (Carillo) H beam lends itself nicely to the threaded rod instead of bolt-n-nut design, which tends to improve the beam-to-big-end transition a lot simply by removing the whopping notch for the bolt head.

RE: Connecting rod shape

I have always used whatever worked best in the application at hand.  In my case I have broken both Carillo "H" beam and Cosworth "I" beam rods on more than one occasion, unfortunately.  Pat has it right on---the Carillo rods were "different" and due to quality construction and strength (compared to OEM stuff of the 50's and 60's) were perceived by the rodding public as 'superior'!  My first choice has usually been Cosworth.  Certainly NOT because of cost but, instead, overall weight was my main concern.  The Cosworth rods could be machined to a lighter (in some cases MUCH lighter) weight.  I still like Carillo and should I get even crazier than I already am, I will probably use them in my vintage mini.  Not a big concern at it's present level of tune!

Rod

PS:  The only rod I have ever seen 'bent' was from an early Chrysler Hemi drag engine.  I think it lost fire and hydraulicked a cylinder.  You will have to ask Pat on that one. V8's are a little out of my area.

RE: Connecting rod shape

Rod

The only noticably bent rod I ever saw was an OEM unit out of a 308 CI Holden V8. Similar story. It had a poorly installed water injection system, which siphoned water into a cylinder while parked on a slope. It hydraulic locked and bent a rod.

My comments re "H" vs "I" for high boost applications is precautionary, as to me the chance of the long thin ribs on the "H" are somewhat more susceptible to flex, and allow the beam to bend.

Regards
pat

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

RE: Connecting rod shape

There appears to be a definite difference of opinion here. It would appear to me that if the rod fails in bending due to off center compressive forces or even elastic instability, it is more likely to bend in the plane of the gudgeon pin (brit speak), since no significant moment can be applied to the rod in the orthogonal (pinned) direction, so surely the H design would be superior in this regard, as claimed by Taylor. MaxRaceSoftware and patprimmer appear to hold the exactly opposite view.

RE: Connecting rod shape

EnglishMuffin

Yes I hold the opposite view.

Even though the ribs on the "H" style are somewhat deeper, which should increase stiffness in the plane of the gudgeon, the ribs are so long and thin, that they are inclind to flex at the outside edge, alowing the beam to bend.

An 'I" beam puts a lot more metal in the area of the outer fibre when the rod is being bent in the plane of the gudgeon.

Much of this is based on intuition and anecdotal evidence.

I would really like to see some finite element analysis of a few structures. I might be converted.

Regards
pat

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

RE: Connecting rod shape

It sounds as though you are talking about lateral buckling of the compression flange, which is covered in Roark to some extent in the beam section. Might be interesting to run the numbers for a specific Carillo rod, or as you say carry out an FEA buckling analysis.

RE: Connecting rod shape

By the way, I am somewhat confused about the I and H terminology. On this site :http://www.billzilla.org/engcrank.htm, the H style is defined to be the more conventional design, like the Ferrari rod in StefanH's post, and the I style is the type with the rod cross section rotated through 90 degrees. That's the reverse of what I thought. Which is correct ?

RE: Connecting rod shape

My understanding is that a Carillo is a "H" beam and a traditional OEM rod is an "I" beam

Regards
pat

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

RE: Connecting rod shape

patprimmer: So we both seem to agree that the guy on that site has his definitions back to front. Since you and he are both Aussies, it appears unlikely that this is a "definition inversion" peculiar to your continent, unless the terms "left" and "right" somehow become interchanged in the southern hemisphere. His error might cast doubt on some of his other assertions, such as the statement that most rod failures occur at the bolts, although that one seems to me likely to be true.

RE: Connecting rod shape

H beam vs I beam, I have been building racing engines of all type for some years now. What I have seen as the cause of rod deflection are 1: elongation ,occurs when the revs are sustained at high numbers. this is evidenced by carefull claying of the piston and dry assembly then cranking engine over ,dissassemble carefully and checking pisto to head clearance. documenting this clearance then running on a dyno dissassembly and claying again. with forged I beams there is almost allways a narrowing of the clearance gap.
 Theory for this is that the pistons on the exhaust stroke are flung to the end of their stroke unrestrained and then jerked back putting large strain on the rod .Solution stronger rods that are lighter and lighter mass at the end of the rod (pistons)
 2: compression failures almost always due to detonation or Pre ignition caused by too much compression and too much spark lead. Or just plain too much fuel load IE nitro or turbo boost too high.

RE: Connecting rod shape

I would have thought that the greatest tensile stresses would occur on the intake stroke, since only in that case are the inertia loads not counteracted by gas forces, although there wouldn't be much in it between the intake and exhaust strokes.

RE: Connecting rod shape

I would have thought that the greatest tensile stresses would occur on the intake stroke
-----------------------------------

on a normally-aspirated engine ,
if the headers are working correctly,

you should/want a negative pressure wave to
be centered on the overlap period

so the rod should be in more tension
during overlap period
in normally-aspirated engines with headers working

Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

RE: Connecting rod shape

Yeah - I guess I'm wrong - wasn't thinking about overlap - the maximum tensile stress occurs at TDC between the intake and exhaust strokes. So I suppose if a tensile failure were to occur in a normally aspirated or supercharged engine, it would probably be just before the intake stroke began. What is the situation on a turbocharged engine?

RE: Connecting rod shape

EnglishMuffin

The only two orientational differences I have seen between Aus and USA are:-

1)  Here the sun travels from right to left, whereas in the USA it travels from left to right. Singapore is quite confusing in this regard though.

2)  You guys drive on the wrong side of the road, as you are probably very well aware, being of British origin.

Oh yes, I almost forgot, we also have the Gay Mardigra, but then again, you guys have San Francisco.

I guess turbos are a bit easier on rods re tensile as there is no effective exhaust scavanging, and there is always some residual back pressure in the exhaust port, as well as boost pressure in the inlet, but TDC overlap is still probably the highest tensile load.

Regards
pat

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

RE: Connecting rod shape

When I was in Australia, the orientation of the moon crescent looked different as well.

RE: Connecting rod shape

Gentlemen,
    I just want to add about the effect of buckling due to the combustion pressure when comparing between I and H conrod.
   If we simulate both the I and H for buckling, we will notice that the H conrod is stronger due to more cross section area at the critical points.
   Overall, it is still difficult to predict which one is better. For turbocharged application, we have to consider a lot about the buckling effect. On the other hand, for NA application like the F1 where it revs up to 19000rpm, definitely centrifugal effect and repeated compression and elongation effect will have to be considered. Not to mention about critical fastener attaching both the big and small ends.
   In addition to that, the effect of bore and stroke will also have to be considered. The Honda F1 turbocharged engine has longer conrod if compared to the new F1 conrod. This will yield different result entirely.

AO

RE: Connecting rod shape

If flange buckling of the H rod ever turnesd out to truly be a problem - and it is not clear that it ever would - one might be able to control it somewhat by leaving one or two small linking portions between the webs at intervals. Of course, that would probably entail machining with a ball nose end mill rather than a form cutter, which is a negative.

RE: Connecting rod shape

for a little more controversy ??   :)

why do majority of Hi-Perf rods use bolts rather than studs ?

especially in aluminum rods where you would think they would be beneficial in protecting aluminum threads
by using studs rather than bolts .

Answer=> ?   .... the stud has a too much of a
"lever-effect" and transmits distortion to the journal hole
(any motion of the stud distorts the journal hole ?)

any other opinions / insight ??

Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

RE: Connecting rod shape

Bolts studs?

Most big end bolts are through bolts with nuts on the end.
Why wouldn't Al rods be counter bored and use the same? Much stronger than threading the Al. It would be a bit difficult to aquire any stretch threading into Al, especially the minimal amount on the big end.

RE: Connecting rod shape

Through bolts sounds better to me. But if there is a difference between non-through bolts and studs, it would appear to me that it would probably have something to do with the fact that studs are usually designed to be tightened until their threads jam against the end of the mating female thread. If this is not done, I can't see much difference between studs and non-through bolts, but perhaps I don't fully understand your "lever effect". In my view, the reason for using non-through bolts rather than studs has to do with the following. I remember once being told that for the best fatigue performance, studs should be designed so that when fully tightened they contact the bottom of the hole instead of being tightened to the end of the thread in the above mentioned manner, but this seems to be rarely done, and using a bolt is the next best thing, since it never reaches the end of its thread when tightened. Although stud bottom contact is tougher to do from a manufacturing standpoint, it should be possible to achieve it by having a short relieved portion of thread at the end of the stud so that it can contact the bottom of the pilot drilled hole in a repeatable manner. I also believe that to help maintain preload one should use the longest bolts or studs practicable, especially with aluminum rods.

RE: Connecting rod shape

The problem with through bolts in high performance rods is that the rod must be notched to make room for the head of the bolt. This creates a weak point in the rod.

One solution is to make the bolts longer so the head is further from the bore of the big end, but this makes the rod and the bolt heavier, and increases the room taken for the crank to rotate at a given stroke. Many high performance engines are pushing the limits re RPM and capacity, so extra weight and bulk are a significant dissadvantage.

Most good high performance rods use a pair of tube dowels to locate the cap on the rod, and to reduce flex at the parting line, thereby reducing fatigue due to bending of the bolt, stud, cap screw.

The above applies to both steel and aluminium rods to varying degrees.

Extra considerations re aluminium are:-

The extra bulk is even more critical as aluminium rods are already bulky

The notch is more critical, as aluminium (to the best of my knowledge) is more notch sensitive than steel.

The longer the bolt, the greater the dimensional change difference, due different co-efficients of expansion between aluminium and steel.

I would think the ideal design would be a stud, mounted in the rod as described by EnglishMuffin, with tube dowels, and possibly a very small pilot hole right through to avoid trapping air under the stud, as this might expel and blow out the stud retaining adhesive (locktite).

This would allow significant strength to be retained in an area where (from anecdotal evidence), rods are weakest , while reducing the weight and bulk of aluminium rods and the bulk and to a lesser extent, the weight of steel rods.

The reduced bulk would be in the area that often interfers with the cam lobes in a typical American V8, and obviously the cam lobe is an area where interference is difficult to measure and correct for.

The linking portion on the "H" beams sounds like a good idea to me in high compressive load situations.

The cross sectional area of the "H" beam is much lower near the outer fibre which is resisting bending.

Regards
pat

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

RE: Connecting rod shape

The rods of the venerable Meyer-Drake Offenhauser had through bolts nested in between local webs, with no notch -just a small counterbore, and didn't look as though they had more bulk than that Ferrari rod, but whether the design would pass muster after an FEA I don't know - as you say, somewhat weak in the critical area. The rods also appeared to be tubular, but how on earth they were made I don't know. The bolts themselves doubled as locators.

RE: Connecting rod shape

Does anyone mind reviewing some formulas for figuring out the loading on the rod at a given rpm? I remember seeing a thread regarding that here some time ago, but I am no mathematician. The reason I ask is I am trying to figure out the load I would place on a rod I wish to use in a buildup of a budget racing engine. I dont want to spend money on billet rods if I dont need to. I know what rpm I want to rev to and I know the piston weight I want to use and the crank stroke. If someone can help me work out the formulas I would be glad to create a little VB script that makes it a plug and chug operation to calculate.

RE: Connecting rod shape

If you take a look on our website centralvwaudi.com and click on the tech button youll see the i beam rods i had made by farndon engineering,and theres a picture of the new rod next to a std rod.I had them made as i bent all the rods in both directions basically shortening the rod by 2mm.

RE: Connecting rod shape

re: EnglishMuffin (Mechanical) Jan 20, 2004
[...] appear to me that it would probably have something to do with the fact that studs are usually designed to be tightened until their threads jam against the end of the mating female thread. If this is not done, I can't see..."

Hmmm, I wonder about your installation procedure.  I suggest you look at the ARP site (http://www.arp-bolts.com/index.html), go to Product, then the Main or Head section, then the How To Install section.  They say to finger tighten only.

Some time ago in Precision Engine (formerly Precision Machining) magazine they said to back off the stud 1/4 to 1/2 turn.

RE: Connecting rod shape

I agree that sounds like a good procedure, but how often is it actually followed in practice, and can one be sure that the stud won't turn during the nut tightening operation? After all, most garage mechanic's tool chests include a stud extractor - a staple of all the major tool manufacturers.

RE: Connecting rod shape

Stud bolt? With the use of the stud bolt, the weak point is shifted to the nut thread or the bolt thread where the two are engaged between one and another.

I guess it is much better to use bolt engaged straight to the conrod upper end. This way the bolt would have longer engaged thread if compared to the nut engaged to the stud bolt.

RE: Connecting rod shape

The length of the engaged thread is not of overriding importance, since the first two or three engaged threads generally carry the majority of the load. You can demonstrate this if you do a detailed FEA. With a correctly designed stud, stud bolt, or through bolt, tension failure will usually occur in the male thread close to the mating member.

RE: Connecting rod shape

Most of the questions that are being asked have been answered by catalogs from manufacturers over the years.  Yes they have thier own agendas but i have found that discussions with thier tech lines will result in answers closer to the truth about what they want to sell and why. Most of the better rod manufacturers will make custom rods to your specs.

RE: Connecting rod shape

Englishmuffin,
    For high revving engine, we want the crankcase and the oil scrapper to be as close as possible to the conrod locus. This will ensure compact design as well as low friction loss from the splashing oil.
    Therefore, with the nut instead of bolt head, I wont be able to keep the oil scrapper very close to the conrod locus. The closer the oil scrapper is to the conrod locus, the more horsepower I would get out of the engine.
    With the stud bolt, I need to use a special nut to enable it to be closer to the oil scrapper. With this combination, the weak point is shifted away from the main conrod structure.
    As for the FEA, yes we did. We did it over titanium bolts (not the stud bolt though). The factor of safety is very close to 1.0. It's just enough for one Sunday afternoon race. To join the 18000rpm NA engine club, the conrod, piston and crankshaft need to be extremely lighweight, stud bolt for the conrod does not belong there.

AO

RE: Connecting rod shape

"A tubular connecting rod construction can be used to provide acceptable stiffness in axial, bending, as well as torsional load cases."

With my very limited engineering backround, I have just completed my first semester of college, I have to disagree.  I would think that the majority of the forces put on a connecting rod are bending forces as a result of the piston pushing downwards on the rod and the rod being at an angle due to rotation of the crankshaft.  I would think that you would want to concentrate a majority of the metal towards the outside of the rod hence the use of I-beam rods.  A tubular rod would have very little metal towards the outside with most of its mass concentrated in the center.  This means the tubular rod would be more prone to bending, especially with small rod/stroke ratios.  Tubular connecting rods are used, though.  Our ship's service diesel generator, about 450kw@910rpm for an idea of size, has solid round rods.  An ideal application for tublar rods would be in slow speed 2 stroke diesels as they have an articulated rod.  The piston rod moves only vertically and has no side loads on the rod.  For the most part they are tubular.  The actual connecting rod which joins the piston rod and the crankshaft is of an I-beam style.

Here is a link to the best example of an articulated rod I can find.  http://www.haskey.com/johnh/railfair99/16_porter.jpg

RE: Connecting rod shape

76GMC
   Interesting. We did several classical calculation comparison between I,H and cross shape conrod. The H appeared to be strongest. The engineer in charge did inform me about the hollow rectangular shape conrod. This will yield better result. My respond to his suggestion was, how can we manufacture the hollow conrod.
   If you can manufacture for me a tubular conrod while at same time having the small and big ends as strong as the conventional conrod, you should get your process patented.
    As for my choice of conrod for high revving engine, my first choice will still be H type conrod. With the I shape conrod, the thick outer metal is in neutral position to resist bending movement. As for the H type conrod, I have two thick structure opposing the bending movement.
    Anyone can disagree, but as a designer I am free to choose the design. The moment of truth comes during the WOT durability test.

AO

RE: Connecting rod shape

The offenhauser ones looked as though they might have been drilled out, with a plug put in from inside the big end bore. But I could be wrong.

RE: Connecting rod shape

I usually see H-beam rods used in forced induction applications like drag racing, while I-beams are used in high RPM naturally aspirated applications such as NASCAR.  The reason being is that the H-beam weighs more and is a bit overkill when you're only making slightly more than 2hp per ci like a NASCAR.

RE: Connecting rod shape

76CGM,
    I dont think it is an overkill unless you overdesign the thickness of the H conrod for the required application. Many racing bikes and cars that rev at very high rpm (some at double the speed of NASCAR max rpm) use H type and there is no clear rule for forced induction or NA engines.
    It is the preference of the engineers, some use I type and get better result and some use H type and get better result. No matter which one we choose, it is the details of the thickness, radius, length, tolerance, manufacturing process, development history, FEA calculations, R&D, etc. that will determine the final result. In racing, the final result is the one that matters.

AO

RE: Connecting rod shape

This is somewhat related to similar posts  on the Maximum Piston speed forum. In a previous position with one of the former Japanese engine suppliers to CART, we used H-beam con rods. We were working with a leading Euro connecting rod manufacturer and at the time our revs were relatively low (compared to our final revs)and they recommended the H section due to our anticipated progress. Our rod bolts were from an American manufacturer of rod bolts, but we didn't use nuts but instead threaded the rod so that we had more threads (~2.5 x d) of engagement in order to obtain more consistent measurements of rod bolt stretch.
We used 2 ΓΈ3.0mm by 6.0mm dowel pins (parallel to bolt centerline) for locating the cap relative to the rod instead of hollow dowels, concentric to the rod bolts.
Prior to this, we were building our own rods and ours looked like they were chopped out of a block of wood (they were designed and built by passenger car engineers who had never experienced what race engineering is about).

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources