20

Does Length Matter?

by Khiem Dinh

Khiem Dinh is an engineer for Honeywell Turbo Technologies at the time of this writing.  All statements and opinions expressed by Khiem Dinh are solely those of Khiem Dinh and not reflective of Honeywell Turbo Technologies.

Someone recently asked, “Why a longer rod?”  Someone else answered, “Longer dwell time.”  I never really thought about it before, so I thought I’d look into it.  Using some basic math, we can prove that a longer rod does indeed create a longer dwell time around top dead center.

Here's a diagram showing the geometry between the crank, connecting rod, and piston.  The math only works out assuming the piston pin location is not offset.  So using some basic trig, we can calculate piston position (S =A+B in this diagram) versus crank angle.

 

This graph shows piston position (normalized so that 0mm is BDC) vs. crank angle for 100mm and 300mm long connecting rods on an engine with an 86mm stroke.  Yeah, you would never use a 300mm long conrod, but I wanted to make it easier to show the difference.

As you can see in the diagram, the piston with the longer conrod spends more time around top dead center (0 degree crank angle) due to a slower acceleration rate towards and away from TDC.  A longer dwell time around top dead center makes better use of the combustion pressure and turning the pressure into torque; a rule of thumb is that you want peak combustion pressure around 15-17 degrees after TDC.  Furthermore, the combusiton pressures are higher while the crank angle is lower resulting in less side loading of the piston into the bore.  Less side loading results in less friction between the piston and bore freeing up power and reducing wear.  The reduced piston acceleration also improves piston ring life as the lower acceleration equals less force on the rings (remember F=ma?).       

The Yamaha YZF R1 is a typical modern sports bike that revs to 12,000 rpm.  Its engine has a stroke to rod length ratio of 2.12:1 which is suitable for high rpm use.  This can be compared to a Nissan SR20DE engine which revs to 7700 RPM with a rod to stroke length ratio of 1.58:1

 

The R1 engine has a pretty tall deck height compared to the general compactness of the crankcase.  In the case of the bike, the chassis is designed around the engine.  With a 110.5mm rod and a 52mm stroke, the engine is tall for its displacement compared to a car engine.
We think automotive engine designers compromise rod length to try to fit engines under low hood lines of modern cars.  This is one of the reasons why the SR20DE ended up with a pretty poor rod ratio for a performance motor.

 

Page 1 of 2 Next Page
Posted in: Magazine, Tech, Engine
Bookmark and Share
Comments
Jasonrg77
Jasonrg77link
Monday, June 20, 2011 5:03 AM
Would one pick the max. rpm, stroke, then rod length for the application? Aside from detonation what beats up a con rod?
Rockwood
Rockwoodlink
Monday, June 20, 2011 7:16 AM
Two double entendre articles in a row Khiem!

Would the variances you see in rod length with engines like the SR20 have much of a negative effective in the lower rev ranges?
spdracerut
spdracerutlink
Monday, June 20, 2011 8:56 AM
Steve, I've just been talking about thrust loading, shaft motion and longer rods, I don't know what you mean....

How much variance are we talking about? But probably very little with regards to rod:stroke effects. I think I'd be more worried about variance in compression ratio?
SixCylinders
SixCylinderslink
Monday, June 20, 2011 9:18 AM
I do think it's funny that the SR20 which is arguably the most popular Nissan engine (and known for being revved over 10K rpm) has a 1.58 R:S ratio. I mean Nissans TYPICALLY are high as hell for car engines, if my numbers are correct a VG30 has 1.86, a VQ30 has 1.84, an RB30 has 1.79, a VH45 has 1.78, a VQ35 has 1.77, and the KA despite being passed off by some enthusiasts as a "truck motor" has 1.72 even with it's 10mm longer stroke.

On the other side of the scale for anyone that cares the 4G63 is 1.70, the LS1 is 1.68, modular 4.6 1.67, 2JZ 1.65, RB26 1.65, K20 1.62, 3S 1.60, D16 1.52, and QR25 1.43. If I missed anything interesting let me know, I try to keep lists of specifications like this laying around for a hobby.
Dusty Duster
Dusty Dusterlink
Monday, June 20, 2011 10:27 AM
@SixCylinders:

The new Ford Coyote 5.0 (which was based on the Modular V8, but heavily modified) has a rod-to-stroke ratio of 1.62:1, slightly lower than the original Modular design.
Rockwood
Rockwoodlink
Monday, June 20, 2011 10:34 AM
Khiem,

Variance as in what you could realistically see in an SR20 or similar motor by raising the pin, not through raising the deck or other more exotic mods.
Mike Kojima
Mike Kojimalink
Monday, June 20, 2011 10:52 AM
Kinda a seek peak, I built a long rod SR20VE and will be doing an article about it soon.

The torque peak was at a higher rpm and the power peak was less peaky, like the hump at peak power was wider with less fall off after the peak. About what it should be in theory. There was only a small difference in peak power but didn't have time do do a lot of dyno tuning, mainly cam timing.

Another funny thing, the engine sounded a lot different.
mxpop
mxpoplink
Monday, June 20, 2011 10:55 AM
FWIW, I have seen where rod length was changed with adverse effect on valve spring life (with no change to cam/rpm's etc). Apparently, the harmonics/frequencies were moved into an unfavorable zone due to the rod length change and made an otherwise reliable motor into a less reliable motor. One change can and will effect other things, lots to consider...

I think I am the one who got your gears churning when I mentioned "dwell" in a previous article's comment section. My expirience with utilizing longer rods came years ago in the early '90s with some Yamaha 2-stroke motocross bikes. The "long rod kit" offered by Yamaha was a win-win for us and this was on a motor that was originally designed as a "race" motor in the first place. I gotta believe that most if not all "street" designed motors will benefit when being converted to race applications but as always, there will be a ripple effect with how it reacts with other factors. Note: We had to use a spacer plate under the cylinder to get the port timing corrected which also gave us larger crankcase volume. In a 2-stroke, crankase volume is part of the intake track length/volume which has great effect on performance as well. I miss 2-strokes as they were the only true race motors widely available to the public.

Micah McMahan
Micah McMahanlink
Monday, June 20, 2011 11:33 AM
Yes JC, you do have a point that most people don't know how or don't know what one change will effect. As I've been doing this since 07; in the EJ, K-series and 4G63 platforms but mainly in the Subaru world; there are several things to consider other than simply fitting in what you can. A well engineered package, and I don't just say engineered like most shop say when they do 9th grade geometry on what fits, I mean engineered. Having customers on the same EJ with no real issues, evident by rebuilds after every other season, it does carry benefits. The main thing I was concerned with is targeting the area of the curve used and shifting the volumetric curve to that region. This is the same reason I can tell JC with certainty that I can make more power with fewer cc's than what they have now. Heck, you guys might cut back on splitting cylinder walls too..but I've already spoken with Peter about this.

Enough Subaru talk...

This article barely touches the tip of the iceburg. I do however see that you snagged a spreadsheet from Gustave's E30 M3 website. I've done more extensive but simple Excel work that I put up for 'show and tell' on NASIOC for people to see. Virtual 4-stroke, Ricardo work, etc is also useless to post.

If you want to ues them for real examples and to further this topic, let me know. I'd be happy to share.

There are pro's and con's to it and because it's an engine, it's naturally a long list...now back to kicking off the next dyno test in the test cell.

SixCylinders
SixCylinderslink
Monday, June 20, 2011 11:50 AM
"Another funny thing, the engine sounded a lot different. "

I was wondering if rod:stroke had that effect, I typically like the sound of big bore long rodshort stroke engines but I've never really payed attention to sound when just the rod length is changed

Mike, I'm looking forward to this engine build for sure. NEO VVL is pure win!
spdracerut
spdracerutlink
Monday, June 20, 2011 11:57 AM
JC, yup, this article is your doing. Inspiration can come from odd places...

Micah, you said, "just the tip." Just messin with ya :) But you're correct, this is a quick intro into the subject. After I did my little derivation, I found an equation for motion on wikipedia that's easier to take the derivative of for calculating velocity and acceleration. Change in distance over time gives velocity, change in velocity over time gives acceleration. So first derivative of the distance/position equation gives velocity. Second derivative gives acceleration. So, for anyone that wants to create their own spreadsheet.

http://en.wikipedia.org/wiki/Piston_motion_equations
mxpop
mxpoplink
Monday, June 20, 2011 11:59 AM
"This is the same reason I can tell JC with certainty that I can make more power with fewer cc's than what they have now."

^^^ absolutely true, hell, F1 does it with less CC's and NO TURBOS!!!!


The oldest saying in motorsports that you can have it fast-reliiable-cheap but can only chose two is as true as ever. I wouldn't bother racing if we couldn't be fast and I have essentially zero racecar budget so some reliability gets sacraficed. We have spent big $$$ in past years on big-name motor assemblies that were no more reliable, even at much reduced power levels that wouldn't even be competitive in todays TA racing.

I would love more longevity out of our package but with 11 wins in the last 11 races including many track records, I would say that we are hitting the targets pretty good. If you are offering to set us up with a 2 season lasting motor while not effecting our miniscule budget, then thank you thank you thank you.... :-)
mxpop
mxpoplink
Monday, June 20, 2011 12:02 PM
"JC, yup, this article is your doing. Inspiration can come from ODD PLACES... "

Hey, I resemble that remark...... :-)
Mike Kojima
Mike Kojimalink
Monday, June 20, 2011 12:25 PM
Firsr long rod motor I tried was a 4AG back in 86 or 87, I can't remember. Had simular results. Micha, please share your data! JC which engine developed valvetrain problems after going to a long rod?
Der Bruce
Der Brucelink
Monday, June 20, 2011 12:35 PM
Steve - Dang you and your French! I was going to joke of innuendo as well but I'll just say well placed "adianoeta" throughout den Artikel!

Khiem - Now you've got me wanting an article on the latest piston top designs!

Mike - I'm excited to see if you guys took advantage of the longer rods by raising the pin and CR, while shortening the skirt, etc for power and longevity!?

Mxpop - 2 stroke powerbands were a blast up to about 250ccs. The 500s were so heavy and torquey, they almost felt like 4 strokes. Nothing compares to the 4 stroke technology of the last 15 years though!
mxpop
mxpoplink
Monday, June 20, 2011 12:55 PM
MK, it wasnt going to a longer rod but actually doing a stroker without a rod change, effectively shortening the ratio. I realize I said the rod length was changed but I meant to say the ratio was changed and I didn't say longer.

DB, I raced a Service Honda (CR500 2-stroke in a 250cc Alum frame) which was the best of both worlds at the time. Talk about harmonics, that bike broke shock/fork springs as well as cracked exhausts regularly. That bike was awesome but the marriage came with penalties as always. Modern 4-stroke MX bikes are pretty amazing (my son races a 2011 Kawasaki 250FI 4-stroke) but wouldn't stand a chance against what a modern 2-stroke would be if the 4-strokes werent given significant displacement advantages.
SkullWorks
SkullWorkslink
Monday, June 20, 2011 1:35 PM
Modern 2-strokes is of course a contradiction in terms when it comes to motorcycles as it is mostly abandoned (for that application) and hasn't advanced like the marine apps (which are barely in use)

Comparing a 250cc 2stroke to a 250cc 4stroke was never fair and isn't now (combustion per rpm vs combustion per rpm/2)

a 250cc vs 500cc is a much more apples to apples comparo,

and you are right the new 4stroke stuff is silly, and needs rebuilds like the old 2 strokes did
mxpop
mxpoplink
Monday, June 20, 2011 2:48 PM
As good as the best 2st were, imagine if the R&D time/$$$ was spent to make them direct injected and all the Ti and Mg money can buy. Comparing a 250cc 2st to a 500cc 4st (or 450 as the AMA does) may produce similar performance but then again, that simply supports my point that a 2st is a better race motor "unless" you give the 4st a huge displacement advantage. The cost/complexity/packaging of the 4st will always keep it behind a 2st when on equal terms (and I like 4st!) Now, if someone would just come out with a proper 3st we could all be happy.... (and I dont mean Wankel)

IMHO, modern 4st MX bikes are no more complicated or time consuming to do typical maintenance rebuilds on than most power-valve equipped 2st. Replacing a piston and valves on a 4st is similar to replacing a piston and powervalves on a 2st assuming no machine work is needed in either case. FTR, factory recomended piston replacement intervals are similar on each type (but most people extend that interval unless a serious racer).

Anyway, all this 2-4 stroke discussion is getting us off track from Khiem's rod length fascination.... ;-)
Mike Kojima
Mike Kojimalink
Monday, June 20, 2011 2:59 PM
JC Service Honda bikes are badass!
Joe
Joelink
Monday, June 20, 2011 3:33 PM
SixCylinders - VQ30 Rod/Stroke ratio is 2.012 with a 73.3 mm crank / 147.4 mm rod. They are the highest R/S ratio I've seen for a mass produced engine, I was always wondering if are any production engines with a higher ratio? I've always wanted to build the spare VQ30 I have to run in the 8K ~ 8.5K RPM range but never had the money or ability to do so.
Joe
Joelink
Monday, June 20, 2011 3:38 PM
It is amazing how high the R/S ratio on F1 engines are nowadays. I recall years ago reading that a 2.1 ~ 2.2 R/S ratio was the norm on an F1 engine, so to see 2.72 is crazy.

Great article! Always fun to learn something new.
SkullWorks
SkullWorkslink
Monday, June 20, 2011 4:31 PM
you'd love to know the R:S in some of the drag B-series stuffs (and now K-series too I Imagine) I have seen some spindly looking rods down at cunningham's that had honda journals/pin diameters...that's all i'm at liberty to say...
Mike Kojima
Mike Kojimalink
Monday, June 20, 2011 4:33 PM
I think they are pretty bad right?
SkullWorks
SkullWorkslink
Monday, June 20, 2011 5:39 PM
1.745 for a stock B16, pretty decent to Start with
stuntman
stuntmanlink
Monday, June 20, 2011 5:52 PM
NSX 3.0L & 3.2L

Bore x Stroke: 90 (&93) x 78mm

Rod length: 152mm

152mm / 78mm = 1.949 !!!!

Common 3.5L stroker kits for the NSX to make them a 3.5L increases the stroke to 86mm

152mm / 86mm = 1.767 (that's a 0.182 difference!)



Pretty interesting stuff. Might explain why the 3.5L stroker kits have issues running even the stock 8,000rpm rev limit.
stuntman
stuntmanlink
Monday, June 20, 2011 5:58 PM
@ Dusty - Those Coyote motors with upgraded valvework have been seen to rev to over 8,000rpm in some magazines! That's a V8 with a 1.62 -kinda neat.
spdracerut
spdracerutlink
Monday, June 20, 2011 6:45 PM
The Ferrari F430 has a ratio of 2.19. Not exactly mass produced, and it's a Ferrari :) LS9 in the ZR1 is 1.73 from what I've found.

Billy, interesting to hear about the 3.5L strokers having issues. I mean, 86mm is the normal stroke for a 2.0L 4-banger and the majority of those have no issues spinning 8k. So I'm curious as to what else is going on in that V6 of the NSX.
Mike Kojima
Mike Kojimalink
Monday, June 20, 2011 6:52 PM
Oh, I was talking about the super stroked out B and K Motors used in All Motor class drag racing. I heard they are up to 105mm.
SixCylinders
SixCylinderslink
Monday, June 20, 2011 6:58 PM
@ Joe I looked up some numbers and you're right about the VQ30DE rod length (which leads me to question whatver source I used when I got my low calculation). I personally love the VQ30 and would love to see one rev to at least 8K and preferably much higher but if my memory serves correctly all variations of the VQ30 used hydraulic lifters, VQ35s used solid lifters, have better aftermarket support and on top of it all their heads flow better (I honestly think the best flow numbers I've ever seen for a head was ported VQ35 heads at 410 intake and 275 exhaust, I want to say I got these numbers from Cosworth but I'm not entirely sure. I've seen some ported Brodix Small Block Chevy heads that did 410/283 but that was at .600" as opposed to .500" and I'd like to think the VQ would be better able to withstand the forces endured by both the top and bottom end than said Chevy.
SixCylinders
SixCylinderslink
Monday, June 20, 2011 7:02 PM
@ spdracerut, is there a possibility it has something to do with the C series being a 90 degree V6? Maybe some sort of balance flaw being exposed by the extra stroke?
Eric Hsu
Eric Hsulink
Monday, June 20, 2011 8:16 PM
JC: wasn't your ECU driving the AVCS solenoid at he incorrect frequency? At least that's what I heard. I was never up to date with your program/engines so I don't know if the issues occured at the same time. On a belt driven engine, the "change" in "harmonics" are not going affect the valvetrain. It can on an engine that has gear driven cams and maybe even chain driven cams, but not a belt engine.

I am genuinely asking and not defending here. I have no idea whether you're referring to the rod ratio change of Cosworth components (although I suspect you might be).
mxpop
mxpoplink
Monday, June 20, 2011 8:40 PM
Mike and Eric, sorry if I misled you two. I was not referring to my Subie motor but rather to another model entirely to which I have knowledge. I am not at liberty to discuss specifics as it was not my build and I only have peripheral knowledge.

While it would seem that a belt driven cam would transfer less harmonics than a chain, gear or shaft system, it seems that it would still transfer harmonics through other ways.

As for the ECU/AVCS question on my motor, I honestly don't recall the exact problem or solution as it was like 2-3 years ago. Besides, I'm just the monkey who steps on the pedal thingys....
jamal
jamallink
Monday, June 20, 2011 11:02 PM
Our current motor has an 83mm stroke. I like the idea of having a longer rod, but to be perfectly honest I'll take the added displacement (we all know the saying and it seems to work for FD guys...).

Yes, I know I can get a 75mm stroke subaru motor back up to 2.5l, but if we go that way I can't get a new set of pistons and rods or headgaskets the next day if (when) something breaks. Last year we had a rod go in testing two weeks before superlap. Thanks to our motor full of off the shelf parts were able to get replacements within a couple of days and have it all together with time to spare.
Joe
Joelink
Tuesday, June 21, 2011 1:51 AM
@ SixCylinders - All VQ engines are solid lifters, VQ30 has shims over buckets. They are very similar so much that VQ35 valvetrain and cams have been swapped into VQ30s for a while now with good results. Though as far as bottom end builds go the VQ35 smashes the VQ30 in parts availability.

I've known people take the VQ30s to 8K RPM with stock bottom ends but the oil pump usually explodes after a few hard 1/4 mile passes, some last a bit longer. Don't know if its a weak oil pump or if there is something more going on with engine harmonics that causes the failure.

Also I believe the C30/C32 NSX engines are 60 degrees not 90.
Micah McMahan
Micah McMahanlink
Tuesday, June 21, 2011 7:46 AM
Mike, all that I have to share up right now are some basic crank slider model derivations that I had worked up for the EJ257 block. The last one is a bit different but is self explanatory. I tossed the EJ20 in there as well for comparison sake.
http://i108.photobucket.com/albums/n22/drummerboy1784/Pistonvel-time.jpg
http://i108.photobucket.com/albums/n22/drummerboy1784/Pistondisp-time.jpg
http://i108.photobucket.com/albums/n22/drummerboy1784/Pistonacc-time.jpg
http://i108.photobucket.com/albums/n22/drummerboy1784/Pistonvol.jpg
http://s108.photobucket.com/albums/n22/drummerboy1784/?action=view¤t=piston_dynamicssubiecomparison.jpg

I see a lot of discussion in here (aside form 2st 4st stuff) skiring on what I believe are resonance/vibration issues. This is something that should be taken into account when modifying stroke and rod ratio of an engine (part of why I've been a long time hater of the 83 and 84mm stroke crank in an EJ). If destroking to gain rod ratio, as I do, you should also take into accont the bore/stroke ratio and loading forces. Also consider the rains in reciprocating, not ROTATING, mass that have been changed. For me design criteria was limited to stock dimension crank (OEM rod journal), factory deck height and then up to 103mm (sleeving options). I arrived at 101.6mm and my just under 1.8:1 rod ratio as "ideal" for not needing custom valvetrain and having a useable tq curve in the road race powerband range of a Subie. If you have lag, the driver chose the wrong gear :p
^^JC, you'll see that's why I've been asking Peter for your track data. I'd like to see how much/what you could really gain, as I'm not a product pusher and turn people away more often than not.

With gained RPM and rotating geometry changes, you can have a large effect (good or bad) on the engine...for both long and short rod ratios :)

Man, I miss the track...someone want to hire me so I can leave this consulting gig I have working on diesel engines? haha
Micah McMahan
Micah McMahanlink
Tuesday, June 21, 2011 7:50 AM
Oh and a belt will transfer vibrations too but with a weaker force than chain...as Eric was touching on. I however don't see how the chain would cause a valvetrain failure, as the vibration has to go through the cam and then into the valve (excluding rocker arm). The fluxuation of lift via vibration causing rotation, I would assume, would be marginal compared to force it sees at initial lift or more so closing.

My concern with vibration and a belt would be premature failure of the belt do to the added forces on the fibers of the belt. That or vibration becoming strong enough to cause jumped timing.
SkullWorks
SkullWorkslink
Tuesday, June 21, 2011 8:29 AM
Mike,

I was talking about the same motors, but Didn't think Jim would appreciate any specifics, so i just pointed out where they start...FWIW they end up at a slightly better ratio too...with all that stroke
Mike Kojima
Mike Kojimalink
Tuesday, June 21, 2011 10:40 AM
I think in CART and F1 engine development, cam drive coupling with crank harmonics was causing a bunch of issues with the valve train, hence the center of engine drives that were developed.
Micah McMahan
Micah McMahanlink
Wednesday, June 22, 2011 5:55 AM
Mike, I'll have to look into it. Might just email Ian Bramsey about it and see what he know's/can share.

Also, in regards to those stroker 3.5L NSX engines, what is the point of failure and limitation with revs?

Are they just running the 91mm stroke crank of the C35A? (93mm bore would put it at 3.7L. Note, the C35A does have a taller deck height, wink wink. I know it's compatable with the Legend's C32A.

Jamal, you do have a point on custom parts. Without paying the 1 week expediting fee (or any other expedited level available), custom pistons and rods do take usually 6-8 weeks. That's why I'm a fan of back up engines for racing. The other perk is if your engine lasts longer, don't those pricier parts pay for themselves by not constantly having to rebuild ;)
Micah McMahan
Micah McMahanlink
Wednesday, June 22, 2011 7:57 AM
Jamal, the one other thing that I just realized I didn't mention was the old addage you are referring too. When looking at making power in a motor, cubes aren't all that matter. Take into account, the dwell, VE's, combustion efficiciency and static compression ratio that can be run on the same given fuel and manifold pressure.
So, for example we'll use an EJ, since that's what WE are talking about. Well take a billet 83mm crank (becoming known for failing) and a 75mm crank and same bore. So the 2.58L has a 9.6% increase in displacement over the 2.33L engine. Now the longer the rod ratio, the higher the static compression ratio you can get away with as compared to the stroker and it CAN also flow more air with the same given duration and lift camshaft. Now take into consideration it's increase in dwell and the fact that you can pull timing out. You now have more area of your combustion pressure happening more RAPIDLY after TDC and more pressure AFTER TDC and not before (like a shorter rod ratio). PCP is fun.
So add up the higher volumetric efficiency of the long rod, it's higher static CR (that it can handle with same fuel and boost), it's better 'use' of combustion pressure and it's inherent lack of abuse at high RPM...so does that outweigh the rebuilding and abuse that goes with trying to make 9.6% more displacement make power at RPM worth while?

I know you're engine package is working well for you guys and I am in your corner (just because JC is funny :p), so please don't take this the wrong way.
Wrecked
Wreckedlink
Wednesday, June 22, 2011 8:34 AM
Can we see a plot of piston speeds for the two rod examples?
Micah McMahan
Micah McMahanlink
Wednesday, June 22, 2011 8:44 AM
Same links as above post from yesterday. First time posting pics on here, so let's see if this workds.

[img]http://i108.photobucket.com/albums/n22/drummerboy1784/Pistonvel-time.jpg[/img]

[img]http://i108.photobucket.com/albums/n22/drummerboy1784/Pistondisp-time.jpg[/img]

[img]http://i108.photobucket.com/albums/n22/drummerboy1784/Pistonacc-time.jpg[/img]

[img]http://i108.photobucket.com/albums/n22/drummerboy1784/Pistonvol.jpg[/img]

[img]http://s108.photobucket.com/albums/n22/drummerboy1784/?action=view¤t=piston_dynamicssubiecomparison.jpg[/img]


Micah McMahan
Micah McMahanlink
Wednesday, June 22, 2011 8:57 AM
well that falied...lol
willscarcast
willscarcastlink
Sunday, June 26, 2011 6:31 AM
great read but a little over my head. good comments too!
mxpop
mxpoplink
Tuesday, June 28, 2011 3:03 PM
"We'll take a billet 83mm crank (becoming known for failing)"

Our only crank failure in 5 years was the result of an unrelated OEM part failure causing our crank to break, not the crank's fault itself. We have successfully used OEM, balanced OEM, Cosworth stroker and BC stroker cranks, all with great results.

We have not had the same non OEM part fail on us twice, we are learning a few things as we push into uncharted territory.

Micah, I welcome you to show us how we can do better within our budget and other constraints. Jamal is free to give you whatever data we have available and I look forward to your suggestions.

Check out the upcoming Modified magazine coming out next week. It should reveal quite a bit about what we are all about ;-)

http://speedhunters.com/archive/2011/06/15/magazine-blog-modified-gt-gt-august-issue-preview.aspx
awdaltima
awdaltimalink
Tuesday, June 28, 2011 11:14 PM
Where can I find the program that allows me to produce the same charts and graphs as your article? I'd like to do this analysis on my own engine.
Micah McMahan
Micah McMahanlink
Wednesday, June 29, 2011 7:37 AM
JC; that comment wasn't aimed at you as I know the details about yall's issue. It was just a simple comment on the many numerous failed BC and K1 billet cranks that I've seen/read of.
When I'm not so swamped I plan to check out your data, well when you have good data...don't recall what the issue was with your May data. I actually did want to see what I can find and make some suggestions, hence why I had asked for it in the first place :)

awdaltima; You can do all the models I did above with excel and simple crank-slider model. I went on to MatLab as I finally got a seat that I could use, again. Actually I think there are some sites online that have the crank model done for you and you just punch in numbers...but keep in mind it'll only spit out numbers. What you did with the numbers/data is what matters.
mike156
mike156link
Wednesday, June 29, 2011 8:50 AM
It was mentioned before, but it is important to recognize the importance of how one change can impact other things. Maybe less obvious things though. People talk about rod ratios and being able to increase RPM range and it isn't that significant.

As an example, 150mm rod in a 4G63 vs. 156mm rod (1.70 vs. 1.77 rod ratio). If you base it on peak rod load, it makes a <50 RPM difference in RPM capability. 159mm rod (1.81 rod ratio) makes a 60 RPM difference. Not much of a difference. Admittedly though, that doesn't account for changes in weights of the piston and rod and on the 4G63 anyway, the 156mm rod and piston does end up overall lighter then the 150mm version which will have a more significant impact on RPM capability.

One thing people seldom seem to look at though is the mechanical advantage difference due to the angle between the rod and crank. You are giving up mechanical advantage with a longer rod at peak cylidner pressures which means less work actually gets done and more work just goes into pounding bearings.

For every positive, there is a negative on this argument.

I think a lot of the 2.7:1 rod ratio in an F1 engine has more to do with the 2" stroke and needing a rod long enough to reach the piston then anything else, the rod has to be long enough to allow clearance between the skirt and counterweight after all and when you start getting that short of stroke, 0.5" more rod to clear the skirt adds pretty signifigantly to rod ratio. The opposite end of this would be pro-stock motors. They use really low rod ratios as keeping deck height low to allow optimal runner design in the head and intake is more important to them then rod ratio, from what I've read anyway.

Lastly, custom long rods and pistons also means down time. If you race competitively, bearing is part of racing and sitting waiting for a set of custom parts instead of shelf-stock items could make or break your season. IMO, if there is a shelf-stock piston/rod setup that gets a higher rod ratio…I’d still go with the setup that meets your other needs first (compression ratio, crown design, valve clearance, etc.) If all that is equal, then I would go longer rod, if the price difference is minor.
mxpop
mxpoplink
Wednesday, June 29, 2011 9:43 AM
MM, as we all kmow, any part can and will fail. It's not a matter of if but when, especially in racing. I too have heard rumors of crank issues but I dont know the whole story. Are these stories from forum heroes or actual racers? Were these cranks the causal failure or the result of something else gone bad like faulty instal/tune/lube/etc or like in our case a hydrolic situation which was not the crank's fault?

We lost our May data cuz I shut off the master when I coasted off track. Seeing steam/smoke made me concerned of a poss electrical fire so I pulled power off just in case. In hindsight I didnt have to do so but I have been in situations in the past where it was neccessary. I believe we have successful data from SLB and Tuner Shootout tho.
spdracerut
spdracerutlink
Wednesday, June 29, 2011 10:47 PM
@awdaltima, I just used Excel as it's easier to generate graphs than Matlab. The equations are pretty simple too. For more complex stuff, Matlab is great; I still have my student edition. I'm going to do a Part II to this to look at a couple more things, so wait for that.
Micah McMahan
Micah McMahanlink
Thursday, June 30, 2011 9:06 AM
Mike156, I like you. You are spot on-on the values by the way. At 9000 rpm the difference in peak acceleration (leaving TDC...the rod killer) is only 40 RPM for this case. However as you did make sure to point out, the lighter piston and reduced reciprocating mass of the rod (% of weight) add to allow for higher RPM. Heck, the EJ257 at 9000 RPM is seeing only 91* of the max accleration that the 4G63 does.
As for the difference in mechanical advantage, I'd have to argue that the point is too vague to be argued. Yes, rod angle is lost but the difference in PCP values and the degree to which it happens after TDC is what would matter. We need to bust out a heat release curve and then calculate the force on the crank pin via vectors.
You are also correct on the F1 using longer than needed rods for packaging. An IC engine is nothing but balancing pros and cons to get you to your end means. However, why would an inline 4 motorcycle engine have the same issue. If rod played little to no effect (just how it reads in your writing) then why would they need to keep such a tall deck height?The could certainly rotate the trans down a little. F1 could run flatter engines with shorter deck heights. Once again they meet the balanced point for the design criteria of the platform.

JC, they are a few guys that I know through forums, racing circles, etc. No foul play in the failure, just defects if I had to guess.
I'll have to get in touch about your more recent data from SLB or Tuner shoot out :)
Carb'd 86
Carb'd 86link
Wednesday, July 27, 2011 8:20 PM
@Mike, on the long rod 4ag were you referring to valvetrain harmonic issues or similar power output with the setup? And what block did you use to create the long rod combination? I ask because I'm in the early stages of building a long rod setup using a 7a block with a 4a crank and head.
Post Comment Login or register to post a comment.

MotoIQ Proudly Presents Our Partners:

© 2014 MotoIQ.com