e3pres

Lars,
I thought I'd direct this question your way since you seem to be the head geek around here. :) Since I'm in engineering school, all of my older hot rod buddies are constantly asking WHY things work.(I guess it makes up for me asking them HOW things work. :D) I can usually give them a pretty good answer, but this one has me stumped.

I was asked by a guy who has built race cars for years (this guy pretty much invented the IHRA Pro Modified class), why engines work better if they're built square i.e. Bore = Stroke? This is what I have so far:

Heat loss due to friction at the cylinder walls is a function of the Coefficient of Heat Transfer and the suface area between the piston and the cylinder in question over the course of the stroke. If you hold Volume constant in your calculations, bore must be increased more than stroke to yield that desired volume when deviating from a 'square' setup. Thus resulting in an increased cylinder wall area. However, when bore is decreased, and stroke is increased to keep the same volume value, the area actually decreases. From all of the calculations I've pieced together so far, all I can prove is that an increased stroke, reduces thermal efficiency, by allowing more heat loss to the cylinder walls. I think I'm working all around it, but I just can't find the connection between Bore and Stroke and its effect on thermal efficiency, which I'm assuming is the deciding factor in all this.

Do you have any ideas? Any insight would be greatly appreciated.

TIA,

[Modified by e3pres, 10:22 AM 10/24/2001]


[Modified by e3pres, 10:24 AM 10/24/2001]

JSB69

Some food for thought: A 3" X 3" cylinder has a displacement of 21.2 cubic inches and a cyl wall contact area of 28.26 square inches. If bore is increased by .125" and stroke held constant, the displacement and cyl wall contact area are 22.92 ci and 29.44 si respectively. If bore is held constant at 3" and stroke is increased by .125", the numbers become 22.07 ci and 29.44 si respectively. This would suggest that it is good to deviate from square by increasing the bore, IF CYL WALL CONTACT AREA IS THE ONLY FACTOR. The point that you also must consider is the increased mechanical advantage as the crank throw gets longer which provides more force (torque) with the same "explosion" in the cylinder. The next step would be to assume the explosive force increases in direct proportion to the piston area increase and compare the force on the crank to the increase in force resulting from the same explosion and the increased lever length (stroke).

e3pres

I did not consider the difference in the moment acting on the crank. Now I'm more confused than ever!

lars

I caught your post earlier today, but it’s taken me a while to think this through properly. You really asked a mouthful, there, and the answer is more than a line or two…!

What you’re asking about are some fundamental engine PARAMETERS. An engine parameter is anything that defines the basic engine configuration, such as bore, stroke, rod lengths, valve sizes, etc. These parameters affect how much power and torque an engine can produce. These parameters have been manipulated and changed by engine designers as long as the internal combustion engine has been in existence, and they continue to be refined and changed to this day. There are thousands of combinations possible, and every designer who ever put together a combination is certain that HIS (or HER) combination is the BEST. So you’ll get a LOT of argument on what’s really the best or optimum combination, but let’s look at some facts and theory, and compare a few engines against each other….

The engine parameters have to be evaluated in some way in order to determine which are “the best” parameters (since this is the essence of your question…). We will do this evaluation by using three “performance factors:”

1. Torque – Twisting force. The moment of a force applied at distance. For a street-driven vehicle such as a Vette, we consider this to be of prime importance.
2. Horsepower – “Motivating Power.” It is a unit of time rate of work (essentially torque times rpm with a correction multiplier). This is of prime importance in a competition vehicle.
3. Volumetric efficiency – How effective of an air pump the engine is. Volumetric Efficiency, or V/E, is defined as the ratio of actual air pumped by the engine to the actual cylinder swept volume.

Important Parameters
To produce more performance, Detroit has been famous for the implementation of the old racers’ saying, “when in doubt, bore it out.” There’s just no substitute for cubic inches on the street. Whenever the public wanted more performance, Detroit gave them a larger engine.

So, it would seem, that if you wanted twice the power, you could build an engine of twice the displacement. But this is not true. Why?

Volumetric Efficiency (V/E) is the common factor for comparing all Chevy V8 engines. It has been learned (not by me, but by Detroit engineers…) that the V/E curve has the same shape as the engine torque curve….. Think about that – it’s important. V/E is influenced by many things, such as cam timing & profile, valve-area-to-bore-area ratio, intake and exhaust port lengths & diameters, etc. But if we keep all other things the same (ports, cams, manifolds, etc.), the only real engine parameter which affects V/E is stroke. Bore sizes in our Vettes engines change only slightly, but stroke changes quite a bit.

For matters of our comparison, let’s say that head design & flow, manifold efficiency, and other top-end factors remain the same for any given engine, since you would pick and use the best components no matter what your swept cylinder volume might be.

If we plot a curve for volumetric efficiency keeping all factors constant (except stroke), we see that the only engine parameter that affects V/E is the mean piston speed (MPS): When MPS is very low (idle and low-end rpm), V/E is low. V/E improves as MPS increases up to a certain point, then it will drop off quickly. MPS is merely the distance that the piston travels per crankshaft revolution times the crankshaft rpm. In formula form: MPS = STROKE x RPM/6.

We find that for most street-type engines, peak V/E occurs at 2000 feet per minute (fpm) piston velocity. Surprise! This is also the MPS at which peak torque normally occurs. In other words, the more stroke, the lower rpm at which the peak torque occurs (and you always wondered why a 427 could melt the tires off the rims at 1200 rpm….). But this is not the whole story, because actual engine torque is also directly proportional to swept volume (your engine’s CID). In formula form, TORQUE = V/E x CID (with a correction multiplier). Therefore, even though a 427’s torque occurs at a lower rpm, it also has a significantly higher percentage of torque than a 327.

We also know that horsepower is essentially torque times rpm (with a constant multiplier). Peak horsepower for most V/E curves occurs at 2800 fpm MPS where V/E = 0.85 (typical V/E for a healthy normally-aspirated Chevy V8). 2800 fpm MPS is a much lower rpm on a 427/455 than it is for a 327/350. Since horsepower is computed from the formula HP = V/E x CID x rpm/5252, we see that more cubic inches produce more torque, but everything else being equal, resultant horsepower is nearly independent of displacement…! In fact, analysis will show that horsepower is completely independent of displacement as far as V/E is concerned.

So what does this mean….??? (now that I’ve rambled on and on…). It means two things:

1. For the street, run the biggest engine you can. You want torque on the street, and the 427/455 has the largest torque output that’s in a street-usable rpm range.
2. For racing, horsepower is not dependent in cubic inches. There are other factors that affect horsepower other than V/E.

So if total cubic inch displacement is not a concern, then what ratio of bore-to-stroke (the parameters affecting displacement) is the best? Almost all successful racing engines are “over-square,” meaning that bore diameter is larger than the length of the stroke. Clearly, a short stroke relative to bore size is beneficial since it produces less piston drag. Additionally, we see that a large bore allows the use of larger valves with less cylinder shrouding. Compare a few Bore-to-Stroke ratios of popular engines:

Chevy 302 4.00 Bore 3.00 Stroke 1.33 B/S Ratio
Chevy 331 4.024 Bore 3.25 Stroke 1.24 B/S Ratio
Chevy 454 4.25 Bore 4.00 Stroke 1.06 B/S Ratio
Mopar 426 4.25 Bore 3.75 Stroke 1.13 B/S Ratio

Another parameter that enters this picture is the ratio of the connecting rod length to the stroke length (R/S). This ratio determines the maximum piston inertial loading and the optimum crankshaft angle (the number of degrees after top dead center at which the crank throw and the connecting rod are at right angles). We can plot a curve showing that the greater the R/S, the less the maximum inertial piston loading, and the more reliable the engine will be (or the greater the redline will be). The greater the R/S, the further past TDC the optimum crank angle occurs and the higher rpm at which the same maximum torque will occur with resultant higher horsepower (since HP = torque x rpm). Thus we can manipulate bore, stroke and rod length to change the torque and horsepower parameters/potential of an engine.

It is correct, as a few members have stated, that thermal losses in the combustion process are greater when you expose more cylinder wall area due to a longer stroke, but these losses are minimal compared to the losses and gains produced by the parameters discussed above.

I have some great ideas for a new ‘Vette engine. If you have a couple of $M to invest, stop on by for a beer or six, and we’ll design the next generation performance V8…..!!

Dalannex

Here's what I think......

you lost me a while back. :eek: :hat :jester


I guess it's good I never claimed to be a genius. I was kind of thinking I had this motor thing figured out too. :)

e3pres

Wow. I guess I was looking at it the wrong way. I was already expecting a result, to a question I didn't even really know how to solve. I had convinced myself that frictional losses in the cylinder wall had to be the determining factor when in fact it is negligible in the overall analysis. Thanks for all the effort you put into this. You gotta start charging for this kind of stuff. :) I was particularly interested in the V/E explaination that you gave. I'm going to check into that more on my own. As for helping out in the development of the new V-8; count me in. You provide the R&D $$$, and I'll provide the beer. :cheers: Anyway, thanks again for the info. and for taking the lead on the nerd stuff! :D

BTW, I forwarded a copy of your Q-Jet tuning stuff to my friend Osiris77. He wanted me to tell you that his car is running like it never has before!

Ganey

In the beginning some thought square was the way to go. As Lars said there has been a lot of debate on this. Longer stroke increases TQ. For performance (forget gas milage) one wants the most C.I. & oversquare (larger bore). :cool: If the displacement is limited, spin it faster!

Something you guys might find interesting is GM specified the C.I. (350) & bore centers (same as SBC) for the Lotus designed LT5. Lotus lenthened the stroke & made the bore smaller than the SBC 350. This is the same B/S ratio as a 385. :cool:


I have some great ideas for a new CORVETTE engine. If you have a couple of $M to invest, stop by and I’ll design the next generation performance V8or12 !!!
:cool:

SWCDuke

First of all I disagree with your premise. This reminds me of the r/L (crank throw to rod length ratio) controversy that has been around for years. You have to look at the intended engine application.

In Formula 1 the name of the game is maximum top end power from a limited displacement and bore/stroke ratios are about 2:1. Beyond about 5000 ft/min mean piston speed the engine can't breathe (volumetric efficiency falls off rapidly) , so short strokes are required to allow the highest rev potential for maximum power. Short stokes also allow the most valve area in a limited displacement class. The tradoff is that the combustion chamber is shaped like a thin disk, which has a poor surface area to volume ratio making the engine fundamemtally thermally inefficient, but because of the very high revs, the absolute time heat has to escape the combustion chamber is minimized and that's one reason why F1 engines only have a usable power band in the top ten or fifteen percent of the rev range.

At the opposite end of the spectrum is the heavy duty engine. They typically have strokes a bit longer than the bore. They are designed for longevity, which means low revs/moderate mean piston speeds, and experience has shown that a bore/stroke ratio in the range of about 0.8 to 0.9 is the best tradeoff.

In the fifties and sixties, automotive bore/stroke ratios increased in the horsepower race, but they proved to be emission unfriendly (poorer surface area to volume ratio, again), so they were reduced and currently average pretty close to 1.0.

The bottom line is that there are no magic formulas or silver bullets. Changes in engine design parameters lie on a a continuum and experience generally dictates an optimum "range" based the intended use of the engine. Those not trained in IC engine design rarely appreciate some of the subtleties involved with various design charactieristics and seem to want to come up with "rules", but they just don't exist. There are too many independent variables, and you can't just focus on one. There are few black and white issues in engine design. It's mostly shades of gray.

I don't know what textbooks are being used in IC engine courses, today, but I consider Taylor's two volume set to still be the "bible". A paperback set is available from SAE, and he has discussions on both bore/stroke ratio and r/L ratio. Anyone who is serious about understanding IC engines should have the set. Volume I is very oriented toward thermodynamics (lots of math) and may not be intelligible to most, but Volume II has LOTS of practical design information and is a great resource for anyone interested in understanding the subtleties of IC engine design.

Duke





[Modified by SWCDuke, 9:42 AM 10/25/2001]

lars

Duke -
You just said the same thing I said - you used different words. We're both saying exactly the same
thing about piston velocity and its affect on VE, and how you manipulate this velocity and the VE
through the bore vs stroke, thus tailoring this to the rpm range of the engine. You thus build
high-rpm horsepower, or produce reliable, low-rpm torque. I don't think we have any disagreements
on the premisis on this stuff - it's pretty much well-established as we both know.

e3pres

Duke,
Are the books you are talking about: The Internal-Combustion Engine in Theory and Practice? I found them on SAE's site. There only $31.20 each with my membership. If they're that good, I might just go ahead and buy a set.

[Modified by e3pres, 1:17 PM 10/25/2001]


[Modified by e3pres, 1:25 PM 10/25/2001]

SWCDuke

Yes. The author is Charles Fayette Taylor. The late professor Taylor managed the Sloan Automotive Engine Labs at MIT for many years. The current edition has some additional information, particularly on emissions. I think my editions are late sixties.

Most of the information in this book was derived during the years that Sloan was involved with research too improve the performance of military aircraft engines in the thirties through fifties, but contrary to what many believe, little had been learned since then other than emission control. The basics of engine performance that we understand hasn't changed a whole lot in the last fifty years. The big improvements have been understanding emissions, better materials, and electronic engine control. It's interesting that some of the engine "secrets" I've read about in magazines are in Taylor's book!

Although I was accepted to MIT I ended up doing my graduate IC engine research at the University of Wisconsin. I regret not having the opportunity to meet Prof. Taylor, but at least I have his book set.

Duke