best engines in the world?

[berfinroy]

I'm probably going to provoke some strong opinions here, but I have had a number of OHC engines-3 Mercedes among them- and I am hard pressed to grasp the advantage of OHCs for a passenger car motor vs a well designed pushrod engine, of which US manufactures have long been the masters. The OHC engines are more expensive to make and maintain, often have cam chain/belt and follower wear issues that pushrods engines don't, and offer little if any efficiency advantages over (and I stress this)a well designed pushrod engine. It's almost as if Mercedes and other builders of high class cars consider the pushrod design beneath their dignity. A case in point: the Buick (GM) 3.8 V-6. Light, compact, efficient, and bulletproof, especially the series II engine that came out in 1996. And the supercharged version makes serious usable power all over the rev band. 250,000+ no hassle miles is considered pretty ordinary for these engines, and there are other American pushrod engines that are just as good. Of course, the US auto industry has been moving away from pushrods lately too, but that's probably more a matter of sales gimmickry than engineering necessity. Maybe the US auto industry should be religated to making engines for the rest of the world!

[JonL]

I don't disagree with you. Overhead cam engines are superior for higher rpm operation. European engines have traditionally been made to be small displacement and lightweight to match the small, light cars they have powered. One way to get more horsepower out of a smaller displacement engine is to allow it to rev higher. So OHC configurations became the norm. American engines, on the other hand, were much bigger, heavier, and matched to big, heavy cars. They needed a lot of power available at low RPM (a "torquey" engine). So they were able to use the easier to manufacture and less expensive pushrod arrangement. So one arrangement is not "better" than the other. It depends on how the engine is going to be used.

[Skippy]

I've wondered about that myself. It used to be that if you wanted good high rpm capability, you had to go OHC. However, they have pushrod engines that will pass 6000 rpm and have for 20+ years. For a typical passenger car engine, and especially a diesel, I don't see the benefit of having an overhead cam. Perhaps someone with engine design experience would care to enlighten us?

[JonL]

Yes, there are pushrod engines that can exceed 6000 RPM, but there are compromises. The whole valvetrain gets flexible and isn't precisely controlled at that speed, so the cam profile has to be compromised to allow the speed. With OHC, you can go much faster. Also, for a production engine, even if it is rated at 6000 RPM, you probably wouldn't want to run a pushrod engine there for very long.

I surmise that OHC for automotive diesels is partly out of a desire to have a more precisely controlled valvetrain. Can open and close the valves more quickly for better performance. Also less noise.

A big part is also what I'll call "industrial inertia." MB makes a lot of OHC engines and has for years. They know how to do it. When they design a new engine, they have an enormous engineering experience base to draw upon. It may not be worth the risk of making a more radical design change even if the cost savings is large. Especially when the perception (correct or not) is that going to pushrods is a step backwards. Marketing plays a big part.

I'm also not sure that an inline OHC engine with a timing belt is more expensive to build these days than a pushrod engine. Especially with four valve heads.

[Tomguy]

I dont know of a GM 3800 engine (Supercharged or NA) that hasn't needed a head gasket around 100k miles. Same with a 3100.

My 4.5 engine had 145k miles before I did the chain. It did have 12° of stretch - but it also had aluminum-backed rails so could have probably gone longer, I figured I'd do it to do it and that it would allow me to write a nice DIY article. Still dont know why MB switched to plastic rails!

My 4.5 was rated for 230 HP and 280 ft/lbs of torque. That's 51.11 HP / liter and 62.22 torques / liter. Compare that to my pushrod V8 5.7 hemi that's rated for 340 HP and 390 ft/lbs of torque: 59.65 HP / liter and 68.42 torques / liter. My cast iron 4.5 was designed and first mass produced in 1970. My 5.7 was designed and first mass produced around 2004. You would think 34 years would net a larger percentage gain!

Now, look at modern MB engines - 5.5L n/a V8 produces a fair amount more than my 5.7 n/a V8...

[mbgingerbread]

I am going to wade in here even though I must admit I don't have engineering degrees or a wealth of technical knowledge by way of the automotive engine.
First, I do agree that the GM engine is a great piece of work-we have had several in different configs over the years. There is an unmistakable feel associated with the fact of getting in and going. The handling soon plays out after time and one feels things getting a little sloppy.
Second, having owned and driven a few MB's there is this sense which speaks to the driver and says: precision. In the generations following each other weight might have declined but handling performance remained static or improved in this relation. The fact that an engine is balancing itself vertically instead of in the V configuration lends IMO the subtle yet dynamic sense of balance which is counterpoint to the constant forces being exerted by changes in direction of the frame and body. As a result if the forces inherent in the inline engine remain overwhelming this gives the driver the feeling that there is a continous force maintaing the balance needed to remain upright.
I asked for and received the wonderful book ' 100 Greatest Cars ' for X-mas one year. Almost without exception the sense of vertical balance comes thru in a way to promote stability and this is made so appealing by the coachwork produced by the great car makers of the world over time.
Turner

[wolf_walker]

The LS motors are hard to beat. Especially for the money. Price a new crate v8 from BMW or Mercedes and price a comparable LS1. It isn't elegant but it works.
I'd love one of the 6+ L LS truck motors with a few mild mods in a w126 personally.
Oh, and with a trans that does not make me want to shoot it sixteen times a day.

[berfinroy]

Originally Posted by Tomguy

I dont know of a GM 3800 engine (Supercharged or NA) that hasn't needed a head gasket around 100k miles. Same with a 3100.

My 4.5 engine had 145k miles before I did the chain. It did have 12° of stretch - but it also had aluminum-backed rails so could have probably gone longer, I figured I'd do it to do it and that it would allow me to write a nice DIY article. Still dont know why MB switched to plastic rails!

My 4.5 was rated for 230 HP and 280 ft/lbs of torque. That's 51.11 HP / liter and 62.22 torques / liter. Compare that to my pushrod V8 5.7 hemi that's rated for 340 HP and 390 ft/lbs of torque: 59.65 HP / liter and 68.42 torques / liter. My cast iron 4.5 was designed and first mass produced in 1970. My 5.7 was designed and first mass produced around 2004. You would think 34 years would net a larger percentage gain!

Now, look at modern MB engines - 5.5L n/a V8 produces a fair amount more than my 5.7 n/a V8...

I believe the HP rating system was different back when the 4.5 was first made--SAE gross vs the later SAE net rating. Also, I think the 230 rating was for the (relatively) high compression euro version of the 4.5. The US version, at least the early ones, had only 8:1 compression and were rated at more like 190-200 HP.

Regarding high RPM operation, NASCAR engines are all pushrods, aren't they?

[Tomguy]

Originally Posted by berfinroy

Regarding high RPM operation, NASCAR engines are all pushrods, aren't they?

Exactly why they don't go over 10k RPMs when F1 engines would be well over 20k RPMs if they didn't limit them in 2006 to 18k!

[Mark DiSilvestro]

It's true that for many years foreign manufacturers built predominantly small high-revving engines where an OHC has some advantage, but today, many cylinder-head designs have three or four valves per cylinder, which is also easier to address with an OHC setup.
My biggest gripe with many OHC designs has been the use of timing belts, especially on interference-engines. Thankfully, some manufacturers have moved back to timing chains.

Happy Motoring, Mark

[deanyel]

Originally Posted by berfinroy

I believe the HP rating system was different back when the 4.5 was first made--SAE gross vs the later SAE net rating. Also, I think the 230 rating was for the (relatively) high compression euro version of the 4.5. The US version, at least the early ones, had only 8:1 compression and were rated at more like 190-200 HP.

4.5s of that era were 230 hp SAE and 200 hp DIN.

[SPDMETL]

Lifelong GM fan here...the LS engines are a work of art ! The current Ford 4.6/5.4 engines are OHC

[Mike D]

Okay, let's do the math.

Push rod engines= 1 lifter, 1 push rod, 1 rocker per valve times cylinder.

OHC engines= 1 rocker arm per valve times cylinders and, uhh, hmm that's it!

DOHC engines add one more piece, another cam to the TOTAL.

Pushrod V-8 = 49 parts working to make the valves go ooompity-boompity.

OHC V-8 = 17 parts.

OHC 4 valve per cylinder = 33 parts

DOHC V-8 = 18 parts.

DOHC 4 valve per cylinder = 34 parts

Now which system do you figure is the most reliable?

Granted the length of the timing chain requires usually 2 or 3 guides and a tensioner so add 4 more parts to the totals if you really want to get picky..

I don't want to hear the talk about timing chains. You guys must have forgotten the fiber gear fiasco's of the 60's and 70's American iron. 60-70K was about norm for a timing chain replacement.

[t walgamuth]

The pushrod engine is much much simpler to build and work on, but in normal cars using normal materials building the parts, you are going to be limited in operation around 6000 rpm or so. Thats about what a production push rod engine will do if it has solid lifters. Varous studebaker engines with pushrods did that back in the sixties and fifties. At that time the gm engines typically were limited to lower rpm.

Racing engines have been overhead cam since the twenties or earlier.

In order to sell cars the manufacturers have gradually built engines with technology that mimics racing engines even though the benefit is debatable.

Designing a pushrod engine?... just design it for high torque at low rpm and all is fine, as long as you gear it to utilize the torque range of the motor.

OTOH if you really want simplicity and don't care about specific output you need to go back to a flathead engine.

Now thats simple.

'Course you would have trouble getting one to run well today with the less volitile fuel, and adjusting the valves down under the manifolds was a triple bit*h too.

Design is all about compromises.

[berfinroy]

And marketing demands "jazz." And that's what overhead cams for passenger car engines was largely about, at least until designers started commonly using 4 valves per cylinder heads (more jazz) which would be pretty hard to do using pushrods!