Short Pushrods and Tall RPMs in Real World Performance

[nicholastanguma]

For the sake of simplicity, imagine two theoretical engines, both single cylinders, air cooled, carbed, counterbalanced, with three valve heads of equal porting, and bore/stroke 94x72 mm, connected to identical 6 speed transmissions. Furthermore, the rear tires on both motorcycles of these engines are exactly the same, too.

In fact, everything about the engines and their respective motos is exactly the same except for the fact one powerplant is an OHV unit and the other is a SOHC unit.

Clearly both powerplants are of big bore/short stroke, so the pushrods in the OHV mill are actually super short compared to those of a typical OHV engine found in Harley-Davidson or BSA or some other vintage tech machine.

I know, I know, the immediate reaction is to say, "the single overhead cam mill will obviously rev higher, so its power numbers are inevitably going to be higher."

But in the hypothetical scenario is the SOHC engine actually going to be a better performer from idle to redline, or do all things being equal coupled with the fact the pushrods of the OHV engine are so short mean the two engines are equally matched?

[Tom68]

The sohc only makes power higher in the rev range if the head flows more air.

The pushrod motor can keep up it it flows the same air and the valve train is stable.

i.e. an expensive well designed valve train will do it, especially with a single cylinder where pushrods won't cripple port dimensions.

But why ? lower center of gravity ? in a bike that reduces agility.

Rode a Vinduro ( Vintage Enduro ) some years back, I'll see if I can a pic of old mate on a methanol twin, no ground clearance, done a good job but.

Edit...found one, Russel on his '66 Triumph.

Benalla Russel Triumph 66.jpg

[englertracing]

If the port limits high rpm power the valvetrain that produces more area under the curve will should run better. Generally speaking.

[PackardV8]

i.e. an expensive well designed valve train will do it, especially with a single cylinder where pushrods won't cripple port dimensions.

The Penske Ilmor Mercedes Indy engine comes to mind.

[Wolfo]

The SOHC engine is the better performer, there are a lot of smal gains compared to a OHC engine - but just compare the friction difference, even with simplified math you'll find the OHC valve train has roughly the double amount of parasitic losses.

[nicholastanguma]

The SOHC engine is the better performer, there are a lot of smal gains compared to a OHC engine - but just compare the friction difference, even with simplified math you'll find the OHC valve train has roughly the double amount of parasitic losses.

I'm assuming OHC is a typo, and you meant OHV, since you were clear in typing out SOHC.

Otherwise good points indeed, thanks for your input.

[PackardV8]

The SOHC engine is the better performer, there are a lot of smal gains compared to a OHC engine - but just compare the friction difference, even with simplified math you'll find the OHC valve train has roughly the double amount of parasitic losses.

OEMs must consider the parasitic losses to be minimal. There are myriad examples of OHCs which interpose rocker arms and hydraulic lifters in the valve train instead of the simple bucket tappet.

[Tom68]

The SOHC engine is the better performer, there are a lot of smal gains compared to a OHC engine - but just compare the friction difference, even with simplified math you'll find the OHC valve train has roughly the double amount of parasitic losses.

OEMs must consider the parasitic losses to be minimal. There are myriad examples of OHCs which interpose rocker arms and hydraulic lifters in the valve train instead of the simple bucket tappet.

and in fact the simple bucket tappet Sohc limits valve placement, it is also relocated T-model Ford valvetrain.
OHV can utilize needle bearings, better than cam on bucket.

[Wolfo]

I'm assuming OHC is a typo, and you meant OHV, since you were clear in typing out SOHC.

You're right, I'm still messing up the terms sometimes, in my native language it's called a "hanging valve train"


For the argument of roller/ball bearings in OHV valve trains...this reduces the losses in low RPM range a bit, but the sheer mass of moving mechanical parts are creating a much higher friction than in SOHC valve trains.
A study from german automotive company Mahle found losses in the OHV valve train to be around 20-45% higher than in a comparable SOHC valve train.
SOHC designs with finger followers are even superior.

[nicholastanguma]

For the argument of roller/ball bearings in OHV valve trains...this reduces the losses in low RPM range a bit, but the sheer mass of moving mechanical parts are creating a much higher friction than in SOHC valve trains.
A study from german automotive company Mahle found losses in the OHV valve train to be around 20-45% higher than in a comparable SOHC valve train.
SOHC designs with finger followers are even superior.

Danke! Gut zu wissen. I didn't know that about finger followers. Thanks for contributing.

[PackardV8]

Percentages are relevant, but how much is 20% of an already small number?

I do know GM Engineering went through the excercise of computing parasitic losses, among other things, as a means of justifying retaining the OHV8 configuration of the LS series. Marketing wanted DOHC just because, but Engineering proved the OHV was more cost effective, better packaging and very little tradeoff in horsepower or fuel economy for street use.

jack vines

[Tom68]

Whilst it's not the OP's question, "better" and "worse" come up in the answers.

Let's not forget he's talking single cylinder and in this case a SOHC is a simple/better way out, simple crankcase, simple barrel with integral chain enclosure and a relatively simple cylinder head, just needs a cam mount over a OHV.
Most SOHC singles are rocker arm anyway giving good valve placement options, they only become an awkward package when they're 4 and 5 valve.

Honda did a simple pushrod OHV that many would assume from outward appearance is OHC, I said simple but I'm sure my Honda MiniTrail motor was simpler with its OHC.

Honda CG125.jpg

honda-125-new-model-engine.jpg.89137b6966ea0a9e24159a5bf3ab9e90.jpg

[nicholastanguma]

I'm assuming OHC is a typo, and you meant OHV, since you were clear in typing out SOHC.

You're right, I'm still messing up the terms sometimes, in my native language it's called a "hanging valve train"


SOHC designs with finger followers are even superior.

I'm again assuming SOHC is a typo, and you meant DOHC, since I don't see how finger followers can be utilized with a single overhead cam design. As far as I know, only double overhead cams can act directly on valve springs without the use of rocker arms.

However, if I am wrong, please correct me, as I would very much like to know how finger follower design could successfully replace the typical rocker arm design typically associated with SOHC engines.

[nicholastanguma]

Percentages are relevant, but how much is 20% of an already small number?

I do know GM Engineering went through the excercise of computing parasitic losses, among other things, as a means of justifying retaining the OHV8 configuration of the LS series. Marketing wanted DOHC just because, but Engineering proved the OHV was more cost effective, better packaging and very little tradeoff in horsepower or fuel economy for street use.

This, absolutely this.

As with any gearhead, my rational mind and my tech spec emotions are always battling over what's REAL WORLD performance on the street vs bench racing performance on paper.

Everyone knows OHC is technically superior to OHV, but for anyone except the marketing department and professional track racers the margins of improvement for REAL WORLD street performance simply don't really matter.

[Wolfo]

I'm again assuming SOHC is a typo, and you meant DOHC, since I don't see how finger followers can be utilized with a single overhead cam design. As far as I know, only double overhead cams can act directly on valve springs without the use of rocker arms.

However, if I am wrong, please correct me, as I would very much like to know how finger follower design could successfully replace the typical rocker arm design typically associated with SOHC engines.

SOHC finger follower designs are quite common, but more in the motorcycle world.
See for example a KTM 450 SXF cylinder head. Or several Honda designs (marketing name "UNICAM") for 450ccm engines.
There are a lot of engines like this, but these are recent examples.
Some theory could be found in Mahle publications and books from Ludwig Apfelbeck.

Too expensive for automotive engines, so only common in motorcycle engine design.