99 replies to this topic

[m9a3r5i7o2n]

V-8 engines & finger followers.

In the book about the old 1939 Tripoli valve/camlobe system there are two pictures of the fingers in the one picture, page #132, the followers have radii on both sides of the follower. The camlobe, page #134, has a straight sided lifting area. That is one on the lift side and one on the down slope side.
Would it not have been better to have made the cam with an all curvature working surface? Then dump the finger followers. After all there are 16 of the little gems along with an unknown amount needle bearing, pins etc. etc.
Just how much these finger followers are used today is probably an unknown.
QUICKSILVER by Cameron Earl ISBN 0 11 290550 First published 1948
HMSO,PO Box 276, London SW8 5DT

M.L. Anderson

[Halfwitt]

Originally posted by m9a3r5i7o2n
V-8 engines & finger followers.

In the book about the old 1939 Tripoli valve/camlobe system there are two pictures of the fingers in the one picture, page #132, the followers have radii on both sides of the follower. The camlobe, page #134, has a straight sided lifting area. That is one on the lift side and one on the down slope side.
Would it not have been better to have made the cam with an all curvature working surface? Then dump the finger followers. After all there are 16 of the little gems along with an unknown amount needle bearing, pins etc. etc.
Just how much these finger followers are used today is probably an unknown.
QUICKSILVER by Cameron Earl ISBN 0 11 290550 First published 1948
HMSO,PO Box 276, London SW8 5DT

M.L. Anderson

To my knowledge most (I believe all) current F1 engines use finger followers in order to get the huge lifts required, where this would not be possible with direct-acting valve actuation. Road cars are also turning back toward finger followers in an attempt to decrease friction.

[cosworth bdg]

Originally posted by Halfwitt


To my knowledge most (I believe all) current F1 engines use finger followers in order to get the huge lifts required, where this would not be possible with direct-acting valve actuation. Road cars are also turning back toward finger followers in an attempt to decrease friction.

Your assumption on F1 engines is very correct , also road car engines are moving that way also, both petrol and diesel versions.....................................

[m9a3r5i7o2n]

Although it is not easily seen without the drawing of the two components the use of an angle is very wrong. Something that the man who drew the original in 1939 would have had difficulty seeing is that an angle would cause interference when either opening or closing the valve if the follower were flat. However he drew it with curves on both sides of the follower. If the F-1 engines have cam lobes with no angles just why would they need fingers followers? Modern cam design could easily stay away from angles on the lobes! All sorts of curves could be used and they would not have to stay with simple radii.
M.L. Anderson

[McGuire]

Could you post a drawing or photo of what you mean? I am pretty sure I am not following you correctly.

[m9a3r5i7o2n]

Quote from McGuire; Could you post a drawing or photo of what you mean? I am pretty sure I am not following you correctly.

I have a very good redrawing of the cam lobe itself which I will post tomorrow. It is a very good sketch and shows very clearly the angle and all the dimensions on that drawing.
However I will try to draw the finger follower of which I will have to draw from an observation of it. These sketches are very small and that particular part has no dimensions so it will be more or less free hand. Remember these sketches in the book are small and are impossible to scale and when “blown” up are too fuzzy to really make a scaling of the part. They very definitely have dual radii. Of that there is no mistake.
M.L. Anderson

[rhm]

I believe the Peugeot/Asiatech engine was the last in F1 to have the camshaft lobes acting directly on bucket tappets. That engine (according to RET) used a small compound valve angle and that machining the lobes for the camshaft was problematic and it also has a lot of friction because the area where the lobe contacts the tappet is moving in two axis instead of one with a more conventional engine.

I expect in 1939, even though the performance of engines was comparatively low then, the engine you're looking at was high performance (for the time). It's most likely that the quality of lubricants they had then and the surface hardness they could achieve on lobe and tappet made direct acting overhead cams impractical above a certain speed. Most engines of the time would have been pushrod based and there the rocker mechanism gives a similar advantage to the finger-follower.

When lubrication and surface finishing improved, the move to direct-acting overhead cams was obvious, but even today the highest performance engines use finger-followers to reduce friction (and in the case of F1 to also deal with compound valve angles). Although direct-acting cams are viable up to at least 15,000rpm (that's what Champcar engines topped out at and that Peugeot F1 engine was past that level as well), in a road-car engine where the mileage requirement is in hundreds of thousands of miles between rebuilds, the level is much lower. Honda VTec engines for example rev to 9,000rpm and use finger-followers (even though they are described as "rockers" in much of the literature, the geometry is that of a finger-follower).

Does anyone know if Ferrari's roadcar V8 engines (from the 360 and F430 for example) use finger-followers?

[Engineguy]

Originally posted by rhm
Does anyone know if Ferrari's roadcar V8 engines (from the 360 and F430 for example) use finger-followers?

I don't know about the 360 engine (it shares NO parts with the F430), but the NEW F430 has direct-acting buckets.



Arguably one of the most exotic modern production engines available (2005 International Engine of the Year, Best New Engine 2005, Best Performance Engine 2005, Best Above 4-Litre Engine 2005), the BMW M5 V10 (and the new M3's V8 version, I believe) has... ta dah... a cylindrically radiused contact surface bucket follower with anti-rotation key, just like the Offy (only a few Offys had flat followers). The key is needed, of course, just like with a pushrod roller tappet.



A close-up of the cam follower... note lasered lobe oiler hole... Offy followers and Ford Indy followers (made by Offy) had key silver-soldered on; I don't know how BMW retains it:




The NEW AMG MB V8 (which is not a modified high-volume MB V8) also has direct-acting buckets.

Three NEW top of the line performance engines from Ferrari, BMW, and AMG-MB... all with direct-acting buckets. Hmmm

[McGuire]

Originally posted by rhm


I expect in 1939, even though the performance of engines was comparatively low then, the engine you're looking at was high performance (for the time). It's most likely that the quality of lubricants they had then and the surface hardness they could achieve on lobe and tappet made direct acting overhead cams impractical above a certain speed. Most engines of the time would have been pushrod based and there the rocker mechanism gives a similar advantage to the finger-follower.

Bucket followers were in fairly common use before 1939. (Ballot, Fiat, Aston-Martin, Bugatti, Miller, Offy, Duesenberg etc.) I think that may be in part where Marion is going here... first of all why a finger follower instead of the cup-type follower that was already prevalent by then?

"Direct-acting" may be a bit of a misnomer. Sound practice is a follower between the cam lobe and the valve stem, be it a bucket, a finger, or rocker arm, so the valve stem does not have to take all the side loadings. Buckets are compact but fingers can offer less reciprocating mass.

Mercedes often used finger followers in that period but Marion seems to have detected something unusual about the setup used in the 1.5 liter V8. I admire Marion's tenacious approach to studying things. He will seize on subjects normally accepted as givens and research them until he knows them intimately from top to bottom.

[McGuire]

Originally posted by Engineguy



A close-up of the cam follower... note lasered lobe oiler hole... Offy followers and Ford Indy followers (made by Offy) had key silver-soldered on; I don't know how BMW retains it:

I dunno... looks like Uber-glue.

[m9a3r5i7o2n]

Quote from engineguy;
cylindrically radiused contact surface bucket follower with anti-rotation key, just like the Offy (only a few Offys had flat followers).

Offy had plenty of problems with the lobes in the 1950s! One only has to read the Book,”Offenhauser”, page # 143 to see the reason. Very flat angled cam lobes with small nose radius and too little open duration for the type of racing engine. There are also two pictures of the lobe for flat and radius cam followers. Very poorly designed
They should have left the cam design to a specially manufacturer. Remember this was before computers because common.
M. L. Anderson

[m9a3r5i7o2n]

Below is the 1939 Intake valve camlobe. Please remember the man that did this originally did not have a computer!!!


http://home.comcast....EDES284DEG.html

http://home.comcast....LLOWER.JPG.html

Done from obsevation only, no dimemsions on sketch.

M.L. Anderson

[hydra]

Originally posted by Engineguy

Three NEW top of the line performance engines from Ferrari, BMW, and AMG-MB... all with direct-acting buckets. Hmmm

Doesn't matter whose name is on the valve cover, that doesn't mean its the best way to do it ;)

[NTSOS]

I would think that the most important factor in using roller finger followers for *ultra*
speed RPM engines would be:

A - a major reduction of valvetrain mass above the valve due to elimination of bucket
weight.

B - followed by lower high rpm friction due to the reduction of lateral forces on the valve
stem and/or oil shear between a lobe/bucket combo.

John

[Engineguy]

Originally posted by hydra
Doesn't matter whose name is on the valve cover, that doesn't mean its the best way to do it ;)

Engine design is, as you know, a whole bunch of compromises. That's why there exists, in currently-produced engines, examples of sliding vs rolling cam contact, linear moving tappets vs pivoting fingers and rockers, etc., etc. There is no "best" solution... I could argue various merits of any of them... I've designed 'em all and love and hate and start over and switch. The valvetrain does not exist in a vacuum... does the 0.03 MPG improvement in fuel economy or 2.4 HP from the friction reduction justify the cost, assembly complexity, cylinder head bulk and weight and parts count that result from rollerized finger followers? How many customers care to pay an extra $400 for that? What if the bulk of the head compromises the exhaust system a little? Or the extra weight is added to already tortured front tires?

Mario Illien said, around 2002, the gain from using finger followers vs buckets was very small even in the IRL/CART/F1 engines they produced. They allow a slightly smaller base circle diameter (less friction, less weight) for a given lift (less so vs curved face bucket) and very slight reduction of recipricating mass,and endless opportunity to fiddle with (or F--- up) progressive ratios (i.e. valve movement not constant ratio to cam profile).

Engine designers are not immune to engineering fads, and going overboard. They should constantly step back and look at the big picture. Outside the realm of the pure racing engine where 753 HP is better than 751, simplicity itself is often a lofty goal that brings desirable aftereffects. Perhaps that's what Ferrari, BMW, and AMG saw in their aforementioned recent design decisions.

[Stoatspeed]

Originally posted by Halfwitt


To my knowledge most (I believe all) current F1 engines use finger followers in order to get the huge lifts required, where this would not be possible with direct-acting valve actuation.

For a while, Ferrari attempted to get around this in F1 engines by developing the so-called "ballistic" cams which actually launched the the valvetrain off the peak of the cam that it deliberately lost contact (and hence giving valve lift in excess of cam lift) at high rpm. Of course, the trick was catching the valve on the way back down without major damage ..... interesting challenge to the mechanical analysis software!
Since Ferrari were doing it, it would be fair to guess that most other teams had some version of the same system - the insight came from a senior Honda F1 engineer of the time.

Dave H

[m9a3r5i7o2n]

This will show just how much camlobe design has come since 1921, Miller intake lobe. Note the red sides are angles! Also notice the very small radius on the nose.

http://home.comcast....248Din1921.html

M.L. Anderson

[Engineguy]

Originally posted by m9a3r5i7o2n
This will show just how much camlobe design has come since 1921, Miller intake lobe. Note the red sides are angles! Also notice the very small radius on the nose.

http://home.comcast....248Din1921.html

M.L. Anderson

Although such a flat will "slap" a flat face follower, transferring the point of contact from the base circle (and center of follower face) to start of the nose radius (and near outer edge of follower face) instantly (and go instantly from zero follower velocity to very rapid lift... i.e. brutal "near infinite" acceleration), there is no such problem when used with a convex curved face follower as the Miller had (IIRC). The curvature of the follower takes up the role of easing into the lift ramp. That's not to say that by some miracle coincidence the simplicity of the cam shown was by any means the ideal profile. If you think about how the pattern for a cam profile had to be made back then (manual mill and offsetting point of rotation on a dividing head?) you can appreciate the simplicity.

[desmo]

Originally posted by Stoatspeed


For a while, Ferrari attempted to get around this in F1 engines by developing the so-called "ballistic" cams which actually launched the the valvetrain off the peak of the cam that it deliberately lost contact (and hence giving valve lift in excess of cam lift) at high rpm. Of course, the trick was catching the valve on the way back down without major damage ..... interesting challenge to the mechanical analysis software!
Since Ferrari were doing it, it would be fair to guess that most other teams had some version of the same system - the insight came from a senior Honda F1 engineer of the time.

Dave H

Never got past the prototype stage apparently. Here's a pic I took in Maranello, the head is dated 1995.

[hydra]

Engineguy,
By "best", I meant best bhp-wise. Let us forget the friction, and valvetrain weight advantages of finger-followers (which are minimal and are of little relevance to most road cars) And you're right, finger followers would probably impose a packaging penalty, making already heavy and bulky heads heavier and bulkier. No, the main advantage of finger followers, one that I don't think is ever utilized on any road car engine, is the ability to use L/D ratios of well over 0.25 to maximize high-rpm VE with little detriment to low-rpm torque... Instead of having a 120bhp/L 9000rpm cam on bucket engine, you could have an equivalent 130 or 140bhp/L finger follower engine running more lift... Or is that not a reasonable argument?

[Halfwitt]

Originally posted by desmo

Never got past the prototype stage apparently. Here's a pic I took in Maranello, the head is dated 1995.

Isn't that a desmo head desmo? The cam lobes seem super wide in parts, and seem to have two profiles in other parts.

I'd heard the 'ballistic' cams were commonplace in NASTRACTOR racing, but I didn't know that they were used in F1

[Engineguy]

Originally posted by hydra
Engineguy,
By "best", I meant best bhp-wise. Let us forget the friction, and valvetrain weight advantages of finger-followers (which are minimal and are of little relevance to most road cars) And you're right, finger followers would probably impose a packaging penalty, making already heavy and bulky heads heavier and bulkier. No, the main advantage of finger followers, one that I don't think is ever utilized on any road car engine, is the ability to use L/D ratios of well over 0.25 to maximize high-rpm VE with little detriment to low-rpm torque... Instead of having a 120bhp/L 9000rpm cam on bucket engine, you could have an equivalent 130 or 140bhp/L finger follower engine running more lift... Or is that not a reasonable argument?

IF you could get that 10% bhp/L increase... but any additional lift beyond what is feasable with the bucket mass/bucket movement characteristic does not yield anything resembling proportional additional power... maybe 1% from 10% lift increase (IF you could get that much). More like an increase from 120 to 121 bhp/L. If you want to tolerate that extra bulk and weight, you'd probably get more power staying with buckets and increasing the displacement slightly.

[cosworth bdg]

Originally posted by Engineguy
Engine design is, as you know, a whole bunch of compromises. That's why there exists, in currently-produced engines, examples of sliding vs rolling cam contact, linear moving tappets vs pivoting fingers and rockers, etc., etc. There is no "best" solution... I could argue various merits of any of them... I've designed 'em all and love and hate and start over and switch. The valvetrain does not exist in a vacuum... does the 0.03 MPG improvement in fuel economy or 2.4 HP from the friction reduction justify the cost, assembly complexity, cylinder head bulk and weight and parts count that result from rollerized finger followers? How many customers care to pay an extra $400 for that? What if the bulk of the head compromises the exhaust system a little? Or the extra weight is added to already tortured front tires?

Mario Illien said, around 2002, the gain from using finger followers vs buckets was very small even in the IRL/CART/F1 engines they produced. They allow a slightly smaller base circle diameter (less friction, less weight) for a given lift (less so vs curved face bucket) and very slight reduction of recipricating mass,and endless opportunity to fiddle with (or F--- up) progressive ratios (i.e. valve movement not constant ratio to cam profile).

Engine designers are not immune to engineering fads, and going overboard. They should constantly step back and look at the big picture. Outside the realm of the pure racing engine where 753 HP is better than 751, simplicity itself is often a lofty goal that brings desirable aftereffects. Perhaps that's what Ferrari, BMW, and AMG saw in their aforementioned recent design decisions.

With the secrecy surrounding current F1 engines, the people on this forum can only guess at what is being done and used in current designs...

[NTSOS]

Does not a larger cam base circle allow a more aggressive/reliable high rpm cam profile to be utilized? It seems that pro stock engines are always increasing the cam bearing diameter so that a larger base circle can be incorporated for reliability at higher lifts and relatively shorter durations.....the overall valve lift and RPM capability of a pro stock motor are truly astounding. I wonder if that relates to a finger type F1 valve train in any way? No expert here, just a question/observation.

John

[McGuire]

Looked around in my pile and found an illustration of the engine to which Marion refers, the Mercedes-Benz M165 1.5 liter "Tripoli" V8...

[desmo]

Originally posted by Halfwitt


Isn't that a desmo head desmo? The cam lobes seem super wide in parts, and seem to have two profiles in other parts.

I'd heard the 'ballistic' cams were commonplace in NASTRACTOR racing, but I didn't know that they were used in F1

It isn't a desmo, but this is:



I'd imagine once an interesting idea like ballistic valvetrain is in the air, word will spread. Ferrari obviously thought it was worth taking past the mule stage which suggests that it probably worked pretty well as a single cylinder mule engine. But then Ferrari tested all kinds of interesting and sometimes, heck often, downright eccentric engine concepts in the '80s and '90s. They'll have an interesting collection on view at any given time in the museum there, apparently the items on view are often changed so no telling what you'll see.

It'll be good though.

[desmo]

One more heh.

Ever seen a Ti block? Same year as the ballistic head.

[desmo]

Originally posted by cosworth bdg
With the secrecy surrounding current F1 engines, the people on this forum can only guess at what is being done and used in current designs...

While that is no doubt the case in a lot of the design details, I think we have a pretty good idea of what a modern F1 engine looks like taken apart.



I'd wager the current designs will look pretty similar in basic forms to this 2002 Ferrari engine. At least within the constraints of what is shown in the drawing. I'd say that F1 has often been fairly conservative really as far as the basic layout of the engines. A lot of detail engineering and development accreted empirically.

[hydra]

Finger-followers need not introduce that much extra bulk to begin with.. Look at the Ferrari heads for instance, they're nice and compact.

I did a quick run on EA Pro with a hypothetical hi-performance 500cc/cyl engine with a B/S of 1.31 comparing finger followers (L/D = 0.36) to a cam on bucket setup (L/D = 0.26 which is quite generous actually) The COB cam duration had a little more duration than the FF cam to compensate for the decreased leverage, giving them the same overlap area, and I fiddled with the head flow tables a little to "optimize" them for mid-lift flow (since it won't be seeing sky-high L/Ds) and the difference between the two engines amounted to ~7% more peak power, 4.5% higher peak BMEP, and up to 8 more VE % points! Not bad huh? Mach Index for the former engine was 0.412, and for the latter engine 0.472 @ redline. Valve diameter was as big as the bore would accomodate, some 0.41B.

[rhm]

Originally posted by desmo


I'd imagine once an interesting idea like ballistic valvetrain is in the air, word will spread. Ferrari obviously thought it was worth taking past the mule stage which suggests that it probably worked pretty well as a single cylinder mule engine.

I've read that Ferrari tend not to bother with the single-cylinder test engines.

[desmo]

There were several single cylinder mules on display at the museum when I visited. I didn't take pics of them then because someone had posted a link (probably now dead) here to photos of the same ones I'd seen here just before heh. That and I only had a lousy 64MB card for my camera then.

[m9a3r5i7o2n]

Quote from engineguy;


If you think about how the pattern for a cam profile had to be made back then (manual mill and offsetting point of rotation on a dividing head?) you can appreciate the simplicity.

Really good machinists can do miracles on a Pratt and Whitney Jig Borer or Jig Grinder and one must remember that the making of cam profiles Grand Masters is not a job for run of the mill machinists. A large flat plate bolted to a Pratt and Whitney Jig Grinder Precision Rotary Table can produce offset radii very easily. However I doubt that Miller or Offenhauser had either. I believe that it was beyond the ability of the machinery and not the machinists and the fact is very likely that Leo Goosen didn’t have the time to spend making really good full radiised lobe profiles. At that time it is likely that the only way to do it was to have a very large drafting board and fit the blend points and tangents points in by hand a very laborious task. All this plus calculating the twenty blend/tangents. I just wonder how the automobile manufactures did it in 1920, my guess would be that the Grand Master and possibly the Working Masters were made by a small very special outside vendor who specially was in doing the Grand Masters and the Working Masters.

But this gets into the old statement that a good workman doesn’t blame his tools but makes sure he has the proper tools to do the job correctly. It’s a ring around the rosy type of argument.
At one time I heard of the shop who made the Grand Masters, in Chicago ,I believe and I am hoping that Maguire can come up with this one as he is the only person I have confidence in for having a piece of information such as this.
M.L. Anderson

[McGuire]

Marion, you have way too much faith in me. Really I have no idea. But there are some old-timers around I can ask the next time I see them.

[McGuire]

It is said that the main thing that made the last Chevrolet Indy engine an unworkable turd (and forced them to go out and buy an engine from Cosworth in order to be competitive) was its bucket cam followers. This was really the only truly in-house IRL engine GM ever did -- the original Aurora IRL V8 had TWR and Geoff Goddard looking in.

When the new engine proved to be a dog they scrambled to redo the cylinder head but then mgt stopped them and said no more fooling around, let's buy an engine from someone else and then we'll know it's right. Pretty sad when you look at it. I salute their their commitment to the teams but that is not exactly a vote of confidence in your own engineering.

[McGuire]

Originally posted by Engineguy



Arguably one of the most exotic modern production engines available (2005 International Engine of the Year, Best New Engine 2005, Best Performance Engine 2005, Best Above 4-Litre Engine 2005), the BMW M5 V10 (and the new M3's V8 version, I believe) has... ta dah... a cylindrically radiused contact surface bucket follower with anti-rotation key, just like the Offy (only a few Offys had flat followers). The key is needed, of course, just like with a pushrod roller tappet.

They say Riley Brett originally came up with that.

[McGuire]

Originally posted by desmo


A lot of detail engineering and development accreted empirically.

Yep, one tiny incremental gain and a zillion dollars at a time.

On a totally irrelevant side note... when I look at that end view of the Ferrari V10, I can't help but visualize how tiny and compact it would be as a fork-and-blade L-head. Silly I know but.

[desmo]

Harley went away from the fork and blade arrangement on the V-Rod engine didn't they?

[Stoatspeed]

Originally posted by desmo
Harley went away from the fork and blade arrangement on the V-Rod engine didn't they?

The V-Rod engine had conventional side-by-side rods - mainly because it was designed for them by Porsche! It was also a styling thing - the edict for the air-cooled engines was that having both cylinders in-plane is fundamental to Harley styling, and they are not prepared to compromise. Since the V-Rod was a whole new thing and also the water jacket disguises the alignment (or lack of), the easier solution was OK.
Just a few years ago, they were rumored to be actively investigating an articulated con rod to allow a single crank journal to carry forces from both cylinders .... this is a step towards being able to use plain bearing bottom ends (and one-piece crank) for easier manufacturing and lower cost than the rollers with built-up crank. That little foray apparently did not turn out too welll .... so probably no change in sight for the air cooled engines

[m9a3r5i7o2n]

Quote fromStoatspeed;
--------------------------------------------------------------------------------
Just a few years ago, they were rumored to be actively investigating an articulated con rod to allow a single crank journal to carry forces from both cylinders .... this is a step towards being able to use plain bearing bottom ends (and one-piece crank) for easier manufacturing and lower cost than the rollers with built-up crank. That little foray apparently did not turn out too well.

Considering the shake problem maybe someone at Harley management had heard about the Sunbeam-Arab V-8 with articulated rods and a 180 degree crank!

M. L. Anderson

[Canuck]

Originally posted by Stoatspeed

The V-Rod engine had conventional side-by-side rods - mainly because it was designed for them by Porsche!

Porsche developed the TwinCam and the Evolution engine as well, though with differing levels of input. Really, the V-Rod doesn't have fork/blade rods because they aren't required. With coming EPA regs, we won't have to deal with air-cooled stuff much longer anyway.

Mid-USA markets a Harley-ish v-twin with a one-piece crank from Falicon, plain shell bearings (obviously), side-by-side rods and KB-supplied pistons with the pin designed such that the cylinders are still in line. Works very well, just too expensive for the average buyer.

[Powersteer]

Originally posted by desmo


It isn't a desmo, but this is:



I'd imagine once an interesting idea like ballistic valvetrain is in the air, word will spread. Ferrari obviously thought it was worth taking past the mule stage which suggests that it probably worked pretty well as a single cylinder mule engine. But then Ferrari tested all kinds of interesting and sometimes, heck often, downright eccentric engine concepts in the '80s and '90s. They'll have an interesting collection on view at any given time in the museum there, apparently the items on view are often changed so no telling what you'll see.

It'll be good though.

That thing looks 5 valve to me.

[vvillium3]

In reviewing which of these two options is better. It seems that neither of these two have that much of an advantage over the other as far as performance goes. So then that brings me to think about which of these two are easier employ in a new design. It seems that a radiused bucket type follower would be much more difficult to make and design around than that of a finger follower system. Not only in the part itself, but also in the cylinder head where it mates to. A corresponding cut would have to be cut into the head for the the key to ride in. Unless a insert was made to attach inside the head. Yes a finger system bring some added bulk to a design, but can be easier machined using a cnc, surface grinder, and wire edm.
This brings me to my next question. How are lobe to follower clearences adressed??? No matter how accurate machining is held to, there is going to be somewhat of a difference in lobe to follower clearance from lobe to lobe. I can see a shim being used with the radiused follower, but how would one go about properly mantaining the correct clearance with a finger system????

[desmo]

Originally posted by Powersteer


That thing looks 5 valve to me.

Good eye.

[Engineguy]

Originally posted by hydra Finger-followers need not introduce that much extra bulk to begin with.. Look at the Ferrari heads for instance, they're nice and compact.

I assume you're talking about the F1 heads... where the fingers need only last a few hundred miles. When you have to design for lower forces on the finger and its pivot, rollers, a production tolerance friendly method of lash adjustment or compensation, and cannot impose the need to disassemble the cams and finger base to change a spark plug, etc., the bulkiness creeps up on you. I've spent hundreds of CAD hours trying to compact finger systems... the equivelent bucket system is always a much tighter design.

Originally posted by hydra I did a quick run on EA Pro with a hypothetical hi-performance 500cc/cyl engine with a B/S of 1.31 comparing finger followers (L/D = 0.36) to a cam on bucket setup (L/D = 0.26 which is quite generous actually) The COB cam duration had a little more duration than the FF cam to compensate for the decreased leverage, giving them the same overlap area, and I fiddled with the head flow tables a little to "optimize" them for mid-lift flow (since it won't be seeing sky-high L/Ds) and the difference between the two engines amounted to ~7% more peak power, 4.5% higher peak BMEP, and up to 8 more VE % points! Not bad huh? Mach Index for the former engine was 0.412, and for the latter engine 0.472 @ redline. Valve diameter was as big as the bore would accomodate, some 0.41B.

GIGO. 40% more lift with fingers? Not bloody likely. Unless you're simply restricting yourself to an existing bucket system that has been given geometric constraints that never foresaw a need for higher lift. To which, again... GIGO.

[Engineguy]

Originally posted by McGuire
It is said that the main thing that made the last Chevrolet Indy engine an unworkable turd (and forced them to go out and buy an engine from Cosworth in order to be competitive) was its bucket cam followers.

At least you tempered this with "It is said" and "the main thing." My peeve with this whole episode was that every race fan in the world was suddenly an expert on valvetrains and "knew" bucket followers are turds just 'cause the most obvious difference was that the Chevrolet had buckets. It also had different intake and exhaust ports, different valve stem contours, different water jackets, different pistons, different bearing diameters, different bearing widths, different clearances, a different ring package, different cylinder sleeves, different dry sump pump, different sump baffling, different intake runners, different exhaust headers, different engine management system, and so on and so on. Any combination of hundreds of features of the engine's design (the method of valve actuation being but one) could have made it come up short... perhaps the real reason the boys at Chevrolet weren't allowed to waste more time just doing a new valvetrain?

There does seem to be a small advantage, as I've said, in pure racing engines at the moment, to finger followers, but I believe the size of the advantage is a magnitude less than popular perception. Which makes it just one more design detail to weigh, pro vs con, in the overall package.

IRL Toyotas were turds compred to the Hondas in 2005. Toyota must, therefore, have reverted back to using buckets in 2005 then, eh?

[McGuire]

Originally posted by Engineguy
At least you tempered this with "It is said" and "the main thing." My peeve with this whole episode was that every race fan in the world was suddenly an expert on valvetrains and "knew" bucket followers are turds just 'cause the most obvious difference was that the Chevrolet had buckets.

No, the people who designed and built the thing were of the opinion that the bucket followers were the chief bottleneck in it. They started that vicious rumor.

As for "popular perception"... for many decades race fans believed that if it didn't have bucket followers it wasn't a proper twin-cam racing engine... nothing further from the truth, really. Look at all the classic engines from Mercedes with finger followers, for example -- the subject of this thread.

[phantom II]

If my memory serves me well, you were of that school also as I was.;)

Originally posted by McGuire



As for "popular perception"... for many decades race fans believed that if it didn't have bucket followers it wasn't a proper twin-cam racing engine... nothing further from the truth, really. Look at all the classic engines from Mercedes with finger followers, for example -- the subject of this thread.

[Fat Boy]

Originally posted by Engineguy
It also had different intake and exhaust ports, different valve stem contours, different water jackets, different pistons, different bearing diameters, different bearing widths, different clearances, a different ring package, different cylinder sleeves, different dry sump pump, different sump baffling, different intake runners, different exhaust headers, different engine management system, and so on and so on.

Ya, maybe, but beside this stuff, it was exactly the same as the Honda!

[NTSOS]

Quick question......is there in fact evidence of any current F1 motor that uses buckets and turns 20k reliably?

Juanito

[hydra]

Originally posted by Engineguy
GIGO. 40% more lift with fingers? Not bloody likely. Unless you're simply restricting yourself to an existing bucket system that has been given geometric constraints that never foresaw a need for higher lift. To which, again... GIGO.

Tell me Engineguy, when was the last time you saw a cam on bucket race engine with more than 0.450" of lift? Ever seen one that can handle 0.600"? I didn't think so...