25 replies to this topic

[Powersteer]

What are the dimensions of the current crop of a formula one engines gear driven cam drive? What material is used for the gears? From the outside they look very thin. Can this approach be used on a high performance production engine?

[J. Edlund]

I can assume that they use some sort of high strength steel, hardened for wear resistance.

I gear train could be used for a production engine. The question is cost and noise caused by the gears. Some truck engines does for example use gears for the valvetrain drive, but the loads on these are higher since the injector/pump combos are driven by the cams, noise isn't that critical either.

[desmo]

I seem to have lost some good pics I had of a Peugeot F1 timing gear tower. From memory, they are very thin, perhaps 10mm. Material will be something similar to nitrided 300M steel.

[Powersteer]

This may sound like going backwards but if shaft driven were to be re-introduced in todays technology, I wonder if it can be beneficial to drive the cams. I heard it is most expensive to engineer. I like belt drive because its light, efficient and quiet but it flexes and disrupts accuracy but by how far? Can't they realign the belt to cam timing using variable cam technology?

[sblick]

For production it is very difficult. The noise is just too much. Most diesel trucks "hide" this noise by putting the gear train on the back of the engine. Same with some injector gear pumps. Most race gears are straight cut and are loud. It is one reason that you have helical gears in your transmission. Helical gears can't handle load as well as straight cut gears when comparing size.
I have seen IRL gears and they are quite thin similar to what Desmo said. I would say some had a diameter of 5-6 inches and were still 10-15 mm wide. They were all made of steel. I think because of rules. The engine I had worked on had gone from a chain drive to gear drive because the timing was a little inaccurate and causing reliability problems. I think that if a chain can't hack it at 10,500 rpms a belt would be worse. I would think the tension on the belt would have to be quite large.
I have seen transmission gears that were made out of some unobtainable steel for a sequential box handling around 1000 hp for off road racing, and they were amazing to behold. Quite thin. even thinner than our Formula Ford gears. The finish and the workmanship were astounding. I thought they were for a "show" trans, but after seeing a drawer full of them I knew they were used.

[Bill Sherwood]

Helical gears can't handle load as well as straight cut gears when comparing size.

I thought it was the other way around?
Since a helical gear always has about 1.5 teeth engaged at a time it can take more load than a straight-cut.
The downside is lateral thrust and extra friction.

[shaun979]

Originally posted by Bill Sherwood

I thought it was the other way around?
Since a helical gear always has about 1.5 teeth engaged at a time it can take more load than a straight-cut.
The downside is lateral thrust and extra friction.

That is exactly what I believe too.

[Wuzak]

Originally posted by Powersteer
This may sound like going backwards but if shaft driven were to be re-introduced in todays technology, I wonder if it can be beneficial to drive the cams. I heard it is most expensive to engineer. I like belt drive because its light, efficient and quiet but it flexes and disrupts accuracy but by how far? Can't they realign the belt to cam timing using variable cam technology?

The problem with belt drives (and chain drives) is that the belts can and do stretch. I imagine the situation would be very bad, particularly now that the engines have to last 2 race weekends.

Also, correct me if I'm wrong, belt and chain drives are less efficient (have greater losses) than a gear train.

[desmo]

Properly lubricated, roller chains are amazingly efficient. Think about racing bicycle drives. I'd be surprised if any gear tower could match roller chains for mechanical efficiency. Isnt' the reason gear drives are prevelant in pure racing engines more to do with accurate valve control and reliability?

[Greg Locock]

At low speeds chains are very efficient, as you speed them up they get less so. Our solar car team did measure it, that's why we dropped chain drive. One pair of straight gears is probably better than 99% efficient, but you'll need at least 4 (at a rough guess), so figure on say 96-98% efficient.

Designing gear pairs so that they are efficient for a wide range of torques is difficult - as the laod increases the gearbox housing deflects, forcing the gear out of alignment.

Incidentally there is afundamental reason why the back of the engine is better than the front for driving the camshaft - on most engines the flywheel is at the back, so it tends to have lower levels of torsional vibration.

[Powersteer]

And what about a titanium shaft driving the cams? It can be extremely compact and only have two point of contact with gears end to end. Can this tech be revived as the two cams would be geared to each other it cam seem quite effecient.

[J. Edlund]

Originally posted by Bill Sherwood
I thought it was the other way around?
Since a helical gear always has about 1.5 teeth engaged at a time it can take more load than a straight-cut.
The downside is lateral thrust and extra friction.

Yes, it's something like that!

Originally posted by Powersteer
And what about a titanium shaft driving the cams? It can be extremely compact and only have two point of contact with gears end to end. Can this tech be revived as the two cams would be geared to each other it cam seem quite effecient.

It's not only four camshafts at two banks that are going to be driven, but also the oil scavenge pump, coolant pump, generator, hydraulic pump and so on. Using shafts will probably end up more complicated than gears. There's also the torsional vibration issue, what if the shaft starts to flex?

[Engineguy]

This topic is sorta meandering around and back and forth from production engines to F1 engines, two pretty well seperated disciplines.

I'll remind everyone that gear driven cams have been used on gazillions of production cars. From the Ford Model T and A, the Ford flathead V8s, GMC and Chevrolet inline sixes, Studebaker V8s, Volkswagon Beetles, buses and Type IVs, Corvairs, British and German Ford V4s and V6s, Pontiac Iron Duke I4s, etc. They're all helical cut (or molded) gears, so noise isn't an issue. The majority of these gearsets have had steel crank gears and either fiber or aluminum cam gears, sometimes both depending on whether the engine was destined for a car or truck. GM used some nylon coated steel cam gears which didn't work out well, and some nylon cam gears that worked OK except for exceptionally high mileage engines. Its been common practice to replace fiber or nylon cam gears with aluminum when rebuilding an engine, just to be safe, particularly if it's a performance build.

A few observations:

Aluminum weighs 1/3 of what steel does. Plastics weigh half as much as aluminum.

The engineering plastics of today are in a different universe from the nylons of the 1960s and certainly the phenolic/fiber composites of the 1930s. For that matter, the same applies to the modern arsenol of aluminum alloys, MMCs, treatments, and coatings.

When I look at the large diameter cam gear on the Ford V4s and V6s (necessitated because the cam sits high in the block with the 60° Vee), I can't help thinking that it would cover the distance, as an idler gear, from a crank to an OHC cylinder head. No heavy steel chain/sprocket set with complex guides and oil pressurized tensioners. No clumsy, bulky, externally mounted timing belt drive. A big aluminum idler would also have the same thermal expansion between the crank and cams as the aluminum cylinder block it's mounted on.

And my final thought:

The latest F1 V10s (sans intentional sump ballast) are well under 200 lbs. Y'all think they have a 40 lb STEEL geartrain on 'em? Anyone got a magnet?

[Wuzak]

Originally posted by Powersteer
And what about a titanium shaft driving the cams? It can be extremely compact and only have two point of contact with gears end to end. Can this tech be revived as the two cams would be geared to each other it cam seem quite effecient.

Are you suggesting a right angle drive at the crank, a (very short) shaft and another right angle drive at the cam? Something along the lines of the bevel drive Ducatis?

If so, I would suggest that the efficiency would be worse than a gear train composed of spur gears. And probably less compact, at least in terms of engine length.

[desmo]

Originally posted by Engineguy
And my final thought:

The latest F1 V10s (sans intentional sump ballast) are well under 200 lbs. Y'all think they have a 40 lb STEEL geartrain on 'em? Anyone got a magnet?

The F1 timing gears I've seen in photos and firsthand at the Ferrari museum sure appear to be steel. There are some recent F1 engines on display in Maranello that one could do the magnet test on- anyone in the neighborhood? I'd bet they aren't Al alloy, Ti is notoriously poor tribologically- what else could they be that looks like steel? They're thin esp the part inside the toothed outer diameter- I'd guesstimate the whole geartrain for both banks wouldn't weigh more than 5kg or so.

[Powersteer]

Sure Wuzak but, why or how so?

[J. Edlund]

Originally posted by desmo


The F1 timing gears I've seen in photos and firsthand at the Ferrari museum sure appear to be steel. There are some recent F1 engines on display in Maranello that one could do the magnet test on- anyone in the neighborhood? I'd bet they aren't Al alloy, Ti is notoriously poor tribologically- what else could they be that looks like steel? They're thin esp the part inside the toothed outer diameter- I'd guesstimate the whole geartrain for both banks wouldn't weigh more than 5kg or so.

There was some guy that made replicas of the "Honda Six" motorcycle. The original had a gearbox with coated titanium gears, but the replicas got steel gears instead which were supposed to be better. The replica engines was made by french JPX, which also made engine parts for F1 (if I remember correctly they manufacture the whole Renault engine, with Renault only doing the engineering).

[ciaoduc1]

Don't gears, and to a lesser extent chain drives, have a really bad habit of transmitting crank harmonics to the valve train? Wasn't that the whole reason for making belt drives?

[Greg Locock]

I don't know, but you could always put an isolator into the gear itself, rather like a TV damper.

To be honest i'd have thought cost and weight were bigger considerations.

[Engineguy]

A little off topic, but probably as good as anywhere:

In the early 1980s I was at the SAE Congress & Exposition chatting with a Gates engineer (I think... could have been Dayco or Goodyear... I've had a few beers since then) about the pros and cons of timing chains versus timing belts. I made an off-hand comment, not thinking it was remotely possible... "What I'd really like is a timing belt that I could package internally in the hot oil bath just like a chain". To my surpise, he replied, "Yeah, you and everybody else." He went on to say several car companies were pushing them and pressuring them to develop just that, but they weren't making much progress.

Fast forward 20-25 years:

Just noticed recently that Comp Cams, in addition to their (dry) timing belt conversion kits, has a small block Chevy internal WET belt drive timing set!

Their blurb:

[llmaurice]

Back to the gear driven thread ,the Japanese (Toyota )have for years employed gear drive cams on some of their engines . They have "scissor" gears whereby a thin section of secondary gear is attatched to the main gear and spring loaded in order to obviate backlash .
Lately , the Famous Honda SP1 and 2 superbikes which gave Edwards supremacy over the rattly old Ducatis in World Superbike carried this technology over from the the 1986 VFR model. In race applications however ,the "quietening " gear is removed to obviate the albeit minimal power loss .

[ciaoduc1]

When I saw your screen capture with the little mouse arrow that wouldn't move I thought I was having one of those too often talked about but not enough seen flashbacks.

Why would you want to oil bath a cam drive? Why can't you put a cam belt in oil?

What was the outcome of your discussion about the pros/cons of different drives?

[knighty]

cam drive gears on race engines are very thin.....the last ones I last saw on the nissan infiniti IRL race engines in 2002 were about 5-6mm thick, they were machined from solid steel and were very highly polished for extra fatigue resistance, sorry I dont know the steel grade. I think Ricardos machined them, therefore f&^king expensive!!!!

there is a 2:1 reduction ratio using the gears on the crank nose, this is then splits off the power tothe scavenge pumps on one side, and the water and alternator on the other side of the engine, chains were used to drive the cam shafts.......quite a conventional layout for a race engine.

Belts are now frowned upon in the race and road car industry, as they stretch when hot, changing timing which effects yor race engine power........or your road cars tightly controlled emissions........the life of chains is now very high........about double the life of a belt

[JwS]

One of the downsides of chains is the dreaded 'chain surge'. You need to carefully design the guide and tensioning/damping system of the chain to prevent waves moving in the system. If you look at a high performance DOHC engine with chain drive the entire chain is usually closely guided, and it seems like there are always issues with the tensioner/damper. On motorcycles it is very common to remove the stock spring/hydraulic tensioner and replace it with a simple threaded tensioner that you can manually adjust. The stock pnes will not keep the chain tight reliably under racing conditions.
I have noticed a trend in the same direction for belts, earlier belts seemed to employ a fixed idler bearing that was adjusted once at installation and checked occasionally, no real tension to speak of. Now many of them seem to have spring tension and hydraulic damping like a chain, probably the same issues of surge that chains have (though the mass is much less)
I have always understood that was the major problem with chains at high rpm, and the reason that gears were prefered, they can have issues with lash interacting with torsional vibration of the crank and cam, but I guess it is easier to live with.
JwS

[sblick]

To correct a mistake I made earlier. In my Mechanical Engineer's Reference Book 12th ed. It says that helical gears are better than straight cut gears for shock loading. It says that, "the teeth mesh gradually so that at no time is the full width of the tooth engaged". Also mentioning that they are quieter. The "drawback of the helical gear is side thrust arising from the helix angle".

Sorry

[Powersteer]

Well, helical gears have a small portion of contact which can result in weaker contact strength where as straight cut has full gear contact making helical a lot quieter. So when high friction happens because of torsion the helical will wear first and then break its gears. But in a sudden power surge helical is stronger in terms of outright strength.

Edit: I'm still on shaft driven.