Cams

Camshafts: larger engines do not need larger cams, assuming that we agree that larger means more lift and more duration. The physical size of the cam may be larger, as the larger displasement engines may be physically larger as well as larger in displacement. One example might be a normally aspirated small V-8 vs. a 500cid Pro type engine. The test was a SB chevy we built for the 24 hrs. at Le Mans with a displacement of 358 ci. Peak torque at 5000 was 532 with max hp of 668 at 7000 rpm., max engine speed was 7800rpm in the interest of engine life. The same year I built a 500cid Ford engine for unlimited off shore competition. The cam worked beautilully at Le Mans and was easy on valve components, and durability was a consideration, so when analyzing the air flow needs for the larger engine the same "grind" looked to be a good place to start. The 500cid engine produced 790 ftlbs and 860 hp at 5700rpm. When we buzzed it to 7000 hp was 992, and just for the hell of it I risked things some by doing a pull at 8500rpm and got 770ftlbs. = 1246hp.
The cam had .656" valve lift with durations of 278 int. and 280 ex. @.050" with a 114 lobe center. These numbers were verified both at the lifter and the valves (both were rollers). Now if you compare hp/cid you get 1.87 for the 358, and 1.98 for the 500, both at 7000rpm. Both engines had "unlimited" induction systems and exhaust systems optimized as well. So here we have a small engine vs. a large engine which have bore / stroke ratios that are identical at 1.14-1, and rod length to stroke ratios which were also equal at 1.75-1, so geometrically they were equal, and the cam (valve timing) identical, but the big engine didn't appear to me to need a cam with more lift or duration.
Both engines had the same CR's at "over" 16.2 - 1, which leads to valve timing vs. CR. The camshaft itself can dictate both the static CR (calculated as combustion space at TDC vs. combustion space + cylinder volume), as the more you lift the valves and longer they're open (lift and duration) often dictate the depth of the valve reliefs in the piston, and the greater the depth of the valve reliefs, the greater the volume of the combustion space (remember: the top of the piston is the bottom of the combustion chamber, and their combination at TDC is the combustion space). Now according to the way we calculate static CR, the greater the volume of the combustion space , the lower the CR, because, although the greater volume is added to the swept volume of the cylinder, it's almost insignificant compared to the OA. cylinder+ chamber space volume. So when dealing with calculated CR, the "larger" the cam the lower the CR will be, unless you put some "huge" dome on the piston, and that's a subject we're not getting into tonignt. There's another "kind" of compression as well which is "running" or "dynamic compression", and the cam has a tremendous effect on this, as the later you close the intake valve, the less the dynamic CR will be. A typical camshaft will close the intake valve(s) during the compression stroke for two reasons. First, you can't open and close valves instantaniously, the cam lobe must open the valve at a rate so that the spring won't allow the valve's inertia to continue opening the valve after the cam's lobe peak, which would cause separation, and sure destruction, and the cam must likewise return the valve to its' seat gently, or you'll pound the crap out of the valve face and the seat in the head with "valve bounce". So duration is somewhat effected by lift since you have to increase opening and closing time. Intake valves are also not closed until the engine's compression stroke because the intake mixture interring the cylinder has mass, and , therefore inertia. If the inertia is great enough many feel that you'll get a little additional cylinder filling, and thus more mixture to burn. I'll admit that at one specific rpm you will achiece this, but who drives at one specific rpm? A good case for the infinitely variable transmission, I suppose. But usually happens is that at lower rpm there's little or no inertia, and the piston rising on the compression stroke will simply push the mixture back up the open intake port, and as a portion of what you were going to compress is gone, your "running" or "dynamic" CR will be less. If you wish to demonstrate this, remove a spark plug and your valve cover. Set the intake valve lash to .020", then screw in a compression guage and rotate the engine with the starter four or five revolutions and record the highest compression observed. Now set the valve lash to .005" and repeat the test, cranking the same number of times, and record the high number. You'll find that it'll be less because you effectively lengthened the cam duration when you lashed the valves tighter, and conversely you shortened it when you increased the lash. In fact you can determine if your engine wants more or less cam by altering valve lash (within reason...+ or - .007") from recomended #'s. Use one of the acceleratometers to compare your car's performance, or a dyno. Let's move on for now.
As for rules of thumb for calculating cam profiles. The .050" valve or lifter value was originally adopted to provide a more meaningful way to measure what some feel is "pertinent" duration numbers because most people don't look at flow #'s below .050", and, as we've discussed cams must gently "ramp" the valve up to the lobe, and gently "ramp" it down to the cams base circle, or 0" lift. Cam measurements of duration taken at 0"or .001" lift are called advertised durarion, and since this number includes the opening and closing ramps, it's always a greater number of degrees than the duration measured at .050". You need to be careful when comparing cams from different manufacturers, as many will use a different check number than .050"...ie. .020", which would make the duration longer on paper even though the cam might be the same. If I were choosing a cam, I'd base my decision on both lift and duration purely on the application, and rpm range where you want the power increase. Application would include auto type, weight, type of trans., gear ratios of each gear, final drive ratio, tire diameter(and compound), flywheel weight, other engine mods., type of induction, type of ECU if injected, compression ratio, and what do you want compared to what you have in terms of performance. I can't forget this: how about emissions, and gas you're going to run.
Cams are certainly important in terms of dictating performance, but like almost everything we've discussed , a change in one area will affect many other areas. There's no such thing as a truely infinately variable cam...yet. So you really have to consider the fact that a cam change will usually give a performance boost in one area at the expense of power in another.....the old give and take. You have to deside where you want the increase. If it's everywhere, I'd recomend no more than a 10-15 degree increase over stock when measured at .050", or 25-30 increase over "advertised" duration, and as for lift 10 to 18% greater than stock, as long as you stay in the duration range. Lobe center-- around 112. You must understand that these numbers can and would change radically if I had some idea what engine we're dealing with. I'm going to move to part II for a while, and then return here to close.
Lifters: All hydralic lifters rely on oil pressure to pump up a piston in the lifter (domestics) or at the end of a rocker type lifter as in some ohc domestics and imports. The sole reason that hydralic lifters were invented was to: 1 eliminate valve related noise, and to reduce or eliminate the frequency of valve setting intervals. They accomplish this through the oil pumping up a small piston inside, or at the end of rocker types which eliminates the clearance or valve lash discussed earlier, so when running there's 0" clearance between the system from cam to valve, and no ticking noises. A roller hydralic cam simply has a roller at the point of contact with the cam which cuts friction some and also allows slightly more agressive "ramps" on the cam lobe. This lifter also is oil dependent to eliminate lash and noise. Flat tappets are lifters which have a flat appearing cam contact area , as opposed to a roller. There are two types, mechanical and hydralic as discussed. Mechanical lifters, and mechanical rollers are the same, do not have an oil pump as do hydralics, and therefore they typically have a clearance or lash which must be properly set, and they do make ticking noises, and in many domestic engines three's a small valve in them which meters oil to the rest of the valvetrain. As they don't pump up, as hydralics, the valve motion is more precise and they can operate at considerably higher rpm than hydralics. Rollers can once again follow more agressive cam profiles than flat lifters. Flat mechanical and hydralic lifters in domestic engines are not really flat. They are slightly convex in shape, and they're positioned so they only contact one side of the cam lobe which is also ground on a slight angle which causes the lifters to spin, and prolongs cam and lifter life. Rollers sontact the entire lobe, and do not spin....if they do, it gets expensive. All cam data should be measured with solid roller or flat lifters. You can not use any hydralic lifter to degree a cam or check specs. You also can NOT run solid flat lifters on a cam intended for hydralics, you can NOT run roller lifters on a cam designed for flat lifters. and you can NOT run a solid roller lifter on a cam designed for hydralic rollers and same goes the other way. If you disassemble an engine, you must put the same lifter on the same cam lobe it was originally on, and never try to run used lifters on a new cam.

Posted by andre (andre@cs.ucla.edu) on October 15, 1998 at 15:26:40:

Jeez, there's some great ideas in there...no flow at .150"...it makes so much sense...why have overlap at all, all you gotta do is get the stuff flowing.... thanks t.o.o. and alloy (for bringing it back).

Andre

Posted by NITRO on October 15, 1998 at 19:29:51:

Andre,
I would simply like to make a statement for T.O.O., and that is: he'd appreciate your response to some of his old writings.
He'll continue to write technical "articles" to be posted here. As opposed to when the articles got "out of hand", I suspect that he will go back to what he'd started in the first place - - "The World According To T.O.O.". His articles are designed to provide as clear a picture as possible without causing the interested reader to need a "dictionary" to understand what he's said.
His time here is growing a bit more limited by the day, and there's a tremendous amount of data both from books and studies, but the "real world experience" really validates what he says and does. When we hit a wall in a program, a simple "educated guess" is all that we need to progress again, and "that guess" didn't come from books, but from all the years in the late 60's when his race cars were his flow benches and dynos. In fact in 1969, his understanding of the importance of dyno testing was "rocked" when FORD's specialists took one of his BOSS 429's and spent a month on the dyno and "found" another 100 hp. T.O.O. went to the next race and the car was a full second and 14 mph off his Tejas combination that FORD had laughed at. He rebuilt the former combination and placed 3rd. in points after 3 days of round-robin eliminations with 31 other FX cars. The next highest placed FORD was 17th. FORD became very friendly after that.
It's simply those times and exposure that he'll rely on when there's nowhere to go.
I hope that what he contributes in his posts will be well accepted and stimulate some to research his findings. He's said many, many times before that he doesn't want others who are interested to need to walk where he's already walked, he wants people to begin walking from where his steps end.
I can't say any more.

Posted by andre (andre@cs.ucla.edu) on October 15, 1998 at 19:19:15:

*arf* -- I owe T.O.O. that. I got some mean ideas about cylinder head design now...he's probably already done them, but...that last thing about no flow at low lift...wow.

If you think about it, it's pretty clear...all you really need to do is get the air flowing behind the valve BEFORE the valve opens...T.O.O. did it by designing for LOW FLOW at low lift. "If I could have designed in zero flow, I would have"...hmmm....I think I know how to do it...anyone else? If you do it, you don't need overlap anymore (I think).

cheers,
Andre

Posted by NITRO on October 15, 1998 at 20:05:59:
In Reply to: big bone ;) posted by andre on October 15, 1998 at 19:19:15:

The best combination is to open the inlet slightly BTDC on the exhaust stroke, then close it at the velocity the piston is traveling..allow it to remain seated, begin opening at piston speed ATDC and then at @12*, slam it open, as the piston will be out of the way. It's a very hard mechanical operation, but it keeps the inlet on it's seat at TDC so no valve reliefs are necessary on the intake side of the piston.
On engines prone to valve train flex, etc. we'll simply begin to open the intake very late on the exhaust stroke to make the necessary valve reliefs as shallow as possible, and that's where T.O.O.'s "magic" One-Way intake valve seats work so well, as they don't allow reverse flow of inert gasses into the intake ports. Nobody has ever seen a port that doesn't look just like new when the "magic" seats are used, but as they don't flow backwards well, they do not flow well at low lifts either. The primary reason for the open-close-open, or the open very slowly mode is to create motion in the inlet port that effectively "stacks up" a good portion of the mixture in the bowl area so when the intake cycle really begins ATDC there's more than the volume of the port and runner's volume due to the elasticity and compressability of the intake mixture.
Combine the extra amount of cylinder filling with no contaminants, a superior combustion chamber/piston configuration, and you are looking at a tremendous amount of efficiency.
Most people who buy T.O.O.'s work will simply tell you that the reason it runs so well and doesn't break, is simply because he studies every component's relationship and role in the "system", and he optimizes tiny aspects that most people take for granted. He studies the application, the package "size", and taylors every single piece to do a specific job in that one arena.
So there's no magic, but there's attention to detail beyond belief, and "The Best Engine Is The Best Series Of Compromises Possible"...and if they defy certain laws during their testing and use, they do and that's not anything we're ever "shooting for".