34 replies to this topic

[Canuck]

In my ignorance, I always believed overlap and duration were the real keys to the camshaft's design but a couple of recent articles in different magazines have emphasized the importance of the lobe separation angle. I've always been aware of the spec, but had no idea it played such an important role. Given that I like to think I build pretty decent engines, and am always looking for more knowledge, I'd like to know how to determine optimum LSA - neither of the articles I was reading gave any real insight into it (unless we're talking about Americna V-8s, which I don't build).

After using the Lotus demo engine software, I dropped them a line enquiring about a license but quickly changed my mind when the least expensive option was to rent it for a month for $1800 , never mind the $42,000 perpetual license. Anyway, getting back to the topic at hand, can anyone point me in the direction of more reading on LCA or perhaps educate me on it's finer points?

[McGuire]

First thing I would like to do is differentiate the terms LSA and LCA. Over the years they have become hopelessly confused.

[Bill Sherwood]

I use neither, as they don't give enough information.
I only use lobe centres, e.g. 100 - 108, etc.

[NTSOS]

With a single cam....I think that the LSA is fixed and ground in at manufacture and the LCA is adjustable plus or minus.

Must be nice to have two cams!

John

[m9a3r5i7o2n]

Quote from Sherwood;

I use neither, as they don't give enough information.
I only use lobe centres, e.g. 100 - 108, etc.

No camshaft catalog I’ve seen uses anything but Lobe centers that I know of and besides it’s the OVERLAP of the valves that causes the interaction of the sound waves, doesn’t it? A small amount of change in the overlap and the Lobe centers almost always accompany one another, doesn’t it?

Canuck , I am very curious! Just what part of the camshaft are you trying to draw? The complete lobe, the complete double lobes, i.e. the Ex. & In. side by side or one from end to end.

Wouldn't it be easier, (and a lot less expensive ) just to get a copy of the old, 1952 Thoren et al., "Cam design as Related to Valve Dynamics." S.A.E. Quart. Trans. 6. (1952).
Measurements of valve motion. Development of the “polydyne” system of cam design.)

Marion L. Anderson

[McGuire]

Originally posted by Bill Sherwood
I use neither, as they don't give enough information.
I only use lobe centres, e.g. 100 - 108, etc.

That's just it... this figure is often confused. Are you referring to the lobe center angle (aka LSA or lobe separation angle) or lobe centerline angle (LCA)?

Both these specs will typically be in the range of 100 to 118 degrees (which no doubt adds to the confusion). However, lobe center angle is expressed in camshaft degrees while lobe centerline angle is expressed in crankshaft degrees.

[McGuire]

As far as I am concerned, if the goal is to study and understand camshaft timing (as opposed to the specific specs of a given camshaft) I would set aside all these wacky and conflicting terms for now and focus on four events: IVO, IVC, EVO and EVC, laid out in crankshaft degrees using a theta diagram.

[McGuire]

Originally posted by m9a3r5i7o2n
Quote from McGuire;

I use neither, as they don't give enough information.
I only use lobe centres, e.g. 100 - 108, etc.

Not my quote, sir. I don't think much about lobe center angle (aka LSA)... unless effective/dynamic CR is an issue.

[m9a3r5i7o2n]

Sorry about that McGuire; the quote is from Sherwood.
M.L. Anderson

[Canuck]

Damn - even though I'd just read it, I still missed the mark didn't I? Yes, Lobe Centreline Angle, not Lobe Separation Angle.

From Stock Car Racing magazine

Myth 1: The most critical aspect of speccing a cam is the intake and exhaust duration.

In reality, the most influential aspects of a performance cam are the overlap period and the LCA or Lobe Centerline Angle (see graphic, "Camshaft Attributes"). For our novices, the overlap is the time that both intake and exhaust valves are open simultaneously around TDC, prior to the start of the intake stroke proper. For a race engine, overlap scavenging is a very important part of the induction process. On a Cup motor, the exhaust's scavenging action is more influential on the induction process than the piston going down the bore.

The amount of overlap an engine needs for optimum output over a given rpm range depends on the rpm range involved as well as the ratio of low lift flow of both the intake and exhaust valves in relation to the displacement of the cylinders involved. Correctly speccing a cam (as opposed to relying on a best guess) should begin with determining the amount of overlap required. From here, the correct LCA needs to be determined. This is not an adjustable feature, as is so often thought. Within a couple of degrees, only one LCA will deliver maximum torque and horsepower over the required rpm range. From the performance point of view, it is better to be 2 degrees too tight on the LCA than 1 degree too wide.

I would not have thought so much of LCA - I've never paid any attention to it.

Thanks Marion - I'll track down the book. It's not that I'm trying to draw it (yet) - I want to fully understand the influence of LCA on the engine (or at least the theory), and how I can apply it to the work I do.

Theta diagram? Example?

[McGuire]

A so-called theta diagram (the term may not be technically correct, strictly speaking) shows all four valve events through 720 degrees of crankshaft rotation, or one operating cycle:

[McGuire]

This little cartoon shows the interesting relationship between overlap and lobe separation angle.

[Canuck]

Right - so the question (still) is - how to determine desired LCA?

From a Vizard article in PHR

What dictates the cam's success in the quest for maximum area under the output curve along with highest peak torque and horsepower is not (as is so often assumed) the duration involved. The most important factor is actually the overlap and the Lobe Centerline Angle, often referred to as the LCA. I realize this may fly in the face of everything you have been told or have read before, but it's not that hard to see it must be so

[Joe Bosworth]

Canuck

I have always preferred to talk opening and closing angles for valve first opening but can manufactures have driven us all to use durations beginning and ending with generally, (but not always), 50 thou of an inch.

As a general rule of thumb, cams ground with 280 degrees of opening generally come in at about 230 degrees measured at 50 thou.

The strongest factor in performance, probably even more important than valve lift once you get lift to a reasonable dimension, is valve overlap at TDC.

A good starting point for a 280 degree shaft is

In O In C Ex O Ex C
30 70 70 30

or 60 degrees of overlap.

Such a cam will generally give max HP in the 6000 to 6500 range assuming decent ramp accelerations.

By comparison, a 15-45-45-15 cam will give max HP at about 5000 RPM and a 40-80-80-40 cam gets you moving up well into the 7500 RPM range.

I have generally found that you can shift things like torque point around a lot by rotating the exhaust opening point by 5 to 10 degrees; easy to do with a twin can engine but obviously needs a friendly can grinder to work with you when doing a single can engine.

The whole deal gets pretty arty-farty from here based on one’s experience. I have never come across a good source for going into this with a lot of science. Either you do the hard yards of dyno work yourself or you really get on first name basis with your grinder who if you are really lucky will let you know of his experiences of what works with what.. At one point in my life I was plugging a cam a week into just one engine trying to unlock the magic bullet.

I hope that this gets you down the path a bit!


Regards

[McGuire]

Originally posted by Canuck
Right - so the question (still) is - how to determine desired LCA?

You can't. That is, there is no "desired" LSA for a particular application. (Sticking to my terms here.) Sure, more overlap will tend to move the power band up the operating range; less overlap will move it down. Broadly speaking.

But beyond that, all I can say is I am not the best guy to make sense of Vizard's approach, because I don't agree with it. I don't think it does make much sense. For example the new LS7 cam runs 120 degrees of LSA, which according to Vizard is totally wrong. To the cam designer, the correct amount of LSA is that which resulted from selecting the correct EVC and IVO for the application in question. To me, tuning a cam by LSA is working from the wrong end of the horse.

[Canuck]

Hmmm...any idea who wrote the Stock Car mag article? If I understand things correctly, they're both saying (in essence) - decide optimum overlap then decide desired LCA. From those two figures you can calculate your duration.

Of course I'm still confused - according this picture,

#6 is the LCA (Lobe Centreline Angle - their words, not mine) - but isn't that also the LSA??? Okay McGuire - in your first post you said you wanted to

differentiate the terms

. Please do so I can slow the swirling in my brain a little.

Now - I don't want to sound ignorant about the contributions made here, but I've found in the past that typical 'car-grind' numbers do not work in our bikes. Certainly they make power but one of the hardest pulling cams on the market is a virtual square-lobe with fairly short duration (240 @ .053*) with 20-40/42-18 timing. The non-existant port length and piss-poor manifold design really like these cams. Old Bobby Wood turned the cam-grinders on their heads when he started putting those out. Some folks fear they're hard on the valve-train but when it comes to light-to-light (which is really all these bikes are suited for), his cams routinely leave the competition sucking dust.


Cranking compression is always an issue with a hotrod V-twin, more-so with a big-inch hotrod. The starter isn't the problem, the battery (lack of space) is.

Regardless, I'm after a better understanding not necessarily a better mousetrap.


* - I don't know why it's .053", just to be different I suppose.

[bobqzzi]

McGuire,

Care to comment on turbocharged cam design? Particularly the neccesity of running low or no overlap. Certainly this is the case with production engines given the design parameters. I'm wondering if you have any thoughts or experiences with high performance or racing engines.

I had one semi-credible source tell me turbo era F1 engines actually had negative overlap...which makes no sense to me.

[McGuire]

Originally posted by Canuck
Hmmm...any idea who wrote the Stock Car mag article? If I understand things correctly, they're both saying (in essence) - decide optimum overlap then decide desired LCA. From those two figures you can calculate your duration.

I am 99% certain Vizard wrote that too. Same thinking; identical flat prose style.

I don't agree with starting with an "optimum" overlap and then selecting a suitable lobe separation angle to go with it. That may be ok in a general way in cam shopping, to help sort through the infinite choices in the catalogs. But in cam design, by FAR the most crucial single point is IVC. Second is EVO. EVC and IVO are least important, and they determine overlap. Hmm.

Now, the thing about lobe separation angle is you can't change it without moving all four valve events. Why would you want to do that?

[McGuire]

Originally posted by Canuck
Of course I'm still confused - according this picture,

#6 is the LCA (Lobe Centreline Angle - their words, not mine) - but isn't that also the LSA??? Okay McGuire - in your first post you said you wanted to . Please do so I can slow the swirling in my brain a little.

I prefer to use the term lobe separation angle (aka LSA) rather than lobe center angle (aka LCA) because it is more obviously descriptive, and in order to prevent the constant and inevitable confusion with lobe centerline angle.

Lobe centerline angle is the distance in crankshaft degrees between TDC at the crankshaft and the midpoint of maximum lift at the camshaft. For example, when we say a camshaft has been ground with an intake lobe centerline angle (or "lobe center") of 110 degrees and has been installed "straight up," the camshaft has been indexed so that max intake valve lift occurs at 110 ATDC. If this cam is advanced 4 degrees (as with an offset key etc) the intake lobe centerline angle is now at 106 degrees.

Lobe separation angle or lobe center angle is the distance in camshaft degrees between the intake lobe centerline and the exhaust lobe centerline. Interestingly, this is the only angle you will find expressed in camshaft degrees. All other specs will invariably be expressed in degrees of crank rotation.

I think if folks study the charts posted here for a moment or two, it will become easy to distinguish between the terms.

You will note that Vizard and others use the terms lobe center angle, lobe center and lobe centerline angle more or less interchangeably. I am sure they can keep track of what they are talking about, but it must be very difficult for newcomers to follow along, which is totally unnecessary.

I know I am fighting city hall on this, but I don't care.

[Canuck]

I get the impression you don't care too much for Vizard

I'm reasonably certain it's Vizard as well - the little graphic above is almost identical in both articles.

His propensity to call the LSA "Lobe Centreline Angle", and not "Lobe Centre Angle" or Lobe Separation Angle has caused me some confusion the last couple of days. Asshole. So now we're back to the theory that timing is far more important than LSA and that LSA is a result of other calculations, rather than dictating other results. Okay.

So how then does one determine desired IVC and EVO - which will also determine overlap of course. Or - do you calculate overlap and then IVC and EVO around that? Chicken? Egg?

Sheesh...I think I need a traditional benefactor to fund all the R&D I'm going to have to do now.

Once again, at the risk of appearing a complete ass - anyone else care to weigh in on the theories here? If two people have two opinions on one subject, will three have 3?

[Joe Bosworth]

Guys

I posted my last message on this subject in an abject state of exhaustion having flown from St Louis, Missouri to Auckland New Zzzzland, close to a 24 hour trip backed up by staying up all day to try to get back in sleeping sync.

I therefore made a couple of mistakes, not being able to differentiate between a n and m in spelling cam and referring to a 15, 45, 45, 15 cam grind when I meant 15, 55, 55, 15.

Anyway let me try to redeem myself by sharing a spreadsheet I made this morning that seems to come pretty close to providing good cam selection data.

As I am travelling away from virtually all of my best data I haven’t tried to validate the spreadsheet for all of my data but it does come pretty close. I have entered a fair range of data that I have on my hand held and it seems

As I am not bright enough to know how to provide a working spreadsheet on a forum I will post a non-working copy and offer to send a working copy to anyone who might be interested enough to request a copy via email to doveratarachdotnetdotau

For anyone who takes a copy I would be happy to hear comments on the sheet.


To Calculate Camshaft Data (ENTER BOLD DATA)
(CALCULATE CIRCLED DATA)
Note: Uses valve opening degrees at lash!!! Degrees rounded to integer values.
Assumes a symetrical cam.
Formulas' for 2V engine, for 4V
double the entry for no. of cyl's.

Sept 2006 using Vizard plus JB Factors LCA =
Lobe Centerline Angle
Value for cu inch disp
/inlet valve dia inch
Inches cc/mm Inlet valve Dia
Engine size 350 5735 inches 2.08
No of cyl 8 mm 52.83


Cu inch/cyl 43.75 717 RPM for that valve dia
5872 @Max HP

per Vizard:
for For that RPM the duration should be
CID/in dia 21.03 268 degrees

Then LCA should be:

108.25

For a symetrical cam that LCA
& duration then overlap =
52 degrees


For that overlap & duration
the grind in degrees becomes
26 62 62 26
IOBTDC ICABDC EOBBDC ECATDC

[McGuire]

Originally posted by Canuck
I get the impression you don't care too much for Vizard

No, I think he's great and have recommended his books etc. many times. I just respectfully disagree with him on a few things.

[Canuck]

Originally posted by McGuire
identical flat prose style.

Wasn't thinking about his content so much when I wrote that

Regardless - Short of experience, is there a ROT to calculate at least baseline/ballpark valve timings for a given scenario?

[m9a3r5i7o2n]

Except for the lack of vertical height of the lift doesn't this show just about everthing that we need to know?

http://home.comcast....NG1940BUICK.JPG

Yours most respectfully. Marion L. Anderson

[NTSOS]

LS7 cam - 505 HP - 28 mpg

211 - intake duration @ .050

230 - exhaust duration @ .050

120 - LSA (lobe separation angle) camshaft degrees

120 - intake/exhaust lobe centerline straight up in crank degrees

123 - ILC (intake lobe centerline) factory installed @ 3 degrees ******

.591 - intake valve lift

.591 - exhaust valve lift

As opposed to a dual camshaft setup, a single cam's LSA will be fixed at birth, it cannot be altered. The ILC/ELC can be advanced or ******** with one affecting the other. Advancing the ILC, ******* the ELC.......retarding the ILC, advances the ELC, but the 120 LSA relationship will never change or will the overlap.

The factory installs the LS7 in the 3 degree ******** position:

123 - ILC - (Intake Lobe Center)
117 - ELC - (Exhaust Lobe Center)
_______________________________________________________________________

Quick overlap calc:

Add the in/ex duration
Divide by 4
Subtract the LSA
Multiply by 2

Example:

211 + 230 = 441

441 / by 4 = 110.25

110.25 - 120 = -9.75

-9.75 x 2 = -19.5

Overlap = -19.5 degrees of negative overlap
______________________________________________________________________

Wide SLA = low, zero or negative overlap for good gas milage and emmision control

Wide SLA = relatively early exhaust valve opening for restrictive exhaust systems

Wide SLA = relatively late intake valve closing for controlling dynamic pressure with 11:1 CR

Massive intake port flow capability allows relatively low 211 intake duration combined with .591 lift

Increased exhaust duration of 230, combined with early exhaust opening and .591 lift helps combat restrictive street exhaust system.
______________________________________________________________________

LS7 Cam profile @ ILC/ELC 120 straight up - 211/230 duration

IO......... -14.5 atdc

IC......... 45.5 abdc

EO......... 55 bbdc

EC......... -5 btdc

Overlap = -19.5 and 120 LSA
_____________________________________________________________________

LS7 Cam profile @ ILC/ELC 123/117 - 3 degree ****** - 211/230 duration

IO......... -17.5 atdc

IC......... 48.5 abdc

EO......... 52 bbdc

EC......... -2 btdc

Overlap = -19.5 and 120 LSA
____________________________________________________________________

Intake Timing calcs:

intake duration = 211

divide by 2 = 105.5

LSA = 120

subtract 120 from 105.5 = -14.5 IO

(if the answer is negative, subtract from 180)

(if the answer is positive, add to 180)

subtract -14.5 from 180 = 165.5

subtract 165.5 from 211 = 45.5 IC

Intake timing = IO -14.5 atdc --- IC 45.5 abdc
____________________________________________________________________

Exhaust Timing calcs:

exhaust duration = 230

divide by 2 = 115

LSA = 120

subtract 120 from 115 = -5 EC

(if the answer is negative, subtract from 180)

(if the answer is positive, add to 180)

subtract -5 from 180 = 175

subtract 175 from 230 = 55 EC

Exhaust timing = EO 55 bbdc --- EC -5 btdc
___________________________________________________________________

IO - -14.5 atdc

IC - 45.5 abdc

EO - 55 bbdc

EC - -5 btdc
___________________________________________________________________

To advance the cam, subtract the amount of advance in degrees from the intake and add to the exhaust.


To ****** the cam, add the amount of ****** in degrees to the intake and subtract from the exhaust.


Straight up 120 - ILC/120 - ELC

IO - -14.5 atdc

IC - 45.5 abdc

EO - 55 bbdc

EC - -5 btdc


Example: 3 degrees advance becomes 117 - ILC/223 - ELC


IO - -11.5 atdc

IC - 42.5 abdc

EO - 58 bbdc

EC - -8 btdc
__________________________________________________________________

Straight up 120 - ILC/120 - ELC

IO - -14.5 atdc

IC - 45.5 abdc

EO - 55 bbdc

EC - -5 btdc


Example: 3 degrees ****** becomes 223 - ILC/117 - ELC


IO - -17.5 atdc

IC - 48.5 abdc

EO - 52 bbdc

EC - -2 btdc
__________________________________________________________________

Note that a dual camshaft obviously does not have a fixed LSA until the ILC and ELC
are dialed in....a nifty tuning advantage over a single cam setup.

The LS7 cam would be great in a turbo street motor due to its negative overlap. A turbo motor cannot build low rpm boost quickly if the manifold pressure is being vented to atmosphere via a late closing exhaust valve.

Chevy engineers knew exactly what they were doing when they specified a cam profile consisting of a very wide LSA, combined with low to moderate duration and extremely high lift for a production pushrod street engine that produces 505 hp and gets almost 30 mph whilst cruising down the boulevard!

My friend Bob apparently has the same vision with his soon to be completed '57 Chevy with C4 suspension and a LS7/Viper/Dana 44 drive train.



John

[McGuire]

Originally posted by bobqzzi
McGuire,

Care to comment on turbocharged cam design? Particularly the neccesity of running low or no overlap. Certainly this is the case with production engines given the design parameters. I'm wondering if you have any thoughts or experiences with high performance or racing engines.

I had one semi-credible source tell me turbo era F1 engines actually had negative overlap...which makes no sense to me.

This is one of those topics where there is often less than meets the eye. For example, see the lift-theta chart for the Comp Cams Magnum 270 camshaft in post #11. Most folks would probably consider this a fairly rorty camshaft for road use. It has an advertised duration of 270 degrees (hence the model name) and at the advertised lift rate (.006") the overlap is nearly 60 degrees.

But if you look at the specs at .050" follower lift, which is certainly more toward realistic in terms of significant airflow, this profile has "only" 224 degrees of duration, and more to the point here, zero effective overlap. So when folks say a given turbocharged combination will or will not work with "high overlap," it's sort of a relative statement. Hard to know what they mean...maybe they don't know what they mean.

Truth is most reasonable turbo combinations will not cry out for a special camshaft, especially if the turbocharger is well-matched to the engine. Now when you get into your more exotic or unbalanced combinations, things can be done with valve timing to suit special needs. With a little engine driving a great big turbo (as in the F1 turbo era) you might bring EVO earlier to spool the turbine, sacrificing some engine efficiency for the sake of turbo efficiency. Or if you are trying to run a relatively high stock compression ratio and a big boost number at the same time (like some turbo kits retaining the stock pistons) you can delay IVC to lower the effective (aka dynamic) CR. All depends what you are trying to do.

[McGuire]

Originally posted by m9a3r5i7o2n



Except for the lack of vertical height of the lift doesn't this show just about everthing that we need to know?

http://home.comcast....NG1940BUICK.JPG

Yours most respectfully. Marion L. Anderson

Yes I think so, and it's simple and compact too. Essentially that could be the theta diagram in smashed down form. Or if you put it in the shape of a circle you would have an old-school Isky-style timing tag.

[McGuire]

For what it's worth, David Vizard has a new article in the December '06 issue of Stock Car Racing entitled "Breaking through the LCA Confusion Zone." (Interesting title.) The story amplifies and expands on his previous positions. One interesting item: in explaining his views on the importance of overlap, he revives the old Isky "five-cycle" concept.

[Learner]

I haven't seen any posts about lift at TDC overlap.

Can't 2 camshafts with the same duration,lift and lobe separation be vastly different in performance ?

I have always timed cams using lift at TDC.

[McGuire]

Originally posted by Learner
I haven't seen any posts about lift at TDC overlap.

Can't 2 camshafts with the same duration,lift and lobe separation be vastly different in performance ?

I have always timed cams using lift at TDC.

That method of cam installation will absolutely do the job, long as you are careful. Personally I prefer the ".050" duration method," which is also recommended by most cam manufacturers.

Great point -- you are absolutely right that two cams can have the lift, duration and LSA and produce different performance; in fact be different camshafts. I started to write down all the ways that can occur but it is a very long list.

[Learner]

To my way of thinking, talking race camshafts, the overlap is a by product.

1st requirement, max lift close to angle of max piston velocity.
2nd & 3rd, IVC and EVO. Depending on rpm range required.
4th, Acceleration rates as high as possible yet retaining structural integrity.

Cams that operate in the 7000/10000 rpm range seem to have half their lift at TDC and off seat timing around 58-60 IVO/EVC and 82-90 IVC/EVO.

Is my thinking way off track?

Road cars are quite different.

[Learner]

The reason I set cams at TDC lift is because piston/valve clearance is normaly in the .040/.060 range.

[McGuire]

I am probably not following you properly. Whatever cam timing method is used, if it is accurate the piston/valve clearance will not change. It is what it is. And I still have to check it on all cylinders, and I still want to check and record it over a sweep so I can know how much advance or ****** I can get away with further down the road.

[jluetjen]

I'm no David Vizard (disclaimer - I'm as salesman in an unrelated field), but I did some analysis of Porsche 911 engines and configurations and wrote this article on another BBS about the different cam shaft charactoristics and their impact on performance. Why the 911? Because it seemed somewhat unique in that there were multiple configurations of the same engine using multiple cam shafts, with factory supplied data no less. It's hardly up to the recent series in RaceTech authored by Gordon Blair et al, but hopefully it will be helpful in laymans terms for understanding the impact of lobe center angle and overlap. Henry Schmidt of Supertec was nice enough to post it on his web site. I also had a bunch of input from John Dougherty of Dougherty Racing Cams and some others on the Pelican Parts Engine Rebuilding Forum.

[McGuire]

Interesting read, good job.