I'm interested in learning more about camshaft design and was wondering if anyone could recommend some online references. Specifically, I'm interested in learning more about the ramp area and the shape of the lobe (more than just about the lift and duration).
Thanks!
Hmmmm... I used to have a link but its dead. Lemme search around a bit and see if I can find it.
I guess I should ask what you specifically want to know. I'm pretty good with cams if you have some specific questions.
Ok, well I can list some of my initial questions here. Some of these might be pretty n00bish questions, which was why I was looking for some more material to read up on, but anyway...
At the moment, I'm interested specifically in how this applies to the 4v Ford Modular engines (DOHC, hydraulic lash adjusters, roller followers).
1) Ramp velocity/acceleration: So, from a theoretical stand-point you would want your valves to open and close as fast as possible, right? But as you move to a lobe profile with faster ramp velocities you put more load on the valvetrain components and need stiffer springs to return the valves without float (I think). What are the practical limits to this? On the closing ramp I guess it's the spring rate. What about on the opening ramp? It seems like you can make the ramp velocity there quite high as long as the loads don't become big enough to damage any components. Is this correct? I guess this is why asymmetric cam lobes exist?
2) Changing ramp: What is the best way to change the ramp velocity? would you move to a smaller base circle, or make the nose of the lobe a little broader?
3) Changes in lobe base circle: To get a faster ramp rate, it seems like you may need to move to a smaller base circle on the came lobe. If you were to do this, can the hydraulic lash adjusters compensate?
Ok, well that's probably enough to start I'm sure I'll think of lots more to ask later.
I guess fundamentally, what I'm trying to understand is, if you could make any lobe profile (no manufacturing constraints), what would it look like and why? Then, what modifications if any would be required to the rest of the valvetrain to use that cam in an otherwise stock engine.
If you want more information on application, I guess I would say aggressive street/mild race. Maybe 7,200-7,500 rpm rev limiter, headers and intake both with runners sized appropriately for higher rpm operation.
I have been mulling over some ideas on how to make custom cam lobes and now I'm wondering if I could do it, what would I change. I think the lift, duration, and lobe separation are pretty well dictated by the engine geometry and operating conditions, so that leaves the ramp as the biggest thing to figure out.
Well not knowing the particulars of the Ford engine, I can't answer the questions specific to the Ford... but ramp rates in general have a toll on valve springs, the valve seats and the valves themselves. The higher the ramp rates, the more abuse the valve spring, the valve head and the valve seat will need to absorb.
Steeper ramp profiles will allow the valve to be fully open for a longer duration. Steeper ramps will give you a snapier acceleration and more throttle response. More suited to race engines than around town easy to drive engines. You have to come up with what you want the engine to do and pick a cam that works in conjunction with the rest of the components you have. I try to stay away from the cams that will just give you a peak HP number and go with the ones that give you the most torque over the entire RPM range. I like to look at the dyno chart and pick the cam that has the greatest area under the torque curve.
In reply to JohnyHachi6:
The one thing I'll point out- there's no real manufacturing limitation on lobe profiles- big, small, convex, concave, whatever- it can me made.
Limitations to the cam profiles are physical. Material ability to open at a stress rate, valves running into heads, cam to pusher (roller, tappet, whatever) geometry, etc. Compromises are the ability to run at specific areas, valve train weight, spring options, etc.
It's not easy, at all.
In reply to mguar:
The goal is to have a time trial and auto-x car that I can drive around town once in a while, but not as a DD. I don't think it's going to be that competitive in auto-x and I already have a couple sets of cams that would work alright for low-mid range torque, so lets focus on the time trial goal. Likely tracks would be Road Atlanta, VIR, and Sebring, among smaller local tracks.
The car should weigh between 2300-2400 lbs with driver and fuel. Transmission is the T45, rear end is a 4.10:1 ratio.
I'm not really interested in a dyno queen, so maximizing peak hp is less important to me than maximizing power across the normal operating range during a time-trial type scenario. (As tr8todd mentioned).
Some years back I went to Smokey Yunicks seminar at the PRI show. He ran a SBC on his Smoketron with valve springs with an installed rate of 75# (much lower than the 200#+ normally used). At 7,000 rpms there was no float at all. His interpretation was that the higher rate springs didn't prevent float, but in his experience often caused the valves to bounce off the seat when closing.
High speed camera valve spring movies:
http://high_speed_video.colostate.edu/#Mechanisms
From what I learned, and this is pretty basic stuff, here's how I look at camshaft design...
NA applications -
Intake - You want the intake valve to open slightly late on the downstroke. This makes the piston create a vacuum in the cylinder and once the valve is open, the air rushes in and fills the chamber quickly. Close it quickly around BDC of the intake stroke and hold your intake charge.
Exhaust - Open it at BDC of the power stroke in order for the cylinder to come back up easily and evacuate spent gasses effectively. Close it right at TDC.
Boosted applications -
Intake - Open the intake quickly right at TDC from the exhaust/intake stroke. Leave it open as long as possible. A little overlap is OK. You want it open as long as possible so that blower/turbo can force feed that cylinder all it'll take!
Exhaust - Open it a little late on the exhaust stroke. Building a little pressure for the exhaust is OK in a turbo cam. This helps increase the exhaust velocity. Close it a few degrees after TDC from the exhaust to the intake stroke to help evacuate the cylinder. Close it quickly.
This is what I remember. I don't know much about lobe centers and all that stuff. Hopefully someone else will chime in.
Cams were created by a VooDoo priest and work on magic. They cannot be understood, only worshiped and revered.
Lobe centers help determine the overlap, or the time that both valves are open. To a certain extent, the less lobe center angle the more overlap, the more lobe center angle the less overlap. The cam lobe profiles are also part of this.
Overlap on a N/A motor can definitely control engine vacuum and thus the way the engine makes power. More overlap = poor bottom end performance but better top end, less overlap is the other way around. Boost can change this; if running a supercharger it sorta overcomes the big overlap and restores bottom end performance. Turbo engines; unless the system is carefully designed the turbo won't make boost at low RPM and this leads to poor low end performance, meaning crappy acceleration from the dreaded turbo lag. IIRC that's why so many of the WRC cars use those 'afterburner' deals to keep the turbo spinning fast in an off throttle situation (that's all the popping and banging you hear as they round a corner, etc).
There was a question about base circles earlier. The regrind hotrod cams which are available for MGs etc are made by reforming the base circle (making it smaller), this increases the lift. It's pretty difficult to add duration that way, unfortunately.
As the cam lobe comes around, if the base circle is not smoothly blended into the ramp the valve follower will experience a sudden change in speed. That can lead to problems at higher RPM.
Rocker ratio is also a big part. A OHC cam with the lobe acting directly on the follower and valve means the cam effectively has a 1:1 ratio, meaning the valve will do exactly what the cam lobe does, no more and no less. This can be a good thing and it cuts down on valve train mass which is why so many sportbike engines that wind to astronomical RPMS are built that way.
OTOH, if rockers are involved the length can be changed and this opens up a whole new world of possibilities, check out the number of different ratio rockers available for a SBC for instance. But it also adds its own problems, such as valve train mass.
So, if I were to make a new set of cams for this engine, it sounds like I'd be best just replicating the OEM base circle and ramp rate, and just make minimal changes to the nose to get the lift and duration I want. Sound about right? I'll leave overlap out of the picture for now since that can be adjusted later (DOHC).
Yeah, but there's one thing to be aware of: changing the duration means you change the ramp speed. Like this: the lobe needs to be 'wider' to add duration. This makes it wider in relation to the base circle. That changes the ramp speed.
If you make your base circle larger you can get rid of some of the ramp speed problem, but now you have another: the valve lifter/cam follower 'installed height'. On a mechanical follower cam, this is not a big deal. You can usually find thin enough valve adjustment shims to fix that. If it's a hydraulic cam, that's not so easy. The bleed holes in the lifter are designed to be within a certain area when the cam is on the base circle, engine oil pressure holds the lifter in contact with the base circle. Once the cam starts to open the valve, the oil bleed hole is covered so oil can't escape and since oil can't be compressed the valve opens. If you change the base circle, you have to adjust the lifter position to replicate this action. One way to do this is to remove the same thickness of material from the tip of the valve as is being added to the base circle, of course keeping in mind the position of the stock oil feed holes in the head. Not insurmountable, just something to be aware of.
To change total lift, you'd need to have a taller lobe. So that means either a new billet or a weld up/regrind. There's a company called Web-Cam that does exactly that, they'll build up and regrind billets for pretty much any engine. http://www.webcamshafts.com/
And yes on a DOHC with a set of degreed cam sprockets and a dial indicator you can set your overlap pretty much anywhere you want.
In reply to Curmudgeon:
Awesome - this helps answer a lot of the questions I had. Thanks!
JohnyHachi6 wrote: 1) Ramp velocity/acceleration: So, from a theoretical stand-point you would want your valves to open and close as fast as possible, right? But as you move to a lobe profile with faster ramp velocities you put more load on the valvetrain components and need stiffer springs to return the valves without float (I think). What are the practical limits to this? On the closing ramp I guess it's the spring rate. What about on the opening ramp? It seems like you can make the ramp velocity there quite high as long as the loads don't become big enough to damage any components. Is this correct? I guess this is why asymmetric cam lobes exist?
Like has been mentioned, yes opening and closing the valve as fast as possible is a good way to get more area "under the curve" or more total potential for airflow. Mechanically (in the real world) you are limited by of course valvetrain speeds much like you are limited by bore/stroke ratio and stroke/RPM ratio with pistons and rods. Moving things too fast makes things 'splode.
You are also limited by the approach angle that is determined by the ramp speed and the size of the roller in the follower. As you increase the ramp rate, you are decreasing the angle at which the ramp meets the roller. As you reach theoretical infinite lift speeds, you reach an approach angle of zero, meaning the lobe contacts the roller at 90 degrees (on the side) instead of retaining an approach angle that lets the roller be pushed away from the lobe.
On the theoretical side, approaching infinite open/close speeds would reach a point of diminishing return in the form of intake vacuum, intake reversion, and a few other things.
2) Changing ramp: What is the best way to change the ramp velocity? would you move to a smaller base circle, or make the nose of the lobe a little broader?
Cam designers have pretty much kept up with the max possible ramp rates for a given design. They have broadened noses, used smaller base circles, and the bottom line is that any more and they can't offer a warranty or maintain their reputation.
The best way for you to alter "ramp speed" is to use increased ratio rockers. That obviously doesn't alter the cam lobe, but it does alter the speed at which the valve opens. The cam lobe plot remains the same on a graph, but the valve opening and closing curve changes.
3) Changes in lobe base circle: To get a faster ramp rate, it seems like you may need to move to a smaller base circle on the came lobe. If you were to do this, can the hydraulic lash adjusters compensate?
In a word, most likely not. To get any appreciable change in the lobe profile you would have to remove more than would be compensated for with lash. You would need to utilize an adjustable valvetrain to regain your correct preload.
I guess fundamentally, what I'm trying to understand is, if you could make any lobe profile (no manufacturing constraints), what would it look like and why? Then, what modifications if any would be required to the rest of the valvetrain to use that cam in an otherwise stock engine.
I would make it big - not in duration or lift, but a physically large lobe and large roller followers. The bigger they are, the more ramp speed you can accomodate without excessive side loading on the follower.
But, then I have the problem of weight and space. Since my current build is a 383 LT1 (cam in the block) I have to be very careful to make sure my standard-base-circle cam doesn't contact the rods on the stroker crank. Many hardcore race blocks are cast with the cam location higher to prevent such contact.
You'll need to log in to post.
