Cam designing

Status
Not open for further replies.
Joined
Nov 26, 2009
Messages
3,054
Country flag
I've been working with John Andrews of Andrews products off and on for a couple years. He has some great cam design software that makes it possible to create and refine camshaft design. There are several input variables you can plug in to manipulate the profile. In the sample image below you can see:

cam lift - green
valve lift – light red
velocity – dark red
accelleration - black
jerk – dark blue


Cam designing

After a lot of careful study, comparisons and adjustments you get the lift data as below:

Cam designing


The JA software offers many features and you can enter valve train details and other requirements. The old Norton cam profiles work great but a little improvement towards perfection helps smooth things out - especially in the ramp jerk area which has improved with modern technology. Its been a long road but its finally paying off.
 
Last edited:
Many years ago Ed Iskenderian wrote an interesting book on cam design.
 
Maybe your talking about the book "Ed Iskenderian and the history of hot rodding"

With the JA cam software I was able to create about 50 cam profiles and one of them was a close match to the PW3. Then I could see the acceleration, jerk and other curves. From there I made a couple improvements - a more gentle and gradual ramp to reduce destructive valve bounce and a more ideal shape over the nose. Not a big difference but enough to prevent your valves from tangling at redline and a smoother transition at peak lift. This is the same cam in Mike Hardings 85HP racer (see thread "who says Nortons can't wheelie")


Cam designing
 
The Iskenderian book was not the one you mentioned about hot rods. It was a book about Iskenderian cams. One of the things we heard in the old days was that if the lift rate was very high, the closing rate had to be slower because if the valve was allowed to snap shut, it was more likely to break. Some BSA singles had cams like that. I think every modern cam I've seen has been symmetrical. But it is not so easy to check accurately.
 

That is very likely the book Alan is referring to. It was updated several times. I think it was originally published in 1958 with 71 pages. The latest edition I've seen had grown to over 100 pages. It is full of basic camshaft design, porting, and valve information. It had a great discussion (at the time) of how he developed his "Polydyne" lobe profile designs, but it's a little dated compared to modern books on valve train design. Still a fun read, though.

Ken
 
The Ashley Book of Knots was as far as I got. You guys are way above my pay grade. I'm jealous.
 
Seems to me that with cam design in a pushrod motor, we are faced with a conundrum. If we fit a cam which raises the usable rev-range to gain more power, we risk tangling the valves when the valve train cannot cope. But if we fit a cam which has slower lift and closing rates and less total lift to stop the valves from tangling , we probably don't get the full potential increase in power. There is probably a happy medium somewhere. To my mind, the timings I have seen for various Commando cams have been nowhere near what could really be considered to be 'full race' in other motors.

BTW, I think you are correct about that book. It appeared in about 1958. I thought it was more about selling cams than designing them, even though what was in it was probably radical at the time. When I was a kid, I used to read stuff like that. I even believed what was in 'Tuning for Speed'.
 
He has some great cam design software that makes it possible to create and refine camshaft design. There are several input variables you can plug in to manipulate the profile.

Did you buy the software or is this with the demo software.
 
Seems to me that with cam design in a pushrod motor, we are faced with a conundrum. If we fit a cam which raises the usable rev-range to gain more power, we risk tangling the valves when the valve train cannot cope. But if we fit a cam which has slower lift and closing rates and less total lift to stop the valves from tangling , we probably don't get the full potential increase in power. There is probably a happy medium somewhere. To my mind, the timings I have seen for various Commando cams have been nowhere near what could really be considered to be 'full race' in other motors.

BTW, I think you are correct about that book. It appeared in about 1958. I thought it was more about selling cams than designing them, even though what was in it was probably radical at the time. When I was a kid, I used to read stuff like that. I even believed what was in 'Tuning for Speed'.


Hey, Alan. I still believe most of what's in "Tuning for Speed." It's still a great read.

Ken
 
Here's an example of what the JA software can do for a PW3 cam - Smoothing out the closing ramp to avoid valve bounce while shortening up the opening ramp. You can see that the velocity (black) and jerk (blue) curves are higher on the opening side but that is usually acceptable with modern cam designs.

You just have to be sure the cam grinder knows what he's doing and doesn't mix up the opening and closing sides (tell him the cam turns counterclockwise as indicated on lower right side).

The input values (lower left area) for the ramp lift and ramp vel (velocity) are different for opening and closing making this an asymmetrical profile cam lobe. Note the shorter left side opening ramp (green) compared to the closing right side.

Cam designing
 
Hey, Alan. I still believe most of what's in "Tuning for Speed." It's still a great read.

Ken
We had a guy in road racing whose mouth was bigger than his brain. He was at Phillip Island and made a disparaging remark about Phil Irving without realising Phil's widow was standing right behind him. What amazes me about Tuning For Speed is it's accuracy about so many things. However it might have just been that era. I often wonder how Chromie and Rex McCandless got the Featherbed frame so right for the Manx Norton. They must have had a lot of painful experiences ? - You could not just design it from theory.
I actually met Phil Irving back in about 1967. He was extremely knowledgeable - or at least appeared to be so to me - I was a callow youth. A friend of mine worked with him at Repco-Brabham.
 
I was under the impression that the closing side of the cam lobe needed to be designed to catch the follower as it returned after being flicked off the nose of the lobe. It being the uncontrolled return which caused valves to break. If you do that, theoretically you can run the cam at higher revs. I think the only such cam I ever saw was in a Gold Star BSA which had heavier valves than most twins. But of course, with a Commando engine, you are getting closer to the situation you have with a big British single. Have you had a look at the cams which are used in 500cc Short Stroke Manx Nortons ?
 
We had a guy in road racing whose mouth was bigger than his brain. He was at Phillip Island and made a disparaging remark about Phil Irving without realising Phil's widow was standing right behind him. What amazes me about Tuning For Speed is it's accuracy about so many things. However it might have just been that era. I often wonder how Chromie and Rex McCandless got the Featherbed frame so right for the Manx Norton. They must have had a lot of painful experiences ? - You could not just design it from theory.
I actually met Phil Irving back in about 1967. He was extremely knowledgeable - or at least appeared to be so to me - I was a callow youth. A friend of mine worked with him at Repco-Brabham.

And Motor cycle engineering another great book of his
 
Look at the black (velocity) and blue (jerk) lines. The left side shows the harsh PW3 ramp - compare it to the smoother modern ramp on the right.

Cam designing

Compare the green line (lobe profile) and see how much shorter the left side ramp is. Many of the british cams had short ramps better suited for low RPM - and many run into valve bounce problems at high RPM when those short ramps are on the closing side. The image above shows ramps that can be located either on the opening or closing side, and many cams have the same ramp on both sides - but you really need the smoother ramp on the closing side.
 
Are you designing a cam which gives more midrange power, or more top end ? It is easy to get more power by moving the operating rev range upwards, - but with most Commando engines, the revs must be limited because of the bottom end. Perhaps there is more to be gained by adjusting the cam timings to suit the exhaust systems ? With four-stroke engines, the technology used with two-stroke engines is still relevant. A 2 into 1 exhaust system on a Commando can probably act similarly to an expansion chamber on a two-stroke.
 
JS

the graphics with your opening post are something of a puzzle. At the foot of the first graphic, the ramp jerk is given as 'constant' whereas it is far from constant. The ramp velocity is nearer to constant, which is to be expected.
This may require a further graphic from JA that magnifies the ramp to better illustrate the 'constant' jerk.

I must say the smooth curves are most impressive, but until the design is actually tested either (preferably) via a Spintron, or by actual installation and test in an engine, whether the dynamic results will be in accordance with the static design is somewhat doubtful.
With a pushrod engine my experience is they never are, and to be effective it is the dynamic variation that must be somehow incorporated into the static design.
How JA does this with his software I would be interested to learn
 
Did you buy the software or is this with the demo software.

I think Jim answered that in his OP. The software was in-house with Andrews Products. Inc. . I would be interested to know the price of the D4camW package in case you know, Jim. There is other similar software available for a download fee of 99 Euro. The package demonstrated has some nice features which the other doesn't have. However, I don't see a capability to perform dynamic simulation of the valve train among their offerings.

-Knut
 
JS

the graphics with your opening post are something of a puzzle. At the foot of the first graphic, the ramp jerk is given as 'constant' whereas it is far from constant. The ramp velocity is nearer to constant, which is to be expected.
This may require a further graphic from JA that magnifies the ramp to better illustrate the 'constant' jerk.

Unless Jim has replaced some pictures, I don't see jerk at (inlet) lobe open/close ramp as requested to be constant - the boxes aren't ticked. Yes, the lifter velocity @ rotation angles defining the ramp (degree -> "A" parameter) is constant for a while, but the important parameter is the acceleration curve (the black one). Jim made a typo in his last posting - the black line is the follower's acceleration at the lobe.

Jim: At www.tildentechnologies.com the following statement is made, quote:

We agree that the importance of jerk is grossly overstated. Valve train dynamics theory shows that acceleration is the highest derivative that needs to be continuous, i.e. jerk remains finite, but can have jump discontinuities. Cubic splines are completely adequate. Most designers are using profiles that are much smoother than necessary. For example, Norton's favorite spline uses quintic functions which have continuous snap (derivative of jerk). The use of overly smooth functions causes a slower opening cam or requires higher acceleration rates to compensate.

We have seen only one source (Hollingsworth and Hodges) that have actually recommended limiting values. They recommend that jerk not exceed 0.0002 to 0.0004 in/deg3 for pushrod engines, whereas direct attack valve trains (OHC, flathead) can use values of 0.0004 to 0.0008 in/deg3. These values are quite large. When used with cubic splines, the profiles are very similar to those with infinite jerk <...>. We have profiled many OHV "performance" cams and typically find maximum jerk of 0.00003 to 0.00004, which is 5 to 10 times smaller. Overly smooth cams are poor performers.

The best way to establish acceptable values of jerk is to perform a valve train dynamics analysis for the cam in question. (End of quote.)

The cited references are
Norton, R.L.: Cam Design and Manufacturing Handbook, Industrial Press, 2002
Hollingsworth, P. and Hodges, R.A.: "The History and Mathematical Development of Cam Profile Design in Rover", paper SAE 914172, (1991).

Jim, in the analysis of the PW3 cam, maximum jerk (JMAX) at opening ramp was found to be 3.67E-4 in/deg^-3, very close to the recommended maximum, wherease at closing ramp JMAX was found to be only 1.81E-4. However, these figures appear to be unimportant. According to the tildentechnology site, the objective should be to obtain continuous acceleration at opening and closing ramps, and forget about jerk at the outset. The acceleration curves above are not continuous, neither at the opening ramp nor at the closing ramp. It seems to me there is still rom for improvements. Would you like to comment?

-Knut
 
Status
Not open for further replies.
Back
Top