New non symmetrical cams finally

Status
Not open for further replies.
I do not claim to be ‘the authority’ on cam design. But I do have new successful cam designs on the market already.

Stan Keyes won the AHMRA bears classic road race championship this year in the US with a new JS2 cam design.

See Mike Harding’s immaculate UK road racer below with new JS2 cam design, the dyno chart, and a vid of him off the starting line.


New non symmetrical cams finally



Mike's 85HP dyno pull with new JS2 cam design.

New non symmetrical cams finally



Mike starting in 2nd place pulling a wheelie.



Others are also enjoying success. As they say - “when the flag drops the BS stops”

***************************************************************************
About John Andrews who created the cam design software I use - this guy developed the cam for the Harley XR 750 - one of the most successfull pushrod twin racing engines ever.
 
Last edited:
  • Like
Reactions: NKN
acotrel
Don,t race but I love the look and sound of twin peashooters, and would never change a Commando signature, just tune whatever I,ve got to suit.
 
JS
by your own admission, your JS2 is the same as the PW3 but with your modification to the ramps. This being the case, the two racers mentioned above would be enjoying no additional performance benefit than if they had been using PW3's, and in Mike Harding's instance, the max.power indicated slightly before 7000 rpm would in no way be causing valve bounce with a PW3 unless associated components of the valve train were so incompatible as to be detrimental in this respect.

Undoubtedly the reputation of the PW3 has suffered badly recently, mainly because of wear issues, and your cams have yet to be proved in this respect, but performance wise, your cams have nothing extra to offer over what the PW3 has allready been providing for years.

Before you set the Harley XR 750 cams on a pedestal, take a close look at some of the recent cams as developed for use in NASCAR, there you'll find modern valve lift designs for pushrod engines, a study of which will reveal just how old and dated the Harley XR 750 design concepts are.
 
Before you set the Harley XR 750 cams on a pedestal, take a close look at some of the recent cams as developed for use in NASCAR, there you'll find modern valve lift designs for pushrod engines, a study of which will reveal just how old and dated the Harley XR 750 design concepts are.

Hi Snotzo, while I support your criticism re. lack of documentation (especially, the pushrod / spring / valve combination needs to be fixed in a verification and subsequent advertisement for a specific gain), I also think your criticism is way too harsh and I am missing a factual contribution on your part rather trying to shoot down a very welcome development effort by Jim. Performance is not everything - the engine also has to stay together in order to pass the finishing line. In this respect, it seems Jim is on the right track with his design, and combined with the BSA lifters wear appears to be very low, at least in a racing application. So - if you have insight into solid lifter and camshaft design used in current NASCAR engines which may be ported to the Norton twin engine, I am sure we all would read your contribution with great interest.

Having said that, in my opinion Jim (JSM) and Jim Comstock is investing time on a dead horse. The future arrived in the 70's by introduction of the roller lifter, used in HD engines since 1984/86. In current Nascar technology, roller lifters win hands down in terms of efficiency, endurance and sometimes power. The logical succession of Jim Schmidt's work would be to design a camshaft suitable for a roller lifter and the associated hardware to provide structural support to such a camshaft. Solid tappets belong in the museum - period.

http://rehermorrison.com/tech-talk-43-the-trouble-with-flat-tappets/
https://www.onedirt.com/news/comp-introduces-new-value-solid-roller-lifters-for-race-engines/

-Knut
 
Snotzo said -

"JS
two years designing cams and you know so much ?

I've been involved in this work for nearly half my life and I learn something new most every day, and one thing I came to understand a long while back is that no one can 'learn' how to successfully design cams in two years !

The emphasis here is on the word 'successfully'.

I am familiar with John Andrews software, having used examples some years back when he made versions available for test from his web site. The software may be very different now, but back then the versions I tested did not contain any provision whatever for adding the complexities of the pushrod valve train into the computation.

The spintron testing you claim has gone into these designs would be interesting to see, because I know of no Spintron equipment that is fully instrumented to enable meaningful results to be obtained for either a single or a twin cylinder engine, and the device you used whereby you motored an engine to make a dubious estimate of cam wear is not and never will be a Spintron in the form that was then used.

IMHO your efforts would be more to your credit had you first engaged in some extensive testing to confirm the soundness of your designs, and then follow by announcing them on your web site rather than use this forum to promote untried and untested products.

Snotzo, Saturday at 4:15 AM"​




"untried and untested?"

More like refinements on the long ago tried and tested.
And, given the market's size, it's not feasible, or fair, to demand your levels of testing standards. WE, are the testers. Not avoidable.

That said, I always read your postings, Snotzo, with respect and close attention.
 
knut
I doubt the feasibility of producing roller followers for a Commando simply because it would require much more than followers.
In any case the roller follower has it's limitations. To continue with the same valve events when switching to a roller follower the cam profile will require some change, and the smaller the roller the greater the change. By virtue of the foot length the standard follower has a certain advantage which if properly explored can result in a very subtle variation of both valve closing action and area under the lift curve. This variation may be done by regrinding the flat foot into a radius, and this radius may be either a continuous radius centred on the camshaft, or it may be offset from the cam centre, or indeed the radius itself can be continually changing across the foot, a varying radius in effect. With any existing camshaft, unless the profile has a negative radius component and so cannot be married to a flat follower, the maximum area under the lift curve will always occur when a flat foot follower is employed.
Anyone purchasing one of the JS2 camshafts and chooses to use it in conjunction with a flat follower will obtain an increase to the area under the lift curve. Would the result be good or bad ? It should be good, but I'll leave that to someone else to confirm or otherwise.

With the NASCAR engines, flat foot followers severely limited the size of the cam lobe, and the effect of the lobe encroaching or overlapping the edge of the foot brought serious issues with wear and breakages, and because the diameter of the follower was mandated and thus could not be increased in diameter, the roller follower was the means of overcoming this initial problem.
My comment re the XR 750 cam was not directed at the use of a roller follower, but at the basic valve lift design typical of that era.
 
Further comment on roller followers.

To run an existing cam with a roller follower which has replaced a flat original, will result in a loss of area under the valve lift curve.
I am not referring here to the JS2, which I assume has this covered via the John Andrews re design, but am here referring to a change made that involves the PW3, a design I am very familiar with.

Just by changing from a flat follower to a roller of 19.05mm diameter, the loss of area under the lift curve is just greater than 29%.
To make allowance for this size of roller and redesign to provide the same valve lift design, ends with a negative component of the flanks, and in the instance where a 19.05mm diameter roller is used, the negative radius is just over 56 mm, so a grind wheel necessary to grind this profile would need to have a radius smaller than 56 mm.
To change from a flat to a radius follower while still retaining the original cam will allways result in a loss of area under the valve lift curve, but in some instances might just result in a slightly more compliant drive.

The other point worth mentioning here is the rocker ratio, generally spoken about as being 1.13:1. In fact this is something that is not strictly correct. For instance, the intake rocker has two extremes, a minimum ratio of 1.111:1, and a maximum ratio of 1.131:1. The exhaust rocker is similar. The greater the valve lift the greater the ratio swing.
This variation of ratio, although small,needs to be accurately taken into account when a new design is undertaken
 
I talked to Stan Keys about his new JS2 cam and he sent me this photo of him cleaning up in Bears division. The top of the RPM band is where you might see more power simply because of its higher rev capabilities.

New non symmetrical cams finally


Nortons on the street don’t need to rev so high but they get the benefit of reduced wear/stress on the cam and valve gear. This also allows for lower spring pressure as Fast Eddie suggested. For example a JS0 street performance cam along with beehive springs will rev to 8300 with only 90lbs spring pressure on the seat. That means you can run 85lbs spring pressure or less for reduced wear because you don’t need to rev beyond 7000 or so on a street bike.

mdt-son - The latest version of the John Andrews software I use does accommodate “Stress Analysis and Valve Float RPM for known valve gear weights”. But I do not rely on them or advertise them because these types of calculations do not account for cam length and cam flex which is a big problem in Nortons because the flexing cam will push back on the pushrod causing valve bounce. Cam flex is also the reason heavy roller lifters will not work in a Norton because adding weight and higher acceleration loads will only make things worse.

Ultrashort stroke 500s can turn up to 9000 RPM with the new cams and racing beehive springs, but its really all about increased reliability and reduced wear.
 
Last edited:
acotrel
Don,t race but I love the look and sound of twin peashooters, and would never change a Commando signature, just tune whatever I,ve got to suit.

I love the sound of separate pipes with megaphones, but that does not mean I enjoy racing with them. What happens with a torquey twin cylinder bike when it is halfway around a high speed corner and drops off the megas ? Do you slip the clutch or do you simply wait, shut-off a bit and hope it does not get nasty when it comes back on song ? Racing with a 2 into 1 pipe is a whole different world. Megas probably give more top end, but on many circuits they are unusable. The thing about 2 into 1 pipes, is that normal cam timings are usually not the optimum for best go, and the gearing of the bike is different. It is not simply a matter of bolt on and go racing.
 
One thing I became aware of when I first began racing in Australia, is that many overseas race circuits are much larger - which makes them in effect power circuits, rather than riders' circuits. So when I read tuning tips in foreign magazines, they often did not work so well in practice on our circuits. What is good on a power circuit, is not necessarily as good on a riders' circuit. The question which still remains in my mind about Commando cams, is about optimisation of valve timings when they are used with 2 into 1 exhaust systems. Does a JS3 cam do the same things to the power band, when used with a 2 into 1 pipe as it does when used with separate pipes ? From various photos, the 2 into 1 pipes made elsewhere are of much bigger diameter tube than my own and the relative tail-pipe size seems insufficient.
 
I'm still down at 59lbs fitted and 199lbs on the lobe and having covered over 3,000 miles all is still good. The lift on the lobe is way over 0.400'' and it is run with radiused followers. The fitted pressure is not what I wanted or expected, but the valve spring seat pocket on the FA head is deeper than standard. Seeing as my engine is apart these days on a regular basis, I thought someone has to start somewhere and so I left at 59lbs. It will stay like this until dyno'd.
 
Madnorton, so what was the fitted length of the beehive spring in the FA head?
Regards Mike
 
mdt-son - The latest version of the John Andrews software I use does accommodate “Stress Analysis and Valve Float RPM for known valve gear weights”. But I do not rely on them or advertise them because these types of calculations do not account for cam length and cam flex which is a big problem in Nortons because the flexing cam will push back on the pushrod causing valve bounce. Cam flex is also the reason heavy roller lifters will not work in a Norton because adding weight and higher acceleration loads will only make things worse.

Jim - I foresaw that argument and counteracted by suggesting an additional camshaft support. Even a valve train consisting of solid tappets would benefit. As for "heavy roller lifters" - the weight game hasn't started yet. There is a lot of potential there, but I agree, they will never be as light as BSA lifters.

-Knut
 
Rollers can obviously be made to work in a Norton.
Dave Nourish fitted them into his motors, which have similar design aspects and space constraints as Triumph twins, ie quite tight.
However, part of the kit he made included stiffer valve springs to cope with the increased valve train weight.
It also, of course, required dedicated cam designs.
It’s all doable though...
 
Just got this photo of Roger M's 500cc shortstroke in Canada with a new design JS2 cam.

“Ran great all season and won every race in the 500 period 1 class... It seems happy to rev to 9000 rpm.” Roger M

New non symmetrical cams finally


The rev limiter was set to 9300 RPM and he missed a shift or two and skyrocketed beyond that (with beehive springs). It also ran on the dyno to 9300 RPM where it reached the limit of its power curve. Note that the JS2 cam started out as a PW3 cam profile but with several added refinements and reshaping that make it capable of extreme RPM without valve bounce problems. Of course you don't have to rev it as high on a 750/850 so at lower revs its going to be easy on the valve train.

Rogers dyno chart below (at end of racing season - used):
New non symmetrical cams finally


The red line was before they adjusted the timing.

Fast Eddie - yes Westlake type rollers could be made to work on a Norton with a center cam support - anything is possible. But I want to stay practical and a faster descending lifter won't help the valve bounce problem at all.
 
Last edited:
A couple days ago I got a message from Rainer Z in Germany who says he is getting 80 HP with one of the new JS3 cams.
 
The same thing happens with valve timings in a four stroke, and the exhaust system type is just as important. ...If a race cam's timings are adjusted to suit a two into one exhaust system , the power delivery is often stronger and smoother. So If I was designing a cam, I would try to develop it in conjunction with a two into one exhaust system rather than with separate pipes.

On a 4T its mostly intake closing and to some extent overlap events.

Would it not be better to adjust the 2-1 merger according to requirements of the camshaft and gas dynamics of the ports??

Kind regards

Christian
 
Status
Not open for further replies.
Back
Top