Worn PW3 cam

ando said:

Thanks Jim. Welding was done by Franklin Cams NZ and the stellite was applied with a hydrogen water torch. Considering their experience I would doubt a bad weld, although that would give me peace of mind. http://www.camshafts.co.nz/AboutUs.aspx

Other cams that I have had done by them are all satisfactory.

The valve springs that I am using are aftermarket beehive springs for an LS1 GM engine and are shimmed to give a little more seat pressure. The cam was covered liberally with run in lubricant as supplied by Franklin. Maximum revs is around 7250 although the engine revs willingly well beyond that. Engine is a short stroke Ariel single and the cam is the factory competition cam with slightly longer ramps. The valve timing measured at .050” lift is 39 68 59 44 (or 39 68 70 30 with an offset foot exhaust follower) which is fairly hot. So valve train control is looking like the problem. I was about to fit a larger 1.85” inlet valve, so I may have to rethink that as well as the valve springs that I am using. Sorry about hijacking the thread a little.

ando

Click to expand...

I would look at the pressure at peak lift.
Look for a spring with less turns and a higher rate.
A lighter valve, spring and cap would be a good thing.
I suspect the real culprit is the cam grind is not suitable for the rpm you are turning but you might fix it with a higher rate spring.

Or go to a dual spring with a dampener. That is what was likely used in the past.

comnoz said:

I suspect the real culprit is the cam grind is not suitable for the rpm you are turning

Click to expand...

An old Ariel single at 7250rpm... you may have a point there Jim !

Ando- your post has really made me think. My ES2 has gone round the clock and the cam and lifter are perfect. The Lobes appear to be radical in terms of lift but the valve timing is not of course. Very much doubt they were made out of anything exotic , wonder if there is something about modern hardening processes that leaves surfaces which oil cannot cling to or is easily wiped off with consequent reduction in film strength..
Actually Ariel singles were tuned to very high outputs and revs in the UK . Very much doubt Ando is doing anything that wasnt done by Hartley back in the 60s.. Would wager his were run on 'R'. Wonder if they were moded to use Hairpin Valve springs ? They were said to be competative against a Manx

Staytite said:

Hi Fast Eddie

Exactly; and written warnings from Norman on the PW3 Data Sheet and a statement that he used Graphogen on my build sheet !

I just wondered if failures correlated with the use or not of Graphogen during assembly?

Click to expand...

Norman is a great advocate of Graphogen, and can tell you some interesting stories regarding the benefits. I have been an enthustiastic user of Graphogen for decades in car and motorcycle builds.

However, I had a PW3 fail after 26 races. I would tell you to use it, but it is a break in element, most of it is gone at your first oil change. It may provide a skin for some time, but that likely degrades after hard use.

On another topic mentioned is too many revs at start up, contrast this with the need to get splash lubrication as you warm through, so running at 2500 or a little above. oN my race bike I usually blip between 2 and 3K, others run a steady 2500.

oldmikew said:

Actually Ariel singles were tuned to very high outputs and revs in the UK . Very much doubt Ando is doing anything that wasnt done by Hartley back in the 60s.. Would wager his were run on 'R'.

Click to expand...

Is that JR Hartley the fly fishing author ...?!

Seriously, that’s a fair point. And a good point about running R, which I think we can almost guarantee any serious tuner would have used back then.

R is amazing stuff, if used within its limitations (don’t run it too cold, don’t run it too hot, change it every couple of meetings). I’m quite sure that R would have prevented such issues back then.

Fast Eddie said:

Is that JR Hartley the fly fishing author ...?!

Seriously, that’s a fair point. And a good point about running R, which I think we can almost guarantee any serious tuner would have used back then.

R is amazing stuff, if used within its limitations (don’t run it too cold, don’t run it too hot, change it every couple of meetings). I’m quite sure that R would have prevented such issues back then.

Click to expand...

Fast Eddie said:

Is that JR Hartley the fly fishing author ...?!

Seriously, that’s a fair point. And a good point about running R, which I think we can almost guarantee any serious tuner would have used back then.

R is amazing stuff, if used within its limitations (don’t run it too cold, don’t run it too hot, change it every couple of meetings). I’m quite sure that R would have prevented such issues back then.

Click to expand...

Surely L P Hartley The Go Between ie we need to know whats going on between Cam and followers.. The Ariel tuner L Hartley I do not think was related though it doesnt seem to have been a problem he faced..

Here's a comparison of copys of the PW3 cam and it shows how some brands have changed the ramp from the abrupt original. If the ramp is too abrupt then the exhaust valve will bounce off the seat and if revved high enough it will tangle with the intake valve. The famous HD XR750 profile has nothing to do with the PW3 but is shown just for comparison. The PW3 (see Megacycle copy) is extreme on the abrupt side and the HD XR750 ML cam (Mert Lawill) is extreme on the gentle side (designed by John Andrews). The HD XR 750 profile shown is corrected for Norton valve train/rocker arm ratio. Some of the profiles shown are for radius lifters. All ramps represent follower lift. None of the cam makers refer to these grinds as PW3 cams - some are copies of copies.

Jim, What are you referring to when you state: "The PW3 (see Megacycle copy)". Specifically what Megacycle cam is a copy of the PW3?

I may have asked this before on another thread but for the benefit of the readers here, exactly how did you achieve the following: "The HD XR 750 profile shown is corrected for Norton valve train/rocker arm ratio"

I am thinking by the nature of the XR design (different rocker ratio) the ramps had to be more gradual plus they were roller cam followers, were they not?

Dances with Shrapnel said:

Jim, What are you referring to when you state: "The PW3 (see Megacycle copy)". Specifically what Megacycle cam is a copy of the PW3?

I may have asked this before on another thread but for the benefit of the readers here, exactly how did you achieve the following: "The HD XR 750 profile shown is corrected for Norton valve train/rocker arm ratio"

I am thinking by the nature of the XR design (different rocker ratio) the ramps had to be more gradual plus they were roller cam followers, were they not?

Click to expand...

THE PW3
and other cams have been copied and changed for radius or flat lifters and etc dating back to Norris cams etc. Master plates have worn and grinders have moved on and no one really knows who made what anymore. Comparing data reveals what is a copy. What I found at Megacycle was a copy of the PW3 but for radiused lifters. This took some time and research and it was like "pulling teeth" to get it. Everybody copies - so PM me for details.


THE XR750 CAM
First I borrowed an XR750 cam and lifter and recorded the lift data with Harley roller lifters. Once you have that data you can use any lifter you want as long as you end up with the same lift per degree at the lifter - as long as the valve train can handle it. But then you have to change everything for the rocker arm ratio (see below) The XR750 got better ramps (the ML cam) to keep from blowing shit up when they exceeded 8500 RPM.

To correct the XR 750 profile for viewing and comparison with a different rocker arm ratio I did several things:

I calculated the difference in percentage between the Harley rocker arm ratio and the Norton rocker ratio. Then I used a spread sheet to multiply all the data with that percentage to give corrected data and applied that to get new cam lobe lift per degree data. I also entered the original measured data to make profile graph in autocad and used the scale command (percentage) to change the shape of the drawing. I had to use both techniques and come at it from both directions until everything added up. I think I changed aspect ratio as well to get it right (can't remember) It was a headache. Using a grid layer you can record all the new data points in autocad for your new lift per degree info. You can also use the spline command to smooth things out similar to "smoothing passes" used in some cam design software.

I also got some XR750 and other data from John Andrews. I'm using his cam design software right now but that's another story. It does all the number crunching for you but it won't spit out the shape you want unless you tweak the variables etc. None of this is easy.

As I do not wish to further hijack a thread so I will start a new thread on my damaged Ariel cam in the BSA section which may be appropriate as BSA took over Ariel.

Ando

ando said:

As I do not wish to further hijack a thread so I will start a new thread on my damaged Ariel cam in the BSA section which may be appropriate as BSA took over Ariel.

Ando

Click to expand...

No. Believe me. It is relevant to the discussion. Did you have a metallurgist do a report on your cam and lifters? If not, I suggest you do. A lot of comments on here but not much relevant to the actual reason for failure. Yours are exactly the same symptoms as mine.

Fullauto

I have started a thread in the BSA section and I think that Jims observation as well as cam graphs done on a similar cam have confirmed there is a problem with the design of the top of my cam which I hope with a follower alteration and better springs will be resolved.in particular i had cam graphs of a similar cam which showed a small problem in the area of the failure and my actions in using a flatter radius follower and unsuitabel springs would have increased the problem. more info in the BSA thread

I did not have any metallurgy done as I had sent my cam back to Franklin for rewelding and regrinding and have it back.
ando

Back to worn cams:

Below is a typical high milage stock cam with normal wear at the lobe nose from normal peak valve spring pressure. You can see a lip at the edge revealing the amount of wear.


Same as above but higher milage with a bigger lip indicating increased wear.


obvious wear on the lobe tip at peak lift - most cams have this kind of wear when they go out.



Stock cam breaking down at peak lift - some cams wear and others start to pit, some do both at this high stress area.


A wiped cam - usually a problem with soft 850 cams. This is not caused by loft or bounce which happens after the peak or after the valve closes - you can see that most of the wear is on the lift side of the cam leading up to the top of the lobe (cam turns counterclockwise).


The cams shown are all stock cams using stock springs with average spring pressure and none show valve bounce problems. Too low a spring rate can tangle valves, too much pressure on a medium RPM street bike will only shorten the life of the cam. A soft PW3 cam is going to wear even faster when you put the highest pressure springs on it - its worth some consideration.

Wear on the nose is most common - everyone has seen this. There are exceptions but looking through boxes of used cams tells the same story of worn out lobe tips.

The last time I looked at my stock cam back in the 20thC it was fine and quite unbenown to me
had been running in an oil with high zinc content.. Its done about 15,000 miles since , still on monograde but one of unproven zinc content. This issue with Zinc is all news to me, and the fact that the oil majors have pulled it likewise, in fact , if not for some chance remarks of Fast Eddie it would still be so..

Could it just be that some of these stock cams have failed because of oil related issues which have exacerbated problems of cam softness?

oldmikew said:

The last time I looked at my stock cam back in the 20thC it was fine and quite unbenown to me
had been running in an oil with high zinc content.. Its done about 15,000 miles since , still on monograde but one of unproven zinc content. This issue with Zinc is all news to me, and the fact that the oil majors have pulled it likewise, in fact , if not for some chance remarks of Fast Eddie it would still be so..

Could it just be that some of these stock cams have failed because of oil related issues which have exacerbated problems of cam softness?

Click to expand...

For sure, we don't know the history , frequency of oil changes or if the oil used was suitable.
My favourite cheap oil story comes from my cousin Bruce who is squeaky cheap.

In the 90s he had a Ford F250 with a good sized V8. He used it to haul a large camper. Safeway used to sell oil labelled " Scotch Brand" On sale it was 89 cents per quart and it was almost always on sale.
Bruce found some and happily changed the old black gunk to some new clean Scotch Brand of the appropriately rated viscosity.
First trip up to the BC Interior in hot summer weather with camper on and boat in tow, the truck is really struggling to make it up the first big climb.
It chugs over the top at about 15 mph instead of the usual 50.
Fortunately there is a gas station not far ahead so Bruce pulled in to get some help and figure out what is wrong.
The gas station mechanic came out, pulled the dipstick to see if the engine had oil. It did, although it was already down on the stick quite a bit.
The bigger problem was the viscosity of the oil, which Bruce said was thinner than water, more like rubbing alcohol.
They changed out the oil for some RPM Delo 400 and the truck had its old pulling power back.
Somehow he got away with the near seizure as the thing ran for years afterward.

Back when I took an engine overhaul course at LeTourneau University we had a section on cam break-in. It seems that VW had a rash of bad cams years ago and did extensive study to see what was wrong. The solution was Moly break-in lube. That was the only thing they ended up changing. You know I don't think that this would have worked if they had a soft cam to begin with but they found that the break-in was the most critical part of cam life. The cast cams seem soft to me just from knowing about cast iron but something about being punished with friction and pressure makes them harder pretty quickly. I have a Megacycle cam from Jim in my bike and also his radiused lifters. I have only had to adjust one valve in around 5000 miles so the wear rate is really low.

motorson said:

I have only had to adjust one valve in around 5000 miles so the wear rate is really low.

Click to expand...

When you adjust the valves you are adjusting the free play with the cam lobe pointing away from the tappet. Has nothing to do with wear on the lobe.

batrider said:

When you adjust the valves you are adjusting the free play with the cam lobe pointing away from the tappet. Has nothing to do with wear on the lobe.

Click to expand...

Very good point, what was I thinking? Megacycle are good hard cams though.

motorson said:

Very good point, what was I thinking? Megacycle are good hard cams though.

Click to expand...

Megacycle cams are just a normal Stellite [or similar] weld up. The Stellite weld up is as good as you can get. Comnoz

comnoz said:

Megacycle cams are just a normal Stellite [or similar] weld up. The Stellite weld up is as good as you can get. Comnoz

Click to expand...

So ideally the cams and followers should be of similar material or hardness .?.. I could be wrong but werent stellite followers introduced circa 72 ie not all commandos had them