Camshaft 850 stock vs. PW3

Status
Not open for further replies.
Joined
Jan 1, 2018
Messages
60
Country flag
I am doing a bit of work on my 850 MkII engine at the moment. Before I took the engine apart, I measure the valve timing. It seemed to be stock camshaft, not too worn imo. Now the crankcase is back together with a PW3. Here a comparison between the two camshafts, same chain, sprockets etc. used:

Camshaft 850 stock vs. PW3


The PW3 is a bit out compared to the spec sheet that came with it (from Andover Norton). Inlet and Exhaust open 3 degrees too early, both close about 7degrees too early. Can I assume that the timing of opening is more important? should I go for getting the opening right or the closing?

Then I put barrel and head on and checked the clearance between valves and piston. Without head and base gasket the inlet valve is nearly touching the piston. I wonder if it is normal for PW3s to be so close. Or am I missing something?

Camshaft 850 stock vs. PW3


I have to admit that I have never done anything regarding valve timing or changing to a hotter camshaft before. So even though I measured and checked everything twice at least there is no garuantee that I have missed anything. I appreciate any advice.
 
Last edited:
You can juggle the timing by 5 degrees at a time by varying the positions of the cam drive sprockets. I've never done it, but someone who has can tell you how to get that PW3 back on point.
 
The head gasket will be at least .040” thick, probably closer to .060” which provide that amount of valve to piston clearance.

Also, if the cam is opening too early, that means it’s open more at TDC than it should be, thereby putting the valve closer to the piston.

As Danno said, you can adjust timing in increments of 5 degrees very easily. That will,put you within 2 degrees of ‘perfect’ which, IMHO, is plenty good enough.

Retarding by 5 degrees and adding gaskets will give you ample valve to piston clearence I think.
 
Have you tried checking the cam timing with the lobe center method? It's generally more accurate than trying to do it with opening and closing points. As near as I can tell from your graph, your intake lobe center is pretty close to 102 degrees atdc as installed, which does seem a little early for something like the PW3. However, your graph doesn't have enough resolution for me to get the figure exact. You probably have better data, and can calculate it more precisely. I don't have the specs handy, but I would expect it to have an intake lobe center around 105 degrees (intake lobe center specs for Commando race cams are typically 104 - 106 degrees, but can be as low as 99 and as high as 108). If so, your cam is a bit advanced. Conventional wisdom is that you will get a little more low and mid-range grunt with that setting, but a little less top end power. I'm not sure you will notice the difference unless you are racing it. If you do feel the need to get it closer, you can either buy a Vernier sprocket, or make an offset key to advance it by the appropriate amount. As mentioned above, you can advance it by 5 degrees easily, but anything less than that requires the Vernier sprocket or offset key. Or, you can cut more keyways in a stock sprocket, but that is a real pain to do. Someone used to sell sprockets with alternate keyways in them, but I don't know if they are still available.

Ken
 
Two options:

1. Check the valve/piston clearance with the gaskets in place/head 'torqued' down to spec before calling it a done deal. If this means you wouldn't be comfortable using a new gasket and re-using it after checking the clearance, then buy two gaskets and do the check with the first and then use the other one after the check.

OR...

2. If you KNOW the COMPRESSED thickness of the gasket - some manufacturers list that - you can just do the check as you did, without a gasket, and add the compressed gasket thickness to whatever the clay shows.

FWIW, I have never done cam performance cam installation (greater than factory lift/duration) on a Norton so this next comment is based on non-Norton engines: We always used .080 as figure for intake valve clearance and .100 for exhaust but obviously there are folks here who have done this a lot with Nortons and if less than that has proven to be no trouble I'd go with whatever is known to work/recommended.
 
Running a web312 with about 5 degrees of intake advance and radiused lifters, no clearance or tangle issues and power comes on around 4000 RPM
 
The cam in my 850 has been reground, but it is not much different from standard 850. The cam timing you use for best power depends on the exhaust system you have fitted. If you advance the exhaust opening BBDC, you get a louder bang which drives the exhaust system harder and moves the power band higher, if you are not using a 2 into 1 exhaust. With a 2 into 1, you get more torque. When you advance the cam, the inlet opens earlier BTDC and you get a bigger gulp of fuel, but more power. The valve opening points are more important than the closing points. I never work with lobe centres when doing the cam timing. Finding the lobe centres is much less accurate than finding the opening points with nil tappet clearance. I usually check both opening and closing points and share the difference which occurs at both ends of the cam travel So if the exhaust opens at 72 BBDC and closes at 53 ATDC, if there is a 4 degree difference at one end, I move the cam 2 degrees to share the error.

The cam in my 850 is advanced 12 degrees, so you are unlikely to tangle the valves in a normal 850, if you get the timing wrong by only a few degrees.
 
Last edited:
My feeling is that the exhaust system on a Commando is not much difference to the exhaust system on a two-stroke. You still have Kadency effects. It you cut the exhaust port up in a two stroke, you move the power band higher, but the shape of the expansion chamber changes the effect on torque. A 2 into 1 exhaust system on a Commando is similar to a fat expansion chamber on a two stroke. In the end, you are playing with two resonating tubes on a Commando. They interact with one another when both valves are open at the top of the stroke. Moving the lobe centres with respect to each other, might be of some benefit. But I think Mr Norton had some prior knowledge of cams before he made Commandos. With the Manxes, there was probably a lot of experimentation by moving lobe centres and changing profiles. They were raced for a very long time. The only British single which was ever better was the 1959 AJS 7R.
 
The power in my 850 comes on at nothing and goes straight through the top. However I do not use it like that when racing. The heavy crank tends to spin up at always the same rate regardless of the gearing. So I use a close box and keep the crank spinning well above 5000 RPM and below 7000 RPM. When you race-change doing that you get extremely rapid acceleration - I ride the top of the torque curve. Gear ratios and steering geometry are important. If your motor bogs down in a corner, you will be slower. On the slowest corner on Winton Raceway, my motor does not drop below 5000 RPM. The slowest corner on any circuit determines what happens on the rest. With many bikes you have to choose between low and high overall gearing, so you either get out-accelerated coming out of corners, or out-gunned at the ends of the straights.
 
Last edited:
Have you tried checking the cam timing with the lobe center method? It's generally more accurate than trying to do it with opening and closing points. As near as I can tell from your graph, your intake lobe center is pretty close to 102 degrees atdc as installed, which does seem a little early for something like the PW3. However, your graph doesn't have enough resolution for me to get the figure exact. You probably have better data, and can calculate it more precisely. I don't have the specs handy, but I would expect it to have an intake lobe center around 105 degrees (intake lobe center specs for Commando race cams are typically 104 - 106 degrees, but can be as low as 99 and as high as 108). If so, your cam is a bit advanced. Conventional wisdom is that you will get a little more low and mid-range grunt with that setting, but a little less top end power. I'm not sure you will notice the difference unless you are racing it. If you do feel the need to get it closer, you can either buy a Vernier sprocket, or make an offset key to advance it by the appropriate amount. As mentioned above, you can advance it by 5 degrees easily, but anything less than that requires the Vernier sprocket or offset key. Or, you can cut more keyways in a stock sprocket, but that is a real pain to do. Someone used to sell sprockets with alternate keyways in them, but I don't know if they are still available.

Ken


Thanks Ken. So if I interpret your words correctly, the lobe center timing is more important than the opening and closing figure? Does make sense to me, considering that the first degrees after opening there is not much valve lift anyway.
If I look in my graph, the intake lobe center is at 100 degrees. If I calculate the figure based on the Andover spec sheet, intake lobe center is supposed to be at 106 degrees. I think I will try 5 degrees retard (moving sprockets and gears) and it should be good.

This will hopefully give me a bit more valve/piston clearance.

I was planning to run without base gasket and a thin JS copper gasket (0.5mm). This will increase the compression from 7.8:1 (stock) to 8.35:1. I wanted to take a bit off the head too to get closer to 9:1 but not anymore. May look for different pistons now.

Thanks everyone for your comments and ideas. I will update once I changed the timing.....
 
The cam in my 850 has been reground, but it is not much different from standard 850. The cam timing you use for best power depends on the exhaust system you have fitted. If you advance the exhaust opening BBDC, you get a louder bang which drives the exhaust system harder and moves the power band higher, if you are not using a 2 into 1 exhaust. With a 2 into 1, you get more torque. When you advance the cam, the inlet opens earlier BTDC and you get a bigger gulp of fuel, but more power. The valve opening points are more important than the closing points. I never work with lobe centres when doing the cam timing. Finding the lobe centres is much less accurate than finding the opening points with nil tappet clearance. I usually check both opening and closing points and share the difference which occurs at both ends of the cam travel So if the exhaust opens at 72 BBDC and closes at 53 ATDC, if there is a 4 degree difference at one end, I move the cam 2 degrees to share the error.

The cam in my 850 is advanced 12 degrees, so you are unlikely to tangle the valves in a normal 850, if you get the timing wrong by only a few degrees.


I am running 850 MKII stock pipes and I am not planning to change that. The engine rebuilt and tuning consumes too much money and time already. But am always keen to learn about how it all works together so thanks for your input.

Your method of timing (taking the middle between opening and closing points) sounds pretty much like timing the lobe center to me. If I think about it, I tend to agree this should be the most appropriate way of timing.
 
Very helpful guide for degreeing: http://www.webcamshafts.com/pages_misc/degreeing.html

Once you've sorted the timing at the stock sprocket arrangement tweak to suit your needs, a good starting point is to aim for slightly more intake lift rather than a straight up split.

RGM sells a cam sprocket with three broached keyways, standard, +3 and +6 degrees. This was immensely helpful when I was tweaking my cam:
https://www.rgmnorton.co.uk/buy/special-camshaft-sprocket-less-weight-adjustable-timing_4641.htm
 
Thanks Ken. So if I interpret your words correctly, the lobe center timing is more important than the opening and closing figure? Does make sense to me, considering that the first degrees after opening there is not much valve lift anyway.
If I look in my graph, the intake lobe center is at 100 degrees. If I calculate the figure based on the Andover spec sheet, intake lobe center is supposed to be at 106 degrees. I think I will try 5 degrees retard (moving sprockets and gears) and it should be good.

This will hopefully give me a bit more valve/piston clearance.

I was planning to run without base gasket and a thin JS copper gasket (0.5mm). This will increase the compression from 7.8:1 (stock) to 8.35:1. I wanted to take a bit off the head too to get closer to 9:1 but not anymore. May look for different pistons now.

Thanks everyone for your comments and ideas. I will update once I changed the timing.....


Be sure you determine the lobe center from the cam data, not a visual estimate from the graphs. It's very easy to be off by a couple of degrees that way. I did it by blowing up your graph, and then measuring the opening and closing points at 2 mm lift. From those you can calculate the actual lobe center. For me, that came out around 102, but if you have a more precise measurement of the actual opening and closing points, you can get a more accurate value. If that' s what you did, and it comes out at 100, then retarding the cam 5 degrees should work out just fine.

Ken
 
Did your cam come with a card telling you the recommended lift on the inlet cam at TDC ? Mike Hemmings supplied his PW3 cam with just such a card. This is the normal recommended way to dial in a Norton cam and I found it a lot easier than trying to determine lobe centres. On a Norton the cam chain tension changes and you need to keep turning in one direction only if you are to get clear repeatable values.

I dialed in my cam using a dial guage on the push rod with the head off. Rotate the engine forward and determine the correct lift at inlet TDC. It is stating the obvious with a single cam if you change the inlet timing the exhaust is automatically changed too.

That RGM sprocket looks a good plan. I don't recommend vernier sprockets for Norton cams. None of them are robust enough for long term use in my experience

Mexico Mike has given some numbers for valve clearance. I have gone a bit closer than that on a 500 race bike but his numbers will be definitely enough in my experience.

One thing I haven't seen is any mention of checking for valve spring coil bind at full lift. If you install a bigger cam you must do this. The infamous Les Emery says you must have over 100 thou but I have used down to 60 thou on my race bike.

If your setting up springs the issue is not so much about installed height but more about clearance from coil bind and control at the peak acceleration and deacceleration of the valve train. Usually you set them up with a target lb over the nose of the cam. Installed height is a short cut for a standard situation assuming everything is just as it came from the factory. Change anything especially cams and valve spring type and they mean very little.

The way I do it is to measure the lb versus lenght of the spring in a spring tester. Any good engine tuning shop will have one. Graph it and subtract the clearance for coil bind. You then have the maximum lb available to work with over the nose of the cam. I use a max of 210 lb on the springs on my 500 Norton running up to 7500 but this could be a bit too much. On standard bikes i think the number is more like 160 - 180 lb over the nose.

I install the valve spring, set the cam to peak lift and then compress the spring the remaining distance to coil bind measuring the distance with a dial guage. You then can calculate from the graph how much packing you need for a target spring lb over the nose.

That's the way I have usually seen guys set up springs on race bikes.
 
Last edited:
Valve drop at TDC is the way later Manxs were timed, but the opening points determine where the power band starts and fuel economy, depending on the exhaust system.. If you use drop at TDC, you might get more power, but it might not be where you want it. Then you start trying to fiddle the gearing again. But you probably have to do that anyway. What I don't like is getting a savage bump when the power cuts in. It makes it too easy to drop the bike in tight corners when you are going fast.
 
The PW3 was designed by Peter Williams to be a smooth torque cam. No big power hits - recognizing it is still a sports cam of course.

Not sure about what you mean by getting more power using valve drop. I'm not talking about tuning the cam. Just setting it up as standard. Lobe centres, lift at 50 thou, lift at TDC all take you to the same place. If your tuning the bike then you will have to do a few experiments and dyno runs to get optimal cam timing. Thats a whole new ball game and I cant see why anyone would bother for a street bike.

I don't have an issue with timing using lobe centres. That can work fine. I just found lift at TDC a much easier way to do it. And with a single cam with 4 lobes once you have set timing on one ramp everything else comes along for the ride - unless you get the hacksaw out of course :).

That said I'm a bit surprised what the OP said about the cam having off spec timings between the front and back of the lobes. When I dialed up two PW3 cams from Mick Hemmings they were text book perfect.

4 S cams from Emery had issues with wandering base lines, strange and different shaped quieting ramps and different lobe centres. eg the left hand inlet lobe had a different lobe centre to the right hand inlet lobe by around 3 degrees. They were ground that way.
 
Last edited:
Thought I should mention my PW3 experience. Chilled cast iron PW3 cam purchased 8/1/2015 from Mick Hemmings, set at 145 thou lift of inlet pushrod at TDC, snapped in half adjacent to and inboard of the rev counter scroll after 15,ooo miles in an 850 mk 2. I also fitted a standard chilled cast iron camshaft purchased from Andover Norton 6/11/2012 which has worked well for 80,000 miles in a 1970 Fastback. Hope all goes well for you.
 
The PW3 was designed by Peter Williams to be a smooth torque cam. No big power hits - recognizing it is still a sports cam of course.

Not sure about what you mean by getting more power using valve drop. I'm not talking about tuning the cam. Just setting it up as standard. Lobe centres, lift at 50 thou, lift at TDC all take you to the same place. If your tuning the bike then you will have to do a few experiments and dyno runs to get optimal cam timing. Thats a whole new ball game and I cant see why anyone would bother for a street bike.

I don't have an issue with timing using lobe centres. That can work fine. I just found lift at TDC a much easier way to do it. And with a single cam with 4 lobes once you have set timing on one ramp everything else comes along for the ride - unless you get the hacksaw out of course :).

That said I'm a bit surprised what the OP said about the cam having off spec timings between the front and back of the lobes. When I dialed up two PW3 cams from Mick Hemmings they were text book perfect.

4 S cams from Emery had issues with wandering base lines, strange and different shaped quieting ramps and different lobe centres. eg the left hand inlet lobe had a different lobe centre to the right hand inlet lobe by around 3 degrees. They were ground that way.

Also note that the values provided by the cam manufacturer are likely based on a specific follower profile, so using radiused lifters with a cam designed for flats will make anything but the lobe center values redundant
 
Status
Not open for further replies.
Back
Top