Bill, they should be used with curved base followers with a radius of 0.75". Made my own followers years ago by getting someone to weld some Stellite on to standard followers and then using a pair of turned discs each side of the follower to act as a guide I hand ground them. Ah the optimism of youth, but it worked. 140 mph at the end of a half mile sprint on a bike ridden to the meeting. They just would not work with flat followers.
E: Engine Camshaft
- Thread starter robin stafford
- Start date
As a follow up to the above; oexing (Vic) sent me his rear short camshaft so that I could analyse it. This is an Andrews version of a Mk 2 The results are below. First I show the lift curves. These were taken from the ends of the valve stems.
Note that I set up the timing so that the cross over point is at about 4 deg BTDC as this seems to be a popular way of doing the timing. Nest I show the start and end of the inlet lift curve at higher magnification.
You will see that there is a quieting ramp at both the start and end of the lift curve. Vincent timing is taken at five thou lift and from these two curve we can see that Inlet opens about 60 deg BTDC and closes at about 65 deg ABDC. Five deg early for the start of the lift and three deg early for the end of the lift.
Below are the same data for the exhaust.
Once again there is a clear ramp at the start of the ex lift and a smaller one at the end of the lift. Timing is Ex opens at about 83 deg BBDC and closes at about 30 deg ATDC., Ten deg early and twenty deg early. If the cam was retarded by six degrees then the first of these three timing would be about right but that would put the cross over point at about two deg after TDC, not four degrees before.
Next I show the velocity graphs.
Note the little blip at the start of the inlet lift. The vertical scale is thous per engine degree.
Finally I show a smoothed acceleration curve. This was smoothed by taking running means of three points.
Note that now the vertical scale is thous per engine degree squared. For those of a metric persuasion
a reading of .00005 inches is 1.27 microns. My digital dial gauges have the fourth figure as either a zero or a five, that is they are only able to read to half a thou and so I cannot do any better than this. You will see how the quieting ramp at the start of the inlet lift shows up on the acceleration curve. The thing which is a pleasure to see it that the negative part of these curves, which is where the valve springs are doing their job, is approximately horizontal, never seen on standard Vincent cams.
The overlap is about 95 deg whereas the official Vincent figure for Mk 2 s is 105 deg. The value for a Mk 1 cam is only 68 to 75 degrees so these are certainly more Mk 2 than Mk 1.
Over to the experts.
Note that I set up the timing so that the cross over point is at about 4 deg BTDC as this seems to be a popular way of doing the timing. Nest I show the start and end of the inlet lift curve at higher magnification.
You will see that there is a quieting ramp at both the start and end of the lift curve. Vincent timing is taken at five thou lift and from these two curve we can see that Inlet opens about 60 deg BTDC and closes at about 65 deg ABDC. Five deg early for the start of the lift and three deg early for the end of the lift.
Below are the same data for the exhaust.
Once again there is a clear ramp at the start of the ex lift and a smaller one at the end of the lift. Timing is Ex opens at about 83 deg BBDC and closes at about 30 deg ATDC., Ten deg early and twenty deg early. If the cam was retarded by six degrees then the first of these three timing would be about right but that would put the cross over point at about two deg after TDC, not four degrees before.
Next I show the velocity graphs.
Note the little blip at the start of the inlet lift. The vertical scale is thous per engine degree.
Finally I show a smoothed acceleration curve. This was smoothed by taking running means of three points.
Note that now the vertical scale is thous per engine degree squared. For those of a metric persuasion
a reading of .00005 inches is 1.27 microns. My digital dial gauges have the fourth figure as either a zero or a five, that is they are only able to read to half a thou and so I cannot do any better than this. You will see how the quieting ramp at the start of the inlet lift shows up on the acceleration curve. The thing which is a pleasure to see it that the negative part of these curves, which is where the valve springs are doing their job, is approximately horizontal, never seen on standard Vincent cams.
The overlap is about 95 deg whereas the official Vincent figure for Mk 2 s is 105 deg. The value for a Mk 1 cam is only 68 to 75 degrees so these are certainly more Mk 2 than Mk 1.
Over to the experts.
When I bought mine, from George Emmerich in 1976, he told me to split the overlap at TDC. We ran them at Bonneville in Bob Guptill's bike on back to back runs with Mk2's. The results were virtually identical.
I really appreciate your tests, Thanks
I really appreciate your tests, Thanks
D
Deleted member 3831
Guest
Bruce
what horizontal line are you talking about ?
Re constant accelleration, it can only be applied to sections, not the overall design. Most often it was applied to the opening and closing ramps, but this concept was generally replaced by constant velocity, again applied to the ramps.
Regarding measurement of a cam profile, or lift at the valve, measurements taken at degree increments of greater than 1 degree will not result in data that can be used in any meaningful way except to determine lift and duration. Similarly, linear measurement of less than 5 decimal places can only be useful for the second derivative, beyond that, with all the best extrapolative aids, it still remains largely guesswork.
Anyone who doesn't believe this should get him/herself a copy of Don Hubbard's book 'Camshaft Reference Handbook 1999', and therein will be found a much more detailed explanation than I have room for in this post
what horizontal line are you talking about ?
Re constant accelleration, it can only be applied to sections, not the overall design. Most often it was applied to the opening and closing ramps, but this concept was generally replaced by constant velocity, again applied to the ramps.
Regarding measurement of a cam profile, or lift at the valve, measurements taken at degree increments of greater than 1 degree will not result in data that can be used in any meaningful way except to determine lift and duration. Similarly, linear measurement of less than 5 decimal places can only be useful for the second derivative, beyond that, with all the best extrapolative aids, it still remains largely guesswork.
Anyone who doesn't believe this should get him/herself a copy of Don Hubbard's book 'Camshaft Reference Handbook 1999', and therein will be found a much more detailed explanation than I have room for in this post
Grey One is correct with regards to the accuracy one need to have available to do a good job. I, and most people, do not have such equipment and in any case with that kind of measuring accuracy one would want to have a temperature controlled environment. Lacking all that one does the best that one can.
The 'horizontal' line referred to by Bruce would be a hand drawn line through the negative parts of the acceleration curves. I do realise that the exhaust is noisy and the inlet has a curve in it but compare these two curves with those obtainable with any of the normal Vincent cams. All the ones I have tested have a gradient along this part of the curve, one gradient for the exhaust and the opposite gradient for the inlet. My understanding, and I do not mind being corrected by an expert, is that this negative part of the curve shows what the valve springs have to try to control and a curve which gives a horizontal straight line is the one to try for. Vincent cams do not satisfy those criteria and these cams are better than the standard cams. Could the curves be improved? Very possibly, but this is what we have at the moment. All I have done is to measure the cams, not designed them so I have no personal bias or vested interest as to their suitability.
I do have the facilities for accurately rotating the engine one degree at a time (I have a rotary table clamped to a special nut which fits to the timing side main shaft outboard of the half time pinion). I could probably even arrange to have a stable temperature. Perhaps Grey One would like to lend me a measuring device accurate to a micron or so and I will find the time to do the job as he would wish.
The 'horizontal' line referred to by Bruce would be a hand drawn line through the negative parts of the acceleration curves. I do realise that the exhaust is noisy and the inlet has a curve in it but compare these two curves with those obtainable with any of the normal Vincent cams. All the ones I have tested have a gradient along this part of the curve, one gradient for the exhaust and the opposite gradient for the inlet. My understanding, and I do not mind being corrected by an expert, is that this negative part of the curve shows what the valve springs have to try to control and a curve which gives a horizontal straight line is the one to try for. Vincent cams do not satisfy those criteria and these cams are better than the standard cams. Could the curves be improved? Very possibly, but this is what we have at the moment. All I have done is to measure the cams, not designed them so I have no personal bias or vested interest as to their suitability.
I do have the facilities for accurately rotating the engine one degree at a time (I have a rotary table clamped to a special nut which fits to the timing side main shaft outboard of the half time pinion). I could probably even arrange to have a stable temperature. Perhaps Grey One would like to lend me a measuring device accurate to a micron or so and I will find the time to do the job as he would wish.
Norman got a copied cam that I did from a genuine Andrews set of cams I had for almost 30 years. Well, some projects take longer than you did think at that time. Anyway, we wanted to see if these cams would be a recommendation for more people as I believe Megacycle still do exactly this version of MK 2 , developed by Andrews - which were taken over by Megacycle and described as "modified " MK 2 s. Just looking at my Andrews I expected at least SOME quieting ramps and judging from the very great finish I do believe the guys at Andrews knew their business well enough to recommend these cams. Norman was able to confirm quieting ramps, albeit a bit less on the closing side of the lobes. There is a lot of measuring uncertainty in our setups so we do not claim to have absolute truth in our findings. But anyway , these cams seem to be the best that is still available and semi-professionally measured as well.
I got three sets from Andrews in the eighties and do not believe there were no other sets in use in bikes whose owners have a look into this forum and could provide some real life experience from road or track mileage ???
We do not pretend to be able to design cam shapes so no need to discuss this matter. Our object was to compare this set with anything that may be around for copying or new production in order to possibly enhance some future spares for better wearing valve train.
So hopefully this is allright for Norman when I ask for more info on any cams suitable for measuring and completing databases ?? I do not think they did everything in an optimal way more than half a century ago at the factory . So a more modern pattern for copying and new production - or just a bit of tweaking the Megacycle cams , should there be minor imperfections - may be seen positive in the club ??
Vic
My copied Andrews cams :
I got three sets from Andrews in the eighties and do not believe there were no other sets in use in bikes whose owners have a look into this forum and could provide some real life experience from road or track mileage ???
We do not pretend to be able to design cam shapes so no need to discuss this matter. Our object was to compare this set with anything that may be around for copying or new production in order to possibly enhance some future spares for better wearing valve train.
So hopefully this is allright for Norman when I ask for more info on any cams suitable for measuring and completing databases ?? I do not think they did everything in an optimal way more than half a century ago at the factory . So a more modern pattern for copying and new production - or just a bit of tweaking the Megacycle cams , should there be minor imperfections - may be seen positive in the club ??
Vic
My copied Andrews cams :
I raced them and used them on the road for a short time. In 1977 I didn't have my race bike ready for Bonneville, so I stripped down my D Shadow and ran it. It went 124.87mph. I used them for a short time on the road but only thought they were "jazzed up" Mk2's and replaced them with the standard cams for two up touring. If I'd had Norman's info then, they might still be in my bike. They're sitting on my shelf.... Hmmm, decisions, decisions..
D
Deleted member 3831
Guest
Sorry folks, but I don't work in half measures, and measuring cam profiles or valve lifts at 5 degree increments and 3 places of decimals, guestimating the fourth data figure is not my way of working.
If you want to compare cam profiles, you need data for every degree of rotation, even the base circle, and to an accuracy of at least 6 decimal places. My linear encoder reads data to eight decimal places, and the rotary encoder to 1/60 th of a degree if necessary. I am not interested in loaning this equipment out to anyone.
I design cams for race engines, but I have measured cams for friends and others where a detailed record of a profile was wanted. I did such an exercise some years ago for members of the Aerial club, they arranged to send me cams, some coming from distant parts of the world, and after measuring all the profiles, they have now a complete set of data for each profile, in a form that can be used for the manufacture of new cams should ever the club decide they are needed.
With elderly cams, some wear usually has to be allowed for, and inevitably some of the old designs left much to be desired. Problems of this kind were eased by using modern smoothing methods, while leaving the original character of the design more or less intact.
If anyone would care to send me one of these Andrews cams, I will measure it and then return it with a data file in S96 format, which can be opened in Notepad or Wordpad to view the data, or in an additional R file, which contains full manufacturing data.
If you want to compare cam profiles, you need data for every degree of rotation, even the base circle, and to an accuracy of at least 6 decimal places. My linear encoder reads data to eight decimal places, and the rotary encoder to 1/60 th of a degree if necessary. I am not interested in loaning this equipment out to anyone.
I design cams for race engines, but I have measured cams for friends and others where a detailed record of a profile was wanted. I did such an exercise some years ago for members of the Aerial club, they arranged to send me cams, some coming from distant parts of the world, and after measuring all the profiles, they have now a complete set of data for each profile, in a form that can be used for the manufacture of new cams should ever the club decide they are needed.
With elderly cams, some wear usually has to be allowed for, and inevitably some of the old designs left much to be desired. Problems of this kind were eased by using modern smoothing methods, while leaving the original character of the design more or less intact.
If anyone would care to send me one of these Andrews cams, I will measure it and then return it with a data file in S96 format, which can be opened in Notepad or Wordpad to view the data, or in an additional R file, which contains full manufacturing data.