Yamaha XS650 Valve Train Geometry

TwoManyXS1Bs

BBQ Hunter
Top Contributor
Messages
14,670
Reaction score
15,273
Points
813
Location
Fredericksburg, Texas
And now, this week's mind-numbing, obsessive-compulsive, winter cabin-fever diversion:

Yamaha XS650 Valve train, Rocker Arm and Adjuster Tip Geometry

Geometries used in the motions of the Cam Lobes, Rockers, Adjusters and Valves in the stock XS650 Head

For your pleasure, or anguish, preliminary educational backround data can be found at these links:

http://www.accessnorton.com/valve-adjuster-contact-geometry-t15583.html
http://www.accessnorton.com/rocker-contact-concepts-practice-t15759.html
http://www.performanceboats.com/gallery/data/500/medium/rocker_geometry.jpg
http://www.flowspeed.com/harland-sharp.htm
http://speedtalk.com/forum/viewtopic.php?f=1&t=3847
http://speedtalk.com/forum/viewtopic.php?t=9978
http://speedtalk.com/forum/viewtopic.php?f=1&t=20333
http://www.thesamba.com/vw/forum/viewtopic.php?t=167378&postdays=0&postorder=asc&start=0
http://mantonpushrods.com/pushrod-info/determine-length/
http://www.shoptalkforums.com/viewtopic.php?f=1&t=75703
http://shoptalkforums.com/viewtopic.php?f=18&t=145307
http://www.hi-flow.com/hp012dvt.html

Pushrod type valvetrains have multiple components and geometries that can be, and must be, configured properly to achieve maximum performance and longevity.
Rather complicated, but affords the tuner some options, and the mechanic some headaches.

On the other hand, the Over-Head-Cam XS650 valve train geometry is rather fixed.

About the only things we can tweak, within reason, are valve stem lengths (within tight limits), camshaft grind profile (base circle and lift), and adjuster tips (spherical or swivel).
Valve springs and collars can be tweaked, which can affect valve train loads, but they don't alter the basic valve train geometry.

Later, we'll try to determine the optimum spherical radius for our stock valve adjuster tips.

Attached is another explanation of rocker arm geometries.
 

Attachments

  • rocker_arm_geometry.pdf
    396.9 KB · Views: 535
Pictures for this thread can be found in this album:

http://www.xs650.com/media/albums/2492/


Starting with an excerpt from the factory service manual for the '447' engines:
full



Annotated to highlight the necessary basic geometric values:
full



For the sake of simplicity, some approximations and averages will be used.

So, we'll just use 8mm (0.315") as the cam lobe lift.

The two documented arm lengths for the rocker are:
40mm (1.575") for the follower (pad) side.
48.41mm (1.906") for the actuator end.

This gives a rocker lift ratio of 1.21 : 1.
Which produces a valve lift of about 9.7mm (0.382").
 
Good post. It appears that the sweep pattern mimics that of a well set up SBC with a solid lift cam and adjustable lash. I was always taught to start the roller tip of a rocker on the valve stem portion closest to the rocker itself and sweep across as lift increases and finally ending up at the valve stem tip portion that is furthest away from the rocker at max lift. Now, this was said to take the greatest advantage of available lift but not necessarily the best for guide wear. The old Comp Cams catalog use to have a little write up of the different ways to set up sweep patterns and the affects.
 
And now, the second half of this thing. Exploring the effects of various adjuster tip radii.
This is still a work-in-progress, but I decided to get some of this in here before I get run over by a truck.

The previous diagrams show that the rockers move through an 11.5° angle, but for the sake of simplicity, we'll use 12° as the rocker movement angle.

I've normally used about 3/8" (9mm) as the radius when regrinding adjuster tips (mostly on Hondas), and have always wanted to explore the effects of different radii on the XS650 valvetrain. The valvetrain geometry is fixed, and a bit strange, as it appears that the valvestem tip is about 2mm too high to get symetrical rocker sweeps.

So, here's the conundrum. The adjuster tip is spherical, and will 'roll' across the valvestem for 12° of rotation. The rocker arm is also doing a 'sweep' over the valvestem, trying to push it outward. The amount of 'roll' and the amount of 'sweep' don't cancel each other out, resulting in 'scrub' across the face of the valvestem. Fortunately, the adjuster tip is offset, and this 'scrub' imparts a 'rotation' onto the valve, which aids valve seat cleaning.

So, the first step in this exploration was to see if different adjuster radii could counteract the 'sweep'. The 'sweep' is also altered by this radius, as its sweep arm origin is at the adjuster tip's radius center.

This spreadsheet shows the results of radius values from 8mm to 30mm. I figured that 8mm should be the minimum to avoid small impact footprint, and 30mm is max because that puts the edge of the 6mm wide adjuster tip in contact with the valvestem.

full


I wanted to upload the spreadsheet, if anybody want's to play with it, or check it for errors, but it fails as it's an invalid file type (.xls).

What this spreadsheet reveals is that the 'scrub' value (0.41mm) remains essentially unchanged for all the possibilities. Interesting...
 
So, I'm considering 19mm and 24mm as candidates for tip radius, because they have the potential to soften the blow, spreading the impact over a larger area thru the use of a broader oil film (boundary layer cushion). Here's a close-up of the 9mm (3/8"), 19mm (3/4"), and 24mm (15/16") tip radii.


full


It looks like they'll remove at least 1/2 of the dead space...
 
What you have represented in your animation appears to jive with my earlier comment. Seems 9mm is very nice. The offset you mention to assist in rotation becomes even more obvious when running the swivel feet! Another area one could look at to assist in rotation would be to run valve locks that butt so that the valve is not "clamped" in place. Usually, a bead lock is used on the keeper when this is put into place. I do wish there was some adjustability with our valvetrain, both on the X and Y axis. What software are you using for your diagrams and animation?...very cool!

valve stem tip 1.png
valve stem tip 2.jpg
 
Hey, Josh! Thanx for the input. I've never heard of those abuting valve locks. More research, catching-up. The software is old stuff, done old-school. Intergraph images meticulously stitched together with Microsoft's old Gif animator. The backround is a 1mm grid. I plan to do one with the elephants foot adjuster as well, expecting a much different sweep pattern.
 
So the natural offset that rotates the valve is enhanced by a larger radius tip? Almost like the teeth in a ratchet?
If I have your drive in doing things like this I'd have the patent for the DW link! (bicycle rear suspension system) I was late to the punch...lol
 
Here's the animation of the XS650 rocker arm, with a 19mm adjuster tip radius.

If you can zoom in on the animations, for both the 9mm and 19mm tip radii, it appears that most of the scrubbing occurs in the first 3° of motion. Given that, I believe that 19mm is about as flat/broad as I'd like to go...


full
 
Here is a revised spreadsheet that includes the swivel foot (elephant foot) adjuster.

full


Notice that, for the elephant foot adjuster, the outward sweep is much less than for the solid adjusters. However, since there's no counteracting 'rolling' motion, all that 'sweep' movement becomes 'scrub' motion, about 1.25mm (0.049"). That's quite a bit more than the solid adjuster's 'scrub' motion of about 0.41mm (0.016").

Here's the animation of swivel foot (elephant foot) adjuster.

full
 
Hi to all. I contributed to 2 of those links mentioned in the first post of this thread.
*The 2nd one at accessnorton
*The one from thesamba (it gets better on page 9)

In both, I made it clear that the angle of the adjuster screw really means nothing and can't always be relied on.
If you want minimum scrub, take a square line of the valve tip and it should pass through the centre of the rocker spindle at 1/2 swing of the rocker arm (normally about 49% lift). That will work OK with a solid adjuster screw, but not with a swivel-foot adjuster. You'll be doing just fine up to about 2/3 rpm. At higher rpm, and if you rarely use less than 2/3, you could set it square at slightly lower lift. That's explained in the links.

If you use a swivel-foot adjuster, you should shorten the valve by an amount equal to to the distance from the bottom of the foot to the centre of the pivot ball. In other words, you are taking your square line to the centre of that ball. If that means chopping too much hardening off the valve tip, you'll need to cut more and use a lash cap.

Adjuster screw radius will not affect side thrust on the valve or guide. It will only affect how far the adjuster rolls across the valve tip. It can increase the amount of roll, but the amount of scrub will remain the same.
It's normal to have about 0.010" scrub between zero lift and mid-lift (pushing the valve away from the rocker spindle). From 1/2 lift to full lift, 0.010" of scrub will be pulling the valve toward the rocker spindle.

Scrub is the difference between how far the screw tip radius would roll if it weren't allowed to skid and how far it actually moves across the valve tip.
 
Welcome to the forum, X-file!

Thank you so much for your contributions on those other threads, and on this thread. They were the inspiration and guidance for this thread. We have virtually no documentation on the engineering geometries for our XS650 valve trains, and have had to resort to somewhat reverse-engineering this design.

The primary impetus behind this is to look into component resurfacing, increased longevity, reduced wear, and reduced valvetrain tapping.

Many of us (especially myself) are neurotic about the top-end clattering, and have pursued various solutions, like elephants foot adjusters, various oils and viscosities, additional oiling jets, ...etc.

These engines are very quiet on cold startup, up until they get hot. After that, we get some top end clattering, and would really prefer that the clattering be a sign of something wrong, rather than typical "just put up with it" sounds.

This picture reveals the typical offset of the rocker arm's adjuster, presumably intended to convert that "scrub" factor into a valve rotation, to enhance valve seat cleaning.
View attachment 47426

This picture of a typical valve stem end shows the doughnut-shaped wear pattern caused by the offset rocker arm adjuster.
View attachment 31557

This picture shows the typical valve, rocker arm, and adjuster arrangement. There is very little valve stem protruding above the keepers and spring collar. And, there is very little extra space below that, between the keepers' bottoms and valve guide. So, shortening of the valve isn't a viable option.
View attachment 47430

This manual extract shows the thin hardened stellite surface, restricting valve stem tip regrinds to a few thousandths.
View attachment 50142
 
Back
Top