1968 250 MK3D rebuild - ex W/C Oil Pump

[machten]

Not sure if it would help or not but it's at least easy to do...
I took one of my stock closers and figured out the radius to be 35 mm. You can set your digi calipers to that and use the inside measuring points to scribe said radius (you obviously don't need the whole circle) into a piece of hard plastic or aluminum sheet and line the rocker face up to it and see if the radius looks ok (bigger, smaller, compound). With a bit of magnification I could easily see the deviation if the scribed radius was made a few mm larger or smaller. You could also just draw it on paper but a scribed line is much more clean/concise.

To measure for relative orientation, with the rocker held in place on the first scribe line, scribe a circle around the inside of the rocker shaft hole and measure from the center of that one to the center of the 35mm radius (the pivot point from the caliper, would be the middle of a 70mm circle). Mine was 48 mm in that dimension.

Bill, any chance you could check the opener for me if you have one readily about? Is the radius the same?

Kev

[wcorey]

Bill, any chance you could check the opener for me if you have one readily about? Is the radius the same?

Yes. same radius, though I would have assumed otherwise as the contact to pivot distance is longer than the closers. Maybe it doesn't matter so long as the radius is linear. Seems like the often seen lowest clearance point variations being discussed here could be just as easily due to rockers as cams.

The other relative circle to circle center measurement is different at 54 mm and also slightly more difficult to set up on the opener because the shaft protrudes out further. I bent a step into the thin aluminum sheet I'm scribing on to compensate.

Bill

[DewCatTea-Bob]

" though I would have assumed otherwise as the contact to pivot distance is longer than the closers. "

____ The c.contact-point to pivot to v.contact-point is referred to as the 'rocker-ratio'.
On the opening-rocker, the pivot to contact-point at the valve is near 4% longer than the cam-follower contact-point to pivot. _ Thus 'actual valve-lift' is a bit higher than the 'lobe-lift'.


" Seems like the often seen lowest clearance point variations being discussed here could be just as easily due to rockers as cams. "

____ I don't see how that could be !
Also, I'm sure that the follower-radius doesn't vary at all notable, unlike the fairly notable variances of the closer cam-lobes.

____ I've included below a picture of a DESMO closing-rocker, showing the forked-end closing-arm appearing as a bit shorter than the rocker's cam-follower.
It would be nice to know exactly what IT's rocker-ratio actually is.
__ I could some day dig-out a closing-rocker to find-out but, since Bill is already set up with a handy method/set-up to possibly check the ratio with, then all he'd have to do is take note of how many thousands of arc-distance the center of the contact-point on the fork actually moves, compared to (for example) .250" of rotational-movement of the center of the contact-point on the cam-follower's contact-pad. ...
In other words, if the exact-center of the follower-pad is moved (around the rocker's pivot-point) exactly 1/4", and we then find -(measure) that the center of the fork's contact-point moved merely .240",, then we'd know that the closing-rocker's ratio is the exact reverse of the opening-rocker's. - (Or of course, whatever it's actual rocker-ratio really might be.)


Dukaddy-DUKEs,
-Bob

[machten]

Bill,

Thanks for the opener measurement, and also for the suggestion of the scribing method with the digital vernier. That worked a treat on a clear plastic CD cover - I thought you said you were starting to miss the obvious?

I checked my closer against your measurement and it was (as close as I can measure) exactly the same. My opener was the same 35mm radius too. I have not yet checked the offset you provided.

though I would have assumed otherwise as the contact to pivot distance is longer than the closers.

and I was hoping for something like that, specifically a smaller radius.

From DCT Bob..

____ The c.contact-point to pivot to v.contact-point is referred to as the 'rocker-ratio'.
On the opening-rocker, the pivot to contact-point at the valve is near 4% longer than the cam-follower contact-point to pivot. _ Thus 'actual valve-lift' is a bit higher than the 'lobe-lift'.

So, if we assume that the opening and closer rockers are actually designed to be the same radius (35mm) , then the compensation for the obviously different rocker ratios in order to execute what must result in a very close to a mirror image on the actual opener and closer valve lifts must therefore be designed in to the relative positions and the specific shapes of the opener and closer cam lobes.(and it seems, maybe not be very well executed in the cases of the cams I've installed for test)

" Seems like the often seen lowest clearance point variations being discussed here could be just as easily due to rockers as cams. "

____ I don't see how that could be !
Also, I'm sure that the follower-radius doesn't vary at all notable, unlike the fairly notable variances of the closer cam-lobes.

Actually I'm thinking Bill's statement is theoretically and practically correct. For example I think I can think of two situations where this might apply:

Too large a radius on the opener rocker so the opener rocker enages too early compared to the closers willingness to release (this was my anticpated scenario)
The state the rockers were originally in - ie they do not have one radius. (kind of esoteric, admittedly)

By end of this weekend I'll be able to test with some stock rockers, but I'm starting to think it will make no difference. Once agin, we'll see.

Kev

[DewCatTea-Bob]

" I checked my closer against your measurement and it was
exactly the same. My opener was the same 35mm radius too. "

____ For any others 'trying' to follow this, it seems that Kev has now found that the radius of his refaced closing-rocker's cam-follower, is virtually the very same as that which Bill had measured from his-own (stock/non-refaced) closing-rocker.
Furthermore (it now seems), Kev also realizes that all the refaced cam-followers (on both of his D.rocker-arm pairs), have pretty-much the very same radius, as stock.


" So, if we assume that the opening and closer rockers are actually designed to be the same radius (35mm) , then the compensation for the obviously different rocker ratios in order to execute what must result in a very close to a mirror image on the actual opener and closer valve lifts must therefore be designed in to the relative positions and the specific shapes of the opener and closer cam lobes.(and it seems, maybe not be very well executed in the cases of the cams I've installed for test)

____ All that seemed to make fair sense (to me) AFTER I had figured-out where I think you should've added (helpful) punctuation.


" Actually I'm thinking Bill's statement is theoretically and practically correct. For example I think I can think of two situations where this might apply:
Too large a radius on the opener rocker so the opener rocker enages too early compared to the closers willingness to release
The state the rockers were originally in - ie they do not have one radius. "

____ Well of course BUT, the spec.figures of the rocker-arms have not ever been so noted to vary any at all,, UNLIKE the measurements of the D.cams !
__ Perhaps YOU had assumed that Bill was only referring to just YOUR refaced-rockers, but since THAT particular (seemly likely) possibility had already been well pointed-out previously, I-MYSELF thus figured that he was referring to STOCK-DESMO rockers & cams.
____ (The possibility of this type of non-singular interpretation of imprecise-wording often leads to such occurrence, which is why MY-OWN [extended type] wording is done in an attempt to avoid that .)


" I'll be able to test with some stock rockers, but I'm starting to think it will make no difference. "

____ Just as I've expected, since your conflict-binding clearance-figures were well within that which have already been known to exist (in at least SOME old D.singles!).
__ So it now seems that you're getting closer to accepting that you'll have to relax your closing-clearance/space (instead of the opening-clearance), so as to properly allow the opening-function to smoothly perform it's (complete) task, (along with properly shimmed v.stem-clearance).


Hopeful-Cheers,
-Bob

PS. My previous posting (also on this page) has had some more wording added to it which I had left-out before (due to a common-distraction which prevented me from having included during the original-time).

[machten]

____ All that seemed to make fair sense (to me) AFTER I had figured-out where I think you should've added (helpful) punctuation.

My apologies, I was rushing a bit at the time.

__ Perhaps YOU had assumed that Bill was only referring to just YOUR refaced-rockers, but since THAT particular (seemly likely) possibility had already been well pointed-out previously

Yes, that was my assumption based on the context of my repaired rockers. Even so, I'd suggest my scenarios presented regarding that particular point are both theoretically possible (and given my rockers have been repaired - practically possible), and given the intractability I'm experiencing resolving this, I'm finding the need to cast my net widely!!

__ So it now seems that you're getting closer to accepting that you'll have to relax your closing-clearance/space (instead of the opening-clearance), so as to properly allow the opening-function to smoothly perform it's (complete) task, (along with properly shimmed v.stem-clearance).

Whilst I accept the principle, I can't rationalise doing it to the degree I would need to do so to remove this bind. The error seems just too large. To quanitify it specifically, at the bind point (just short of removing it - so that I can still feel the resistance point) I have an opener clearance of zero (and it needs to be 0.1mm min when used in anger) and closer clearance of less than 0.04mm, c-clip retainer getting tight - call it zero (with zero closer shims installed). At TDC this results in an opener clearance of 0.65mm and closer of 0.25mm. Given:

1. I've now been able to prove the rocker radiuses and offsets as being "correct"
2. There is no evidence of weld that might indicate attemped repair or dislocation of rocker spindle sockets

...this seems way too high a clearance to me (total 0.9mm at TDC and I need another 0.1mm on the opener yet). At the moment, with the information I have, I can't come to any conclusion other than I have tested two dud cams (one being mine). There may be a very good reason this other lightly used stock cam has been sitting on a shelf for the the last thirty - forty years.

I have 3 more B&W cams organised to test against Tuesday night, I guess I'll know more after that.

Kev

[DewCatTea-Bob]

" I'd suggest my scenarios presented regarding that particular point are both theoretically possible (and given my rockers have been repaired - practically possible), "

____ And I still continue to agree with that past (& quite sensible) speculation,, (although your provided rather low out-of-concert/incoherent desmo-clearance figures continue to lead me to blame only the D.cam[s]!).


" Whilst I accept the principle, I can't rationalise doing it to the degree I would need to do so to remove this bind. The error seems just too large. "
at the bind point (just short of removing it - so that I can still feel the resistance point) I have an opener clearance of zero
and closer clearance of less than 0.04mm, c-clip retainer getting tight - call it zero (with zero closer shims installed). At TDC this results in an opener clearance of 0.65mm and closer of 0.25mm. "

____ Alright then, I'm understanding that in order to reduce the opening/closing binding-conflict to it's bare-minimum (without ANY slop!), you've chosen to use shimming which provide clearances of .0256"/.65mm for the opening-rocker (@ TDC), and .0098"/.25mm for the closing-rocker (also @ TDC)...
Therefore to maintain the very same in-concert/coherent -(merely content-relationship) between the opening & closing rockers during the D.cam's most-close to binding-conflict area/spot,, (it then seems) you need to re-set your opening-clearance to .004"/.1mm, AND let your closing-clearance go-out to (at least) .039"/1.0mm AT TDC, thus then providing (with your particular D.cam) the correct 'desmo.clearances' (of .1mm open, & near-zer0 close)!


" this seems way too high a clearance to me (total 0.9mm at TDC and I need another 0.1mm on the opener yet). "

____ Many of the old DESMO-Single D.cams leave this amount of closer-slop, and it's really nothing to be concerned with, (since the stock included v.springs will have the valves held closed at TDC) !


" I can't come to any conclusion other than I have tested two dud cams (one being mine). "

____ Well the term "dud" may be justifiably applicable if ya really expect so very near PERFECTION (in constant closing-tolerances), however, in so far as those DESMO-cams performing that which the main-function is primarily intended to perform,, they're no-doubt actually not 'duds' !


" There may be a very good reason this other lightly used stock cam has been sitting on a shelf for the the last thirty - forty years. "

____ Yep, it's quite possible that some other desmo-perfectionist thought the same as you have.
(Too bad in a way.)


" I have 3 more B&W cams organised to test against Tuesday night, I guess I'll know more after that. "

____ That's really great ! _ Then we can get a closer idea of the percentage of D.cams which are less perfect than most.


Hopeful-Cheers,
-Bob

[machten]

Therefore to maintain the very same in-concert/coherent -(merely content-relationship) between the opening & closing rockers during the D.cam's most-close to binding-conflict area/spot,, (it then seems) you need to re-set your opening-clearance to .004"/.1mm, AND let your closing-clearance go-out to (at least) .039"/1.0mm AT TDC, thus then providing (with your particular D.cam) the correct 'desmo.clearances' (of .1mm open, & near-zer0 close)!

Yes, for this camshaft (ie the loan one i have - not mine) that is it in a nutshell.

____ Well the term "dud" may be justifiably applicable if ya really expect so very near PERFECTION (in constant closing-tolerances), however, in so far as those DESMO-cams performing that which the main-function is primarily intended to perform,, they're no-doubt actually not 'duds' !
____ Yep, it's quite possible that some other desmo-perfectionist thought the same as you have.
(Too bad in a way.)

Well, it's not just me being a perfectionist. My friend who owned his own ducati devel mechanic shop in the eighties and his own singles for many years inspected my head assembly for me as a fresh set of eyes and reckoned the clearances were the highest he's seen to remove a bind.

And the following from an email exhange with Ian Gowanloch last week...

"Single Desmo cams were often a little imperfect. Desmo cams in general have become consistently better with each new model produced.
It is necessary to leave a small clearance on the closing rocker, between .05 and .10mm. Set them to zero as is done on Pantahs and there is a problem binding. It is quite possible you are looking at a refinishing problem. Please add the thoughts above to the equation. A clearance of 0.20 to prevent bind is a little outside my 40 something years experience working on them".

(0.2mm closer and 0.48mm opener is one permutation that removes the bind for my cam - and I can move some of the opener clearance to the closer, but run out of shims to remove before I can get opener clearance anywhere near 0.1mm (note this is my cam, not the "other one" which is detailed in my previous post for which clearances required are in fact higher).

All of which prompts me to think it is at least worth a look at these other camshafts, particularly as they are easy for me to access.

Kev

[DewCatTea-Bob]

" Well, it's not just me being a perfectionist. "

____ I too, have always been somewhat of a perfectionist but in this case it's not up to ourselves, it's up to the D.cam...
When I stated "desmo-perfectionist", that was in regard to those who expect that 'zero-clearance' ought be maintained (by the D.cam) at all points during the closing-lobe's full-function. _ And if that were actually always the case, then we'd all more certainly be quite content,, but actually for no good reason ! ... As such closing-clearance consistency doesn't improve reliability or lead to extended engine life, (and if not maintained by a good-mechanic, could possibly lead to the reverse), and also, once the valve has been shut, it doesn't even help improve performance (to have the valve mechanically held shut) !!
So why should anyone care if the closing-rocker has a full-mm of slop at TDC (so long as it's clearance is set just-right where it most matters) ??


" My friend who owned his own ducati devel mechanic shop in the eighties and who inspected my head assembly for me as a fresh set of eyes reckoned the clearances were the highest he's seen to remove a bind. "

____ A "fresh set of eyes" -(meaning somebody who has not had their 'mind-set' contaminated with the existing thoughts of others), by a fellow DESMO-expert would've been a good thing to have gotten related input from ! _ But I'm concerned that such opportunity may have been lost, cuz it seems that it would've been irresistible to not fill him in on what the deal seems to be, before presenting the questioned-situation to him and merely only ASKING little more than: "what's up with this?" .
So I suspect that he didn't really get the chance to ponder all on his own that which he was presented with, before coming to any conclusion entirely of his very-own !
One clue to back-up that notion is the wording: "the highest he's seen to remove a bind" ... Cuz any good DESMO-mechanic will not have found himself in the situation of having to "remove" or deal with any kind of 'bind' ! - (Unless given such a created-problem by somebody-else.) _ As such issues no-doubt never came from the factory that way, (as far as I know) !
__ As for myself, I believe I only once came-across a D.head with near-about .040" of closing-slop at TDC -(which is actually irrelevant since TDC was not the point of minimum [non-binding] clearance). _ (Most I recall btw, were with far more acceptable run-outs.)


" And the following from an email exhange with Ian Gowanloch last week...
"Single Desmo cams were often a little imperfect. "

____ Right but, not all ! _ I would've guessed the ratio to be about 1 in 8 that would not satisfy a 'desmo-perfectionist'.
You've already found one, perhaps two,, so the chances are that the rest you check ought to check-out as average, if not near perfect.


" (0.2 is what is required for my cam - .
Additionally it requires an opener clearance of 0.48mm to remove the bind and provide.1mm opener clearance at the bind point. "

____ I understand this to mean (in other words) that with your-own 250-D.cam just snugly slip-sliding through & past the point closest to bind-up, that currently happens to coincide with the closing-clearance set at .008/.2mm and the opening-clearance set at .015"/.38 (both measured at TDC).
__ So if that's correctly understood, then in order to get the proper desmo-clearance settings, the closing clearance must then be increased to .023"/.58mm at TDC, in order to end-up with zer0 closing-clearance at the point closet to bind-up, (although I've agreed that the closing-clearance should be a bit greater than only zer0).


" I can move some of the opener clearance to the closer, but run out of shims to remove before I can get opener clearance anywhere near 0.1mm "

____ As I've mentioned before, the C-clip valve-keeper collet-shim comes in different sizes from (I think) 2mm to 5mm thick,, so chances are the one you're using is not the thinnest. _ What thickness is it ?


" All of which prompts me to think it is at least worth a look at these other camshafts, particularly as they are easy for me to access. "

____ It's a fortunate circumstance that you've been so concerned with this DESMO-clearance issue, cuz now this could be the first time anyone has ever checked-into this particular matter with so many D.cams all at one session and shared the info with more than just those directly involved !!
I'm sure all DESMO-Dukers will appreciate your unique & helpful contribution !


Hopeful-Cheers,
-Bob

[machten]

What has now become a bit of an aside, but may be of interest to some people

...originally from page 2 of this thread and concerning desmo valve dimensions...

I said..

Never having worked on a desmo single before I'm not familiar with the specific dimensions of the valves. I got a friend to take my valves with him to compare against his 450 desmo valves. I don't think they are original either, but he reported back that my inlet valve overall length was exactly the same as his, but the distance from the valve head to the collet recess is 2mm greater than on my valve.

From DCT Bob...

____ I've never before happened to note any such variance, although that's not to say I've never encountered such (unknowingly).
Unfortunately, I've never been able to accumulate DESMO-head parts, so as to have them all laying-about for possible comparison. _ Otherwise, I would've already been quite aware of the alternate location for the C-clips on the valve-stems.
There actually shouldn't be any alternate locations in the valve-stem (for the keeper-clips) due to whether the DESMO-head is made for a 250/350 or a 450 !
__ That your 250's valves are set-up with the shorter length from valve-head to keeper-slot, goes hand&hand with your low-lift D.cam ! _ And well correlates with my notion that older (pre-450) DESMO-models had a milder-lift D-cam.

...and...

__ You had mentioned that you had a valve which had it's C-clip/v.keeper-slot spaced 2mm further-down from the valve-stem's tip-end, (compared to a 450's)... Was that the exact-same case with both your valves? ...

To which I responded...

Comparing the inlet valves: same length overall, with my c-clip recess 2mm closer to the valve head
Comparing the exhaust valves: My valve was 2mm longer overall, with my c-clip recess 1mm further from the valve head (the head being the the valve seat end)

I have now received some new valves for comparison from Ian Gowanloch. Photos below...
Both valves from Ian came packaged and sealed like so...

Valve packaging.jpg

Comparing my inlet valve with the one from Ian....new one at top...
Length of new I valve = 86mm. Distance top of valve to centre collet recess = 4.0mm
Length of existing I valve = 85mm. Distance top of valve to centre collet recess = 5.0mm

Inlet Valves.jpg

Comparing my exhaust valve with the one from Ian....new one at top...
Length of new E valve = 85mm. Distance top of valve to centre collet recess = 5.0mm
Length of existing E valve = 85mm. Distance top of valve to centre collet recess = 5.0mm
(i.e. Identical)

Exhaust Valves.jpg

So it seems there are a few different permutations of available valves, which is useful to know when you have limited spring retainer and collet collar sizes and shims!!!

Kev