Valve adjustment as per factory manual

[schrader7032]

Kurt......... the head fasteners couldn't be in any better environment and if you toy with some fasteners..... you will see it is possible to torque from 49 to 50

and even 49.5 to 50

I can show you published procedures which do actually specify backing off head bolts prior to retorquing which leaves it reasonable to assume the manufacturers are capable of specifying exact and concise procedure

bmw sure is and they don't on the airheads

E.I. -

You posted this on a different thread...not sure what 49 to 50 is?? Thought I'd bring it over here...

I talked to a BMW trained mechanic who's been working on Airheads since the '60s, /2 and later as well as the newer models. He doesn't know of a specific reference to backing off the nut, but he said this is what they were taught in school. Back it off a couple of foot-pounds, one nut at a time, and recheck the torque. It doesn't affect the head gasket...unless you were to loosen all fasteners at the same time.

Having said that, he mentioned a couple of things, which I believe makes sense. First, if this is your bike and you were the last one to set the torque, then it's reasonable to just put the wrench on the nut and show that the torque is at least the specified value. If the nut was loose, then it will get tightened to the correct value. If the nut is already at the correct torque, there's little reason to think that it got tighter. In that case, leave things alone.

Second, if this is an unknown bike and you don't know who torqued the heads previously, then it makes sense to establish the baseline torque. And the only way to do that is to individually release a few ft-lbs of torque and reset the value. After that, revert to the first point.

That seems like a reasonable approach to me. BMW didn't necessarily get things right in all situations, either on the design of the bike or in their manuals. There are service bulletins which indicate there were errors in their manuals, specifically in terms conversion from SAE to metric units. I think we need to do what makes sense to each of us...if we got it wrong, the bike will let us know, hopefully not at some inopportune moment!

Kurt in S.A.

[EuroIron]

back off a couple foot pounds............... is that exactly 2?

now how do you propose to do that if you don't believe a head nut can be torqued from 49 ft/lbs to 50 ft/lbs

[schrader7032]

Come on, Euro Iron, you know I didn't mean EXACTLY 2 ft-lbs. One of the definitions in Webster's for "a couple" is an indefinite small number. It's already been stated that backing off to the tune or 1/8 or 1/4 of a turn is all that's necessary.

You said "...if you don't believe a head nut can be torqued from 49 ft/lbs to 50 ft/lbs". I didn't say anything like that. I don't know what you're talking about. And what is ft/lbs? Is that feet per pounds? What unit of measurement is that? See, I can read things too literal, too. Gotcha!!

I don't know what engine uses 49 to 50 ft-lbs (assuming that's what you meant) on head bolts, certainly not a /2, /5, etc. Still, the idea is to lessen the torque to whatever feels comfortable, a small percentage of the required torque, so it can be retorqued with a running motion.

That's it...I think a lot of information has been provided to the original poster, if he's even still reading. You can read the manuals word-for-word, you can read things on the web, you can talk to people who have been doing this work for years. Each person should then take away what works for them. Then they go riding. Hopefully that's the case here.

Later...Kurt in S.A.

[EuroIron]

so if a head nut is actually torqued to 24 ft lbs

can it be directly torqued to 25 ft lbs without having to loosen it.......... a couple ft lbs or even a 1/4 turn?

based on the incorrect notions of many, any nut or bolt would have to be backed off AT LEAST 6.24 ft lbs, in order to torque if to 25 ft lbs with mathematical certainty

these guys are telling you 25% stiction

why 6.24 and not 6.25?

if 6.25 that would mean the fastener was at spec and not needing any further attention


my only point is that BMW's procedure is very specific and it does not specify backing off the head fasteners to "retorque" or check them for tightness

have you thought any more about your theory on valves bedding in/seating as opposed to why the head nuts and cylinder flange nuts are actually part of a valve adjustment?

once you understand that, then it makes it easier to understand why loosening is invalid and counterproductive......... detrimental even

so humor me and kick that about.......let's not worry so much about to back off or not to back off the fasteners for now

[schrader7032]

so if a head nut is actually torqued to 24 ft lbs

can it be directly torqued to 25 ft lbs without having to loosen it.......... a couple ft lbs or even a 1/4 turn?

based on the incorrect notions of many, any nut or bolt would have to be backed off AT LEAST 6.24 ft lbs, in order to torque if to 25 ft lbs with mathematical certainty

these guys are telling you 25% stiction

why 6.24 and not 6.25?

if 6.25 that would mean the fastener was at spec and not needing any further attention

I think you're getting way, way too numerical here...we don't have that much information to go on. There is too much error and variables in our torquing process to be that numerical. One of my points is that the nut must be moving in order to get a true torque reading. If you can do that by backing off to 24 ft-lbs and then go to 25 ft-lbs, fine. Most likely the increase in 1 ft-lb would happen so quickly that you wouldn't have time to focus on the torque wrench before you end up at or beyond the ending value.


my only point is that BMW's procedure is very specific and it does not specify backing off the head fasteners to "retorque" or check them for tightness

To me, BMW's statement to "check for tightness to 25 ft-lbs" is NOT very specific. That's why I asked someone who was trained by BMW where they do cover the specifics on each mechanical task.

have you thought any more about your theory on valves bedding in/seating as opposed to why the head nuts and cylinder flange nuts are actually part of a valve adjustment?

once you understand that, then it makes it easier to understand why loosening is invalid and counterproductive......... detrimental even

so humor me and kick that about.......let's not worry so much about to back off or not to back off the fasteners for now

I don't recall stating a theory...I might have mixed up some words regarding things bedding in. I understand that as part of a fresh engine build, things like the head gasket are going to compress over time. Things wear in to some degree. Thus, it's necessary to reestablish the proper tension in the head bolts in order that the mating surfaces remain firmly in contact. That has to be done in order to be able to set the valves properly. The dimensional relationship between the head, cylinder, and engine must be the same otherwise setting the valves won't be accurate.

From my vantage point, the loosening approach is not invalid, counterproductive, or detrimental. OK, the loosening does create a tension load cycle in the bolt that wouldn't normally be there. But the bolt is steel, it doesn't happen that much in the life of the bolt, and the delta tension (loosened vs torqued) is so small that the fatigue of the bolt is virtually unaffected. We're operating in the elastic region of the bolt's material properties. If we were near the plasticity region, then that would be a different story. I don't know the specific grade of steel, but I'm sure basic fatigue properties of the bolt, for the small stress cycle being introduced would suggest that this process could be repeated thousands and thousands of times before failure would occur. I should live that long!!

Kurt in S.A.

[Darryl.Richman]

4. Working in the reverse of the tightening sequence, slacken the cylinder head retaining nuts by 1/4 turn each; do not slacken them further or the rocker arms may require realighning. Working in the order of the tightening sequence, tighten the nuts to the appropriate torque setting.

While it is true that the BMW manuals -- shop and owners -- do not specify a procedure for applying torque, it's not beyond the realm of possibility that the shop manuals assume that the mechanic already knows how to perform the operation and merely calls it out to refresh his memory; and that the owners manual is not written to encourage the owner to do the work, but to understand what his local dealer is doing for the money.

If you check http://www.boltscience.com/pages/tighten.htm, you'll see that the actual torque distributed to clamping force (thread extension) can be as little as 12% of the total applied torque; the rest goes into overcoming friction on the threads and the nut face.

Since static friction has a different, and generally larger value than sliding friction, one could be excused for coming to the conclusion that applying a certain torque to a fixed nut may not guarantee that the designed clamping force is actually being applied.

The latest model BMWs don't rely on a fixed torque spec, but rather ask for continuing to turn the nut through a particular angle after reaching a specified torque. This is supposed to be much more accurate than what is possible with a simple torque spec alone. Such a specification avoids a lot of the problems related to friction.

[EuroIron]

I am BMW factory trained

GM and Chrysler too

Certified Master Tech

anyhow, you still have not figured out why this checking for tightness, which is quite concise, is called for every 4K

but you are getting close......... so very close in fact I'll ask you to consider what the studs thread into



however, despite how this may appear to you........ this is strictly academic and nothing personal against anyone

[Darryl.Richman]

I am BMW factory trained

GM and Chrysler too

Certified Master Tech

anyhow, you still have not figured out why this checking for tightness, which is quite concise, is called for every 4K

but you are getting close......... so very close in fact I'll ask you to consider what the studs thread into
Well, depending on the model, we are either talking about screws that thread into cast iron cylinders, or through-studs that are mounted to the cast aluminum crankcase.

In the former situation, I can't see how multiple cycles of loosening and tightening will have much effect.

The "airhead" bikes do have a certain reputation for losing the threads in the crankcase; this has been a tidy little business for TimeCert and Helicoil.

(A dealer's mechanic pulled a stud out of my R1100RS once, as well.)

however, despite how this may appear to you........ this is strictly academic and nothing personal against anyone

I can readily appreciate the Socratic approach to learning.

But let me ask you a question: is the torque spec in the manual to be used when initially installing a head? If it is, doesn't simple physics say that applying the torque to a loose nut possibly result in higher clamping force than would be the case if the nut were already stopped and fixed?

[EuroIron]

does it matter?

if the fastener has been torqued according to the specs, with a good torque wrench, and you apply the correct torque to it every 4000 miles, and it doesn't move...... all is well

if you carte blanche back off prior

you don't know all is well, or even that the fastener was correctly torqued, even if after loosening it...... the fastener has to be rotated past it's previous resting position to obtain the correct torque

why? because the fasteners nor the threaded items are made of diamond

if you replaced the BMW studs and nuts with ARP fasteners, would you follow their procedures?

they will tell you not to back off......... just hit them with the torque wrench

[EuroIron]

Here is the bottom line

if the procedure was to loosen prior to checking for tightness

BWM, and others, are quite capable of telling you so and would do so in a very clear and concise manner

their procedures do not take into account that somebody unqualified has attempted to service the machine prior

what would BMW tell you if you said, "well I don't know if it was originally and /or subsequently correctly torqued to the published spec" ???

have your torque wrench calibrated then install and new head and/or base gasket(s) and follow their procedures

that is the BMW factory I worked for and if a technician put in writing or verbally admitted to not following specific procedure, adding to procedure, etc........ and there was a problem........ that same technician would eat it

overkill or even absolutely necessary?

probably not but academically and technically correct

unverified but allegedly Suzuki does actually specify backing off head bolts, one at a time, prior to torquing during valve checking procedures......... on one of their early 80's air cooled 550 or maybe 600 cc inlines

yet Suzuki do not specify this procedure for all their engines

look at the 650 Savage.......... specifies removing ALL cylinder head nuts in a criss cross pattern, apply oil, then reinstall and torque to spec in a criss cross pattern

this is specified to be performed prior to valve clearance check every 4000 miles, and keep in mind some of these engine/frame combos require the removal of the engine to removed the cam cover as necessary to access these nuts.......

wonder how many of them have been correctly service even by dealerships

I'll think about posting some more specs