Alfa Romeo Forums banner

1 - 20 of 21 Posts

·
Registered
Joined
·
1,233 Posts
Discussion Starter #1
After dragging to do this for many months, I took the valve covers off to do some inspection and maintenance ...

I noticed some green coolant next to a head stud bolt. I can't think of where this coolant can come from except the likelihood that it has come up from the head stud :confused:!

I took the head stud nut off and it broke loose at about 80 ft-lbs (I guess this could be a little bit loose for the 24V head!). I'm planning to put a couple rubber o-rings under the nut to seal it. In the picture below, one o-ring is shown between the steel washer and the head nut. The steel washer is thicker than the rubber o-ring so one will not be thick enough.

What do you guys think?
 

Attachments

·
Premium Member
Joined
·
15,849 Posts
Just a thought. I can't imagine what will happen to the rubber o-rings when compressed under a washer and stud nut torqued to somewhere around 80 lb-ft of torque or whatever. Unless there is a bit of space for the rings under the nut inside of the washer, the poor things will get squished beyond recognition and probably not seal anyway in such a distressed state. But you could try it, but it is only a bandaid of sorts.

I recommend however that you should try retorqueing the heads first to see if that stops the leak. This is a problem with the 12v engines, as after many thousands of miles down the road, the heads loosen up just a little, and the gaskets can start to leak cooolant, most of the time fixed by retorqueing. Otherwise you will have to replace the head gaskets. I don't think the bandaid approach will get you far. The only other time I've seen this is when a pin hole casting flaw opened up in a head, and squirted a very thin line of coolant into an exhaust port. Had to have the head welded up, for a price of course.

You should also change the oil, as that puddle is significant enough to worry about having leaked coolant contaminating the oil over time. I'm surprised that you don't have chocolate mousse on the inside of the cam cover.
 

·
Administrator
Joined
·
12,536 Posts
The most important thing a head stud has to do is correctly hang on to the head ... a rubber grommet is hardly going to help that.

Is the surface under the head stud washer nice and flat? ... then I'd retorque all stud properly.
Pete
 

·
Premium Member
Joined
·
2,811 Posts
Definitely re-torque.
270 degrees, IIRC.
 

·
Registered
Joined
·
5,466 Posts
I agree with others

I don;t think rubber will help longer term BUT some gasket sealant or other thread sealant might!

bob
 

·
Registered
Joined
·
1,233 Posts
Discussion Starter #6
Thanks guys! Yup, totally agree with all your comments!

A rubber o-ring might get crushed to nothing and what I afraid of is pieces from the o-ring might come off and cause trouble! However, it appears that there is some room there to accommodate a thin o-ring between the head, stud and washer!

So, I first put a few dab of RTV silicone sealant around the bottom of the thread. Slip a R07 (3/8" ID, 1/2" OD) o-ring down to the bottom. Put the steel washer in then nut, torqued to 30ft-lbs, back off the nut and remove (pic below, o-ring under washer already). I'm going to let the RTV cure a little, then I'll apply some oil and re-torque the nut to 18'#+240º !
 

Attachments

·
Registered
Joined
·
5,466 Posts
nice camera!

photos look good ! Agree with the course of action although a piece of O ring prolly will get ground up pretty easily with no damage!
 

·
Premium Member
Joined
·
26,992 Posts
I wish I could see coolant inside head on Sara's 164B then I would know where it is going.

Not getting in oil but mayo keeps coming back under valve covers.

As for your idea of trying to seal head stud. See if you can find a large statoseal metal washer with neoprene molded into it and try that under steel washer.
 

·
Registered
Joined
·
1,233 Posts
Discussion Starter #9
I either need more arm power or a cheater bar!

Torquing the nut down ... with my torque wrench set at 130ft-lb, it clicks at about 220º :rolleyes: ... good enough? I don't think using 18ft-lb to set the initial starting point is that accurate either! ... ok, I'm calling a time out for the day, ... time to go out to re-fuel (my stomach that is ;)) !

So, do they really use stronger and/or larger head studs on the 24V? :confused:
 

·
Premium Member
Joined
·
26,992 Posts
Studs/nuts same size.
 

·
Premium Member
Joined
·
15,849 Posts
Could be a difference in stud/nut thread shear material strength, as they are different part numbers between 91-93 and 94-95 studs/nuts. Also could be difference in length of thread engagement. This would determine the total engaged thread load capability at either end.

Also, it does look a little from how they appear to be drawn in the eper that the shank dia might vary between models. Since they are all steel, the modulus of elasticity won't change, thus...

elongation of a length of rod = PL/AE, or... applied axial load times effective length all divided by the cross sectional area times the modulus of elasticity.

Interesting that 91-93 use two different stud numbers, whereas 94-95 use one. The nuts between the 91-93 and 94-95 are different as well.
 

·
Premium Member
Joined
·
9,776 Posts
Torquing the nut down ... with my torque wrench set at 130ft-lb, it clicks at about 220º :rolleyes: ... good enough? I don't think using 18ft-lb to set the initial starting point is that accurate either! ... ok, I'm calling a time out for the day, ... time to go out to re-fuel (my stomach that is ;)) !

So, do they really use stronger and/or larger head studs on the 24V? :confused:
Make certain that the nut threads are clean clean clean. Any thing on the threads can give you a false torque. Just some advise.
 

·
Premium Member
Joined
·
26,992 Posts
Two different lengths on 12v engines.
 

·
Registered
Joined
·
1,233 Posts
Discussion Starter #14 (Edited)
Post-torque report

I have finished re-torquing all the head nuts a couple days ago. I have found that I could break loose all the nuts at about 80 ft-lb. So, I guess somebody was there before and re-torqued the nuts up using the 12V spec :mad:!

To try to find out "torque vs angle" relationship, I wrote down the torque angles at various torque settings. I set the torque wrench at 60(ft-lb), 70, 80, ... and so on. The final settings was at around 110(ft-lb), 120 or 130 (for a couple of nuts). The final torque angles for the nuts were between 210-230 degrees. I stopped short of the 240 degrees because I think the 240 as specified is too high and too close to some "yield point". When I was torquing down a couple of nuts, I think I actually had a gnarly feeling that something was starting to give at the end (could be the threads) :eek:!

I'm sure 240 was ok when it was done in the factory (with robot arms and much more precise tools) and on brand new parts etc. I'm sure the Alfa engineers had done the calculations and tested the parts on the load-test machine! However, in the field (where we don't have robot torque arms :p), this spec is just too close to (if not already at) the safe limit and leave very little (or no) margin for errors (if we want to call that)!

Below is a "Torques vs. Angles" graph from my data. The data points are torque angle at the various torque settings (e.g. 60,70,80,...). For each set (e.g. @60ft-lb), an average value is computed. A straight line is then fitted using the averages. Extending the line to the x-axis, it intersects at just about 18.4 ft-lb (bingo :))! At +240 degrees, the trend line shows the torque would be at about 140 ft-lb :eek:!
 

Attachments

·
Registered
Joined
·
1,233 Posts
Discussion Starter #16 (Edited)
looks fantastic -- what happens if you include the outliers?
I gather you want to see the data distribution! I have updated the chart (in the previous post) to include recorded data points. Now, you can see the ugliness and how inaccurate my torquing technique was :D! When I was recording the data, some data appeared to be all over the place! Let's look at the data @100ft-lb, the lowest and highest torque angles was -18% and +21% away from the final computed average :eek:! However, most of the other data are generally within 10% ... and, with a larger set of data they converged to some nice averages (as indicated by the R**2 number of the line fit)!
 

·
Registered
Joined
·
5,466 Posts
I don't think thats torqueing tech nique

Bob I think there is inherent variability in the materials ; I suspect only a small amount is 'user error'. Most of this has to do with (in my opinion) the varying yield of the studs, the threads themselves, and whatever you are compressing. My guess is that you would end up with similar error bars with 'automated' torque technique
 

·
Registered
Joined
·
8,982 Posts
try this. it works well on the 2 liters.. get a pane of glass, and get some fine emery cloth, put the emery cloth on the glass then add a few drops of oil, then get the washers that are under the head bolt, and 'lap ' them in so you get a perfect gloss pattern on 'both' sides, now do this with the head bolts, but only on the washer side, this should seal it much better... every time i do a 'head' job on my car or my feinds i do this , allways works
 

·
Registered
Joined
·
1,233 Posts
Discussion Starter #19 (Edited)
> I don't think thats torqueing technique
Haha, I was sort of joking! I knew there's going to be "quite a bit" of variations. If I loosen and re-torque the same nut, I'm not surprise that I'll get numbers that is, say, +/-10% off!

Nothing (except perhaps at the very microscopic level of the material surface) should be yielding (yet!) at the lower range of the torques. The greatest contributor to the variation is material friction! Material surface is not perfect - to name a couples possibilities: dirt and deformations left from previous torquing. When the nut turns, the threads have to slide over each other. Friction is estimated by coef. of friction (some fraction of the tensile load in the perpendicular direction) and there is a range (static to dynamic/sliding).

Torque angle is a better specification to achieve a certain load on the studs. By tightening the nut down through a certain angle, it shorten the stud by a certain ;) proportional amount. This shortening displacement will add a certain :cool: compression load (clamping force) to the head and tensile load to the stud rod. Anyway, I'm certain :p that this is a better method (well, for most situations :confused:) because this completely ignores friction!
 

·
Registered
Joined
·
1,233 Posts
Discussion Starter #20
try this. it works well on the 2 liters.. get a pane of glass, and get some fine emery cloth, put the emery cloth on the glass then add a few drops of oil, then get the washers that are under the head bolt, and 'lap ' them in so you get a perfect gloss pattern on 'both' sides, now do this with the head bolts, but only on the washer side, this should seal it much better... every time i do a 'head' job on my car or my feinds i do this , allways works
Thanks for your comments! I gather this is a suggestion on how to prepare the head bolt, washer and stud before re-torquing. Yes, cleaning them (polishing them in your case? :)) will certainly give a narrower range of outcome with less variations. However, the smoother surface probably will shift the resulting load curve/line higher because the torque wrench will likely click later resulting in larger angular displacements (torque angles)!
 
1 - 20 of 21 Posts
Top