Head gasket torque

[belchfire]

I know that there are a million postings about this subject, but unfortunately, also a million answers too. I have a 1.6 turbo running about 15# of boost. Gasket started leaking and water was getting into the oil. I pulled it down and cleaned things up. The third and fourth bolts were rusted pretty bad where the gasket failed. This also happened to be right where the big oil return hole is (hyd head). Cleaned everything up and checked the head. I couldn't even get a .0015 under the straight edge so it's pretty well dead nuts. There seems to be a slight corrosion around the problem bolt area, so I put some permatex on the gasket to ensure sealing. Supposedly, the trick set-up is to run the metal AAZ gasket,and maybe the PD 150 (arl) head bolts but none were to be had. I ended up with a Victor-Rentz gasket (1notch stock 1.6) and 12 mm bolts that are the same for the 1.6, AAZ & TDI: so they said. The original torque procedure says to go to 55 ft/lb then 1/4 turn more. This one says 30 Newton/meters (22ft/lb?) then 60 N/M (44ft/lb?) then 90* then another 90*. Then warm it up and ANOTHER 90*. THEN drive it 1000km and tighten ANOTHER 90*! Good God, That's 1 full turn total! When I did the second 90* I almost had to use a cheater bar. Do they relax enough so that another 1/4 turn won't snap them off? Also, on a couple of them, while turning,  seemed to give. Just a little and then they came back. Was this the cylinder ring "seating", or did I pass the yield point? It "feels" like they're in the 100 to 120 ft/lb range.

[Quantum TD]

If you're running the fiber gasket (not the MLS one), then follow these procedures:

Stage 1: 40 nM

Stage 2: 60 nM

Stage 3: 180 degrees. You can do it in 90 degree increments

Stage 4: Warm up car to 50C. Retorgue an additional 90 degrees

Stage 5: With the newer gaskets, this is "optional". If you see seepage, you might want to do it. Also, if you're running higher than average boost, you'll want to do this.


The "give" you felt was likely the bolts stretching. If they continue to stretch, I'd say don't torque them anymore.

This is a bit of a gray area for VW. I'm not sure why they couldn't just give a simple sequence with a final torque value like just about every other car manufacturer. I doubt that they followed their own procedures when the built the motors in the first place.

While there are a lot of factors involved in the final torque values, namely the type of lubricant used on the bolts (WD-40 is best), tensile strength of the bolts (they should all be the same, but with bolts made in China, who knows), etc., I've found that the finished result hangs in the 100-120 ft-lbs range. Sometimes a bit higher, but rarely lower than about 105. After the complete cold sequence, I usually go over the lot and make sure my wrench clicks at about 105 ft-lbs. If not, I tighten them up to that point. After the reheat, I set the wrench to 115 and do the final 1/4 turn. If they click before then, then fine, if not, I push them up to 115. That's just me being a bit anal, but it works for me. Since you've already done the cold sequence, I'd set the wrench to 115 for the reheat and see if you get it to click. If not, go until it does. If you feel any bolt give, then just do the 1/4 turn.

If you used anything other than simple WD-40 on the bolts (or whatever pre-lube came on them), then you may have already exceeded the tensile strength for stretch.

[belchfire]

   I didn't do  WD40 but 30W instead. I don't think that 44 ft/lb is too critical for the initial torque. A friend suggested that I may have bottomed out and that is a real concern. I cleaned & tapped & blew the living crap out of the holes so that they would be clean & empty. There is a possibility of hydraulic lock but it's kind of late now. I guess I'll take my chances and hope she don't blow. Almost got her back running. When I do, I'll warm it up then check and re-torque at your 115 ft/lb. If nothing snaps, then I'll call it good.

[Quantum TD]

You should be OK with 30w. You just don't want to use anything that will drastically reduce surface tension. My mistake was to use synthetic  . I had to completely redo the job when 1/2 of the bolts started to give on the first 90 degree torque.

You are probably right on the initial pass. Nothing to worry about.

As for bottoming out, I doubt it. The bolts only project past the head about 1-1.25 inches. There's about 2 inches of thread depth.

On the oiling: if you use a thicker oil, just be sure to rest the bolts on some clean, fresh newspaper so that the oil runs off and gets soaked into the paper.

[745 turbogreasel]

I have had some  cheap aftermarket bolts break before reaching spec.

[vanagonturbo]

Quantum TD is on the money. Not sure about using synthetic causing a problem. I have used synthetic, wd-40, 30W, 20w50, 10w30 and pretty much anything slippery and have never had a problem.

[belchfire]

Re-torqued after getting to temp (then cooled) to 115 ft/lb. 3 turned slightly the others didn't. so far, so good.

[745 turbogreasel]

You are measuring the wrong thing, those bolts are designed to a specific amount of stretch, easily predicted by # of turns, not a  turning torque which can be affected by any number of factors.  It will be fine for awhile either way.

[belchfire]

I understand the theory of the stretch bolt. I had a race pinto and used a dial indicator to stretch the con-rod bolts .006" . A little hard to do in this application. It just seemed odd that the earlier spec was only 1/4 turn additional and the new was a full turn more. The old set torque was 55 ft/lb and the new-44 ft/lb. I seems that an additional 3/4 turn would impart more than 10 ft/lb of force. It might make more sense if the bolt bottomed out and then there was the 1/4 turn that would produce a consistent clamping force but with the variables in thread lubricity and bolt-to-washer-to-head friction, the actual force would vary. By going with a real number, it seems that the variables would be compensated for and that there would be an even clamping force on each bolt. If done correctly, the bolts would still retain their stretch characteristics and be able to adjust for the differences in thermal expansion. I guess that with studs, you torque the beejezzus out of them and hope that's good enough.

[745 turbogreasel]

The head bolt design and procedure were changed to correct a recurring problem, that didn't do it, so they went to the metal gasket.
If the bolt bottomed, it would end in disaster.
using bolt stretch does eliminate those variables.
Studs don't stretch appreciably, and are typically not torqued as tight.  Without stretch, clamp force is easily calculated as a function of turns x thread pitch.  More machining precision is required, and greater clamp force is available if needed at a lower fastener torque.

[belchfire]

Yeah, forget I mentioned the bottom out thing. I can see the revision as the previous procedure obviously didn't work. From what I've read, the metal gasket is the trick set-up but a 1 notch was not available so this one's a composite. Only 100 miles, but running good so far.