While doing the the second 90 degree turn on my stroker heads I felt a slight give for a few degrees for a couple of the long bolts and then gave the feedback that was normal for the last 45 degree turn. Memory is hazy. On the PS head the top middle bolt and on the DS bottom left are the ones I remember. Could have been one other.

I did use 220 sandpaper to surface the head facing side of the washers in cross hatch pattern. Bolt threads were cleaned with brass brush and followed with brake cleaner and coated in oil prior to installation. To add to this thread the head gaskets are Cometic, though that shouldn't really matter.

Heads are off an '89 that was supercharged, filled a cylinder with intercooler fluid and hydrolocked. Most of the bolts had their anodizing.

I may undo all the work and redo it and see if the results are the same.

 

Kevin,

Your attention to detail is usually superlative so please take my comments for what they are and apologies if you have all this covered already:
1. You are using heads from an 89 so presumably they are the ones with the modified [longer] bolt length and you are using the appropriate length bolts [and not studs].
2. Lubricating the threaded part of the bolt is usually not a good idea. The tightening procedure is designed to get the bolts to a specific level of tension- the first stage with a torque wrench is basically to take up the slack and little more. The next stages apply a fixed degree of rotation on the bolts to achieve the desired tension that is usually in the range of 70% to 90% of yield thus allowing a safety factor as it were. Torque wrenches are inferential measures of bolt stress but they can be corrupted by inappropriate lubrication of the threads as deeper engagement will be achieved with lubrication. Whether or not this can create a real world issue remains to be seen- there is margin for error built in to the system.
3. Assuming bolts have been used after initial torquing there should only be two stages of tensioning using the 90 degree increments. The procedure for studs is different in that three stages of controlled rotation are specified- one presumes you have not got the two different procedures confused. By design the bolts should be sacrificial if they have been inadvertently over tensioned [one hopes].
4. As I am aware the gaskets are designed for a single installation and if you remove the gaskets the manufacturer will advise they be binned so removal should be avoided unless you have clear doubts about one of the points raised herein..
5. I also noted that you stated the heads came from a hydro-locked engine. It should be rather obvious that when such happens there is clear potential for the head bolts to have been over stressed local to the locked cylinder and thus potential [but not obviously apparent] damage to the heads around the bolts. Sadly such damage may only become apparent when the bolts are torqued and the heads are under compression from the bolts. Your description of concern at two different locations suggests this is not an issue but...?
6. Did you use new head bolts or [possibly] ones from the hydro locked engine?
7. Were the heads skimmed at all/checked for flatness?

Trust the above thoughts vaguely helpful. to stimulate discussion

Hopefully our engine building experts can give you more specific guidance.

 

A lot there to digest. First, I think you overestimate my attention to detail.

Heads and bolt length are matched - heads have the boss on them. Heads were lightly skimmed to remove surface corrosion. Heads and bolts came from the hydrolocked 89.
To the best of my recollection I've always coated the head bolts with oil prior to installation. And there have been no failures and the cars are still in my possession.
Cometic head gaskets are layered - in this case there are 5 layers. Installation steps are the same, but you do them twice and unbolt the head in between installations. This fully compresses the head gasket.
I'll circle back to Cometic on reuse.
I did not apply any lubricant between the bolt head and the washer as Greg does.

Thanks for your input.

 

Kevin,

I doubt the oil on the threads would make any serious difference but as I am aware it is not the "done thing" as it were. The Cometic gasket procedure is interesting- tends to fly in the face of all I know from using many thousands of composite gaskets in my oil industry career but completely open minded on that one. I have always used the argument that gaskets only work if they are elastic and if they are, should be able to be reused as long as they are not plastically deformed but that is another matter.

If and when a bolt is plastically deforming it is usually quite distinct to anyone with a sense of feel. Basically you are turning the wrench and the resistance has flat lined or is collapsing when near to failure. Ironically the only bolt failures I have ever had were both using a torque wrench and came about because I had the wrong torque figure. I came close when reinstalling the lower front A arms- instead of checking the torque values I used the number in my head [85 ft lbs] which was correct for the bolt grade at the rear set but not for the front set- fortunately I could something was not quite right at abut 70 ft lbs and was relieved when the new bolts I put in torqued to 50 ft lbs- one can never be too careful!.

If you do remove the bolts perhaps you can check the free lengths to ensure they are not plastically deformed as they should be dumped if they are. My expectation on a hydrolocked motor is that the rods will bend but maybe such can deform the bolts?

It wiil be interesting to see if any of the team have more constructive inputs to make for you.

 

I would replace with new bolts. The old bolts are probably what's at fault. 220 grit should be rough enough, but I like to use a grease pen as insurance.

 

$30 a piece for the long ones does temper one's leap toward buying all new bolts.

 

I go in dry on the threads, clean the head side surface for the washer, sand the washer and use ARP Ultra Torque lube on the upper face on the washer. Works well for me.

 

The following are my thoughts on head bolts....not head studs.

I assume that any piece of hardware that has been sitting in a "stretched mode" for 30 years, has been through thousands of heat cycles (getting stretched and relaxed multiple times) is no longer what it was originally. And if they are discolored....they fail the very first test. (Visually different is pretty obviously no longer the same as they were originally.) Original 25-30 year old cylinder head bolts get replaced, 100% of the time.

Head bolts that have been replaced before and are removed to freshen an engine (race engines/high output engine/experimental engines) can be reused. However, they are carefully monitored for actual torque, as they are angle torqued. We just freshened an engine with 35,000 miles on it over the period of about 8 years. I replaced 2 or 3 bolts that did not have the torque I expected after the first 90 degrees of rotation. (One of those bolts was discolored and did not pass the first test, but I was interested to see if it was actually "altered" other than just the change in color.....it was.) The remainder were still visually the same. I angle torque with a torque wrench that reads the amount of foot pounds required to get to that angle, once you are done pulling on the wrench. The "first 90" degrees of rotation must have all torque values within 10% for me to continue.

If a cylinder head has to be subsequently removed, I replace the gasket. The price of the gasket, compared to the amount of hours it takes to go back and replace a gasket that leaks is moot. (I don't use Cometic gaskets. LIke Kevin, I would call them and see what they say.)

I lightly oil the threads, sand the washers, clean the head surface just before assembly (even though the heads are almost always cleaned from the "restoration process"), and use the best thick never seize I can find between the washer and the bolt. Each bolt is marked in relationship to the head and if it slips, I stop and "repair" that issue, before continuing.

 

Greg, so what I described is a 'slip'? I don't mark the washers prior to installation. I think I read that you do.

When you say you repair, do you just replace the bolt and continue with the tightening process or do you back it all out and start over? I know the two bolts that 'slipped'.

I mark my bolts with paint pen for each time I do a 90dg turn. I missed the anti seize under the bolt head, though.

Thanks for chiming in! I was waiting on you to return and join the conversation.

 

If you suddenly feel a "give" in the amount of torque you are applying, that is a slip.

"Yielding" occurs slower....it's a different feeling. Yielding can easily been "seen" with a torque wrench like I use. For example, if you have 75 ft lbs. at 1 x 90 degrees and the torque drops to 60 ft lbs at 2 x 90 degrees, the bolt has "yielded" and will be permanently stretched. (The bolts are designed to stretch, but they need to be able to return to their original length. If they are unable to return to their original length, they have yielded and are junk.)

I remove bolts that have "slipped" all the time...it is very difficult to not have this occur, on "used" engines. I stop the process, pull that one bolt out, repair what I did wrong, and re-install the bolt to the point it was at, when it slipped. If the bolt washer "slips" on the final torque (most often this is when they slip), I note the bolt, continue to tighten the rest, and go back and "fix" the one/ones that slipped.

Remember, this is an angle torque process. Because of that, the bolt "next door" to the one that slipped has no idea of what is happening to the one that slipped.

In short, pull the 2 bolts, fix the reason they slipped (oil between the washer and the head), and re-tighten those two bolts.

Move forward.

 

When doing the first torque to 20 Nm it makes sense that a washer turning may result in improper torque.

When doing the 90° turns, why does it matter if a washer turns or not?

 

The tightening procedure is a very logical method to ensure the bolt stress falls into the desired "drop zone".

The first step [15 ft lbs] is there to level the playing field as it were and is intended to ensure that the starting point has more or less the same bolt stress. At this level of torque any bolt to bolt stress difference is not going to have a major impact on the final result. The logic is simple enough- supposing after the first step the actual stress in the bolt varies by say 30% [just for purposes of discussion]. Assuming the head bolts are made of grade 8 type steel the maximum design stress will be around 100ksi. If we assume the torque required to achieve this stress level is 75 ft lbs then at the first step level of 15 ft lbs the bolt tension should be around 15ksi. Now, if we assume the first step error function results in variances of 30% the first step stress range could theoretically vary from 10ksi to 20 ksi. On the face of it such sounds rather a lot but once the two steps of bolt rotation are applied those differences remain constant and the bolt stress range then becomes 95 ksi to 105ksi and in real world terms such difference is not really relevant at those stress levels. In practice the first step should result in all the bolts being tensioned to within about 10% and thus when fully torqued the resulting spread of bolt stress is going to be within about 1% thus the logic behind the tightening procedure.

Thus when preparing the system for tightening, the roughness of the bolt to washer surface will impact the first step to some extent but the error introduced is likely to be irrelevant and especially so if all bolt positions are in the same state. Once the first step has been completed, what happens at the bolt washer interface is pretty much irrelevant and the only thing that matters is the degrees of rotation. If I was making these joints I would implement the first step and then mark each bolt with a paint mark at the 12 O'clock position before commencing the two rotational steps. Once this has been done it is more or less impossible to go wrong.

 

When thinking deeply about this it is important to remember that the purpose is to create compression via the bolts. (In the case of the heads. For a ‘normal’ bolt the purpose is to ensure that vibration and temperature changes don’t ‘back’ it out.) The target compression is an input for determining the target stress on the head bolts. The target stress, results in, with the material properties of the bolt and hole as additional inputs, a calculable elongation of the bolts. It is then this elongation that is the target for measurement: When the target elongation is produced the stress on the bolt is sufficient to provide the required compression of the two surfaces. It may also be that the elongation target is higher in order to keep the bolt ‘tight.’

Torque is a measure of dynamic friction of the bolt’s threads versus the bore’s threads along with the bolt head against the washer. It isn’t a measure of bolt elongation and it is affected by many factors (e.g. dirt, etc.) Since, however, it’s rather difficult to measure bolt elongation in a blind hole another method must be devised. Given knowledge of the thread pitch and the properties of the two materials, elongation can be computed as a function of rotation. This target rotation is then our goal during installation.

Friction during rotation (torque) isn’t directly relevant to final compression load of the two deck surfaces. It is, however, important viz-a-viz causing damage to the threads and the physical act of tightening the bolts: We want a ‘smooth’ turn and to minimize galling. Thus, we want clean bolts and holes and a bit of lube on the threads.

Last, it follows that any change to the pitch or material properties of the bolt will change the tightening procedure as a different target rotation is likely.

 

Is there a syllabus for this class?

 

Thank you, Mr. Worf!