New head gasket - to retorque or not to retorque?
#32
Banned
Join Date: Nov 2006
Location: Knoxville, TN
Posts: 1,018
Likes: 0
Received 0 Likes
on
0 Posts
torque and bolt stretch are not equal but torq+ angle will stretch the bolt the proper amount every time.
I just torqued the head bolts on my brother's '99 Cummins he rebuilt using the torque + 90° method. I was using a new Snap On 1/2" drive torque wrench. There was a 20 ft-lb variance.
--Eric
#34
Banned
Join Date: Nov 2006
Location: Knoxville, TN
Posts: 1,018
Likes: 0
Received 0 Likes
on
0 Posts
Could it be due to bolt length? Would be interesting to know which length bolts feel harder to turn vs the others.
--Eric
#35
Registered User
Thread Starter
Then why does this allow for so much variance in torque? If friction is essentially the same (clean threads, clean bolts, everything lubed), why does 90° feel like it's about to snap some bolts, and like it doesn't adequately torque others?
I just torqued the head bolts on my brother's '99 Cummins he rebuilt using the torque + 90° method. I was using a new Snap On 1/2" drive torque wrench. There was a 20 ft-lb variance.
--Eric
I just torqued the head bolts on my brother's '99 Cummins he rebuilt using the torque + 90° method. I was using a new Snap On 1/2" drive torque wrench. There was a 20 ft-lb variance.
--Eric
I measured all my bolts, chased the threads and the holes and used assembly lube.
#36
Registered User
That's a good point, and it could be. That being said, if the desire was to achieve a uniform elastic deformation (stretch) of the bolts, why wouldn't there be different angles for the different length bolts? I.E. shorter bolts get torque + 75° and longer bolts get torque + 90°, etc.
--Eric
--Eric
#37
Banned
friction and bolt stretch are not the same. if you torque a bolt dry ( no lube) you will reach a set torque (on your torq wrench) WAY before the bolt has stretched the proper amount. this is from the FRICTION between the bolt head and the cyl head and the FRICTION on the male and female threads. if you use the PROPER ( most mechanics have no clue what the manufacturer recomends or will use whatever they have handy) lube on the bolt head and threads, and the male and female threads are clean you will come closer to the proper stretch. the two torq specs before the 90* turn are threshold specs to settle the gasket and do not begin to stretch the bolt. the stretch all comes with the 90* turn, or whatever degree the manufacturer recomends. when manufacturers build thousands of engines that they have to warrenty they find the BEST way to do it. that`s why they all use torq+angle or torq to yield. the degree of turn will ALWAYS stretch the bolt the same amount weather it is dry or lubed. you`re right people will always use their own pet cures, and some get away with them for years and others blow up the first time. back in the `70s I was having problems with big block chevy rods coming loose on some race engines. had my torque wrench checked for accuracy, and it was spot on. then I got to reading the manufacturers specs, not the chiltons or motors specs that I had been using. guess what? the chilton and motor manuals said 50 ft lbs. the GM manual said use 30wt oil and 50 ft lbs, or .006" stretch prefered. so I checked and at 50 ft lbs the bolt only stretched about 003". I had to go to about 70 ft lbs to get .006" stretch. that ended the loose rod caps problem. torque and bolt stretch are not equal but torq+ angle will stretch the bolt the proper amount every time. enjoy the new year.
Do you really believe the "imperfections" on the bottom side of the bolt head accounts for 20,000lbs (from a GM test you mentioned earlier) variance from bolt to bolt?
Most manuals give a lubed and a dry spec - even for the torque-turn method. There are two specs to help us mechanics to eliminate the error caused by dry-torquing a bolt with a wet-torque spec. The same manual also generally tells you what lube the manufacturer wants you to use and some even tell you how to lube them, all for obvious reasons.
Why were you using a Chilton manual to build a race motor? Chilton writes books for back-yard fella's who like to maintain and tinker on their family movers. And how did you measure the bolt stretch when they were installed? I've never seen a stretch spec quite like that before to be honest, but did you notice GM's spec and Chilton's spec was the same? Means it probably isn't wrong.... I actually don't think your problem stemmed from under-torqued bolts. GM's 50 ft/lb lubed with 30wt oil is a "factory" spec. One has to assume that because it's a "race" engine you're working on, it would be turning a higher RPM than from the factory, yes? In that case, the rod bolts should be torqued a little tighter anyways, just to cope with the extra stress of higher rpm. I can see why they would loosen off... Other than that, I think your torque was probably just fine. Not trying to tell you how to build motors, just saying....
Torque-turn gives the exact same "distance" of stretch every single time. Yep, no doubt about it. I agree 100%. It does NOT however, set the bolt to the exact same TENSION (or clamping force if you will) every single time on used bolts. The stretch vs. tension value depends COMPLETELY on the material of the bolt. If it has been fatigued at all (used bolts are fatigued to a point, like it or not) they will give easier. That means stretch remains the same but tension is less. With a little experience, you can physically feel the bolt's tension in the handle of your torque wrench (or johnson bar). You can feel the bolt stretch as you turn it, and yes, you can feel some bolts stretch a whole lot more than others before the desired tension is achieved. Some never get there - and you can feel the bolt begin to give in when this happens - it's all very distinct. This happens because the bolts have fatigued over time due to heat and their own tension. The bolt's material strength has been compromised some. So I almost know whether the 90 did the trick or not, you can feel it in the bolt. Again, this is why most manufacturers recommend using NEW bolts if you're going to use the torque-turn method. They know it's not completely accurate with used bolts too.
Yep! Like I said before, it's one of those debates that will never end... Like I also mentioned earlier, it's important to remember that torque-turn wasn't invented because the torque method wasn't working. It was invented to eliminate inconsistencies and errors in the procedure from one mechanic to another. Done correctly, the torque method isn't crap like it's being made out to be. It's true that torque-turn seems to be the preferred method for head's and a mere few other things these days, but don't hang up your torque wrenches just yet 'cause there's still a million other places where a terribly unreliable, inaccurate and erroneous torque spec is the only spec you get! (sarcasm)
New or used bolts?
Smooth torque action? Even on the back ones?
Did you use adapters on some bolts and not on others?
How about different sockets for the hard to reach ones in the back?
Was the 90 done in a single complete and smooth swing?
Recommended or jobber lube?
Lubed correctly? Keep in mind "correctly" depends on the lube.
All of these things affect torque. Any variance in procedure shows up on the torque wrench.
Last edited by 9812vram; 01-12-2013 at 09:13 PM. Reason: Correction and Spelling
#38
Registered User
Join Date: Apr 2009
Location: Banning, ca same house 73yrs
Posts: 173
Likes: 0
Received 23 Likes
on
16 Posts
9812v, I measured the rod blot stretch with a micrometer, you can also use a dial caliper, or a bolt stretch gauge that have been sold for the last 15 yrs or so. it has a dial gauge on an aluminum frame. you put a long box wrench on the bolt head or nut , zero the gauge and turn the wrench till the gauge reads the stretch, it is so simple even you can do it, for blind holes you have to use the torq+ angle.for a million years all manufacturers used torq specs only `cause thats what they were used to. just like the soft composition head gaskets. and with the composition gaskets, GM and most others from the `30s on reccomended a re-torq after a heat and cool cycle. of course all engines of the day were 8 to one compression or lower. in the late `50s GM and others started making 10 and 11 to one compression engines GM found that a steel shim gasket needed NO re-torq,and would hold up, so thats what they used for years on their factory engines. even fel-pro recomended a re-torq on their composition gaskets. then they invented the perm-a-torq head gaskets in the late `70s. things have been better since then. I used the motor and chilton books because the had the same spec`s as the GM FACTORY spec`s. GM had a lot of rod failures in the `50s and early `60s. that`s when they did a LOT of research and published a HIGH PERFORMANCE book. that`s the FIRST place I read the .006" stretch prefrered OVER the 50 ft lb spec. it was the first use of torq angle, only they didn`t call it that. you never read a GM high performance book. that`s why you never heard of the stretch perfered method. CAT and detroit engines have been using torq angle since the `50 and early `60s. the manufacturers KNOW the composition of the bolts, and how much the clamping force varies between new and used bolts. that`s why they have a spec for how much the used bolt can have and still have the proper clamping force on the GASKET. they have done more research than you to determine what method works better to make gaskets last in their engines. they have TESTED the clamping force on the GASKET with torq angle on dry and lubed bolts, they are about the same.( 3,000 lbs variance). the clamping force with torq only on the GASKET with dry and lubed bolts varied 20,000 lbs. torq+angle and torq to yield has been used by all manufacturers since the middle `80s and there are still lots of people that do not understand the the reason why it is so much better than torq only. my only reason to posting here is to educate people on the reason this new technology works and is better than what was used in the past. just like ARP`s new ultra lube is LIGHT YEARS ahead of the old ARP assembly lube, and the new MLS ( multi layer steel) head gaskets will hold more pressure with NO re-torq than previous no re-torq gaskets. it`s called progress. enjoy and embrace it ` cause it`ll make your life easier.
#39
Banned
9812v, I measured the rod blot stretch with a micrometer, you can also use a dial caliper, or a bolt stretch gauge that have been sold for the last 15 yrs or so. it has a dial gauge on an aluminum frame. you put a long box wrench on the bolt head or nut , zero the gauge and turn the wrench till the gauge reads the stretch, it is so simple even you can do it, for blind holes you have to use the torq+ angle.for a million years all manufacturers used torq specs only `cause thats what they were used to. just like the soft composition head gaskets. and with the composition gaskets, GM and most others from the `30s on reccomended a re-torq after a heat and cool cycle. of course all engines of the day were 8 to one compression or lower. in the late `50s GM and others started making 10 and 11 to one compression engines GM found that a steel shim gasket needed NO re-torq,and would hold up, so thats what they used for years on their factory engines. even fel-pro recomended a re-torq on their composition gaskets. then they invented the perm-a-torq head gaskets in the late `70s. things have been better since then. I used the motor and chilton books because the had the same spec`s as the GM FACTORY spec`s. GM had a lot of rod failures in the `50s and early `60s. that`s when they did a LOT of research and published a HIGH PERFORMANCE book. that`s the FIRST place I read the .006" stretch prefrered OVER the 50 ft lb spec. it was the first use of torq angle, only they didn`t call it that. you never read a GM high performance book. that`s why you never heard of the stretch perfered method. CAT and detroit engines have been using torq angle since the `50 and early `60s. the manufacturers KNOW the composition of the bolts, and how much the clamping force varies between new and used bolts. that`s why they have a spec for how much the used bolt can have and still have the proper clamping force on the GASKET. they have done more research than you to determine what method works better to make gaskets last in their engines. they have TESTED the clamping force on the GASKET with torq angle on dry and lubed bolts, they are about the same.( 3,000 lbs variance). the clamping force with torq only on the GASKET with dry and lubed bolts varied 20,000 lbs. torq+angle and torq to yield has been used by all manufacturers since the middle `80s and there are still lots of people that do not understand the the reason why it is so much better than torq only. my only reason to posting here is to educate people on the reason this new technology works and is better than what was used in the past. just like ARP`s new ultra lube is LIGHT YEARS ahead of the old ARP assembly lube, and the new MLS ( multi layer steel) head gaskets will hold more pressure with NO re-torq than previous no re-torq gaskets. it`s called progress. enjoy and embrace it ` cause it`ll make your life easier.
Now, before you run out and tighten all your 12-valve's head bolts to 125, know a few things. First off, bolts will actually stretch. When you tighten them down past specs, two things can happen-one, the bolt will stretch only slightly, and its clamping force will be greater than stock, or two, the bolt will stretch too much, and weaken. If the bolt weakens, it may have less clamping force than stock, or in a worst case scenario, break.
So how do you know if one of your head bolts is in the danger zone? For one, the bolt will feel "buttery." It will keep turning, but won't tighten up any more than a certain amount. To keep empirical track of any butter bolts, we first torqued them all to 100 lb-ft, and found that some had to be turned 1/8 to 1/4 of a turn to even reach the 100 lb-ft mark, indicating that 20 years of running may have stretched our bolts already. Once we got to 100 lb-ft, we marked all the bolts to keep an eye on any that were turning excessively. The engine was still hot as we tried the first bolt, which we took up to 110 lb-ft, then tried for 115 lb-ft. In trying for that next step, the bolt kept turning more and more without reaching the 115 lb-ft "click" of the torque wrench. We decided to leave it be, and let the engine cool off. We loosened it up a little, then put it back to 100 lb-ft."
That come's from HERE. What's the torque-turn spec for over-torquing head bolts in the performance world? There are some things ya just can't do with torque-turn. I'm pretty sure I stated earlier that I use the torque-turn on factory setups and generally only disregard it for the performance stuff (admittedly not always though). You can pre-determine bolt stretch on new bolts via a complicated formula, but ya can't use it to measure the stretch on a used bolt. I think DP explains why pretty well.
Most of the engines I work on have blind holes on the rod bolts. Makes it pretty difficult to "check the stretch". I'd be willing to guess though, that after you found your 0.006" "performance" spec was = to 70 ft/lbs, you used the torque wrench to do the rest! Turning the rotating assembly back and forth to measuring every bolt sounds pretty inconvenient.
#40
Registered User
Join Date: Apr 2009
Location: Banning, ca same house 73yrs
Posts: 173
Likes: 0
Received 23 Likes
on
16 Posts
"From the factory, Cummins uses a torque-angle method of tightening head bolts, which usually will result in a torque spec of a little more than 115 lb-ft. We'd heard racers were having success at wrenching down the stock bolts past factory torque specs, and were running 600, 700, or even more horsepower on stock head bolts without blowing a head gasket. Everyone seemed to have a different idea of how things should be done, or how tight you should torque down the factory hardware, but a common consensus seemed to be that 125 lb-ft of torque was a pretty good number.
Now, before you run out and tighten all your 12-valve's head bolts to 125, know a few things. First off, bolts will actually stretch. When you tighten them down past specs, two things can happen-one, the bolt will stretch only slightly, and its clamping force will be greater than stock, or two, the bolt will stretch too much, and weaken. If the bolt weakens, it may have less clamping force than stock, or in a worst case scenario, break.
So how do you know if one of your head bolts is in the danger zone? For one, the bolt will feel "buttery." It will keep turning, but won't tighten up any more than a certain amount. To keep empirical track of any butter bolts, we first torqued them all to 100 lb-ft, and found that some had to be turned 1/8 to 1/4 of a turn to even reach the 100 lb-ft mark, indicating that 20 years of running may have stretched our bolts already. Once we got to 100 lb-ft, we marked all the bolts to keep an eye on any that were turning excessively. The engine was still hot as we tried the first bolt, which we took up to 110 lb-ft, then tried for 115 lb-ft. In trying for that next step, the bolt kept turning more and more without reaching the 115 lb-ft "click" of the torque wrench. We decided to leave it be, and let the engine cool off. We loosened it up a little, then put it back to 100 lb-ft."
That come's from HERE. What's the torque-turn spec for over-torquing head bolts in the performance world? There are some things ya just can't do with torque-turn. I'm pretty sure I stated earlier that I use the torque-turn on factory setups and generally only disregard it for the performance stuff (admittedly not always though). You can pre-determine bolt stretch on new bolts via a complicated formula, but ya can't use it to measure the stretch on a used bolt. I think DP explains why pretty well.
Most of the engines I work on have blind holes on the rod bolts. Makes it pretty difficult to "check the stretch". I'd be willing to guess though, that after you found your 0.006" "performance" spec was = to 70 ft/lbs, you used the torque wrench to do the rest! Turning the rotating assembly back and forth to measuring every bolt sounds pretty inconvenient.
Now, before you run out and tighten all your 12-valve's head bolts to 125, know a few things. First off, bolts will actually stretch. When you tighten them down past specs, two things can happen-one, the bolt will stretch only slightly, and its clamping force will be greater than stock, or two, the bolt will stretch too much, and weaken. If the bolt weakens, it may have less clamping force than stock, or in a worst case scenario, break.
So how do you know if one of your head bolts is in the danger zone? For one, the bolt will feel "buttery." It will keep turning, but won't tighten up any more than a certain amount. To keep empirical track of any butter bolts, we first torqued them all to 100 lb-ft, and found that some had to be turned 1/8 to 1/4 of a turn to even reach the 100 lb-ft mark, indicating that 20 years of running may have stretched our bolts already. Once we got to 100 lb-ft, we marked all the bolts to keep an eye on any that were turning excessively. The engine was still hot as we tried the first bolt, which we took up to 110 lb-ft, then tried for 115 lb-ft. In trying for that next step, the bolt kept turning more and more without reaching the 115 lb-ft "click" of the torque wrench. We decided to leave it be, and let the engine cool off. We loosened it up a little, then put it back to 100 lb-ft."
That come's from HERE. What's the torque-turn spec for over-torquing head bolts in the performance world? There are some things ya just can't do with torque-turn. I'm pretty sure I stated earlier that I use the torque-turn on factory setups and generally only disregard it for the performance stuff (admittedly not always though). You can pre-determine bolt stretch on new bolts via a complicated formula, but ya can't use it to measure the stretch on a used bolt. I think DP explains why pretty well.
Most of the engines I work on have blind holes on the rod bolts. Makes it pretty difficult to "check the stretch". I'd be willing to guess though, that after you found your 0.006" "performance" spec was = to 70 ft/lbs, you used the torque wrench to do the rest! Turning the rotating assembly back and forth to measuring every bolt sounds pretty inconvenient.
#41
Banned
No offense, but where I work, you'd be let go for poor efficiency, carefully measuring 12 rod bolts for 0.006" and adjusting accordingly after torque-turning them. But that's beside the point... I guess you'll never have a come-back.
That's all good and well and fine. I have no argument with that now or the last two times.
This is where I have an issue. The bolt will hav X amount of stretch per X amount of turn. Got it. Good. But. The bolt is stretched. Why is it stretched? Because it's material makeup has been compromised and the "same stretch" will equal less tension. That's the problem - less tension. Who care's about stretch at this point.
Yes, this is what I'm trying to say minus the gauge limits. See, the thing is, that means there's a "tolerance" right? I've never figured it out, but that tolerance is likely the same ~5% you get with a crappy torque spec.
For the third time, how does a stretch gauge tell me how much further I have to turn a used head bolt in order to gain the proper tension? Torque turn method is near-perfect so I want all my bolts less than 1% difference in clamping force for my 600hp 12valve. Remember, you said yourself you can turn it to the 90 spec but it will have less tension than a bolt that's not stretched...
Nothing miraculous about it. The feeling is very distinct. Sounds like you have some experience with engine building so quit acting like you don't know what I'm talking about.
Tale? As in "fairytale"? Carefull now... There's a whole lot of folks on this forum running some pretty impressive hp/tq numbers on nothing more than over-torqued head bolts. As far as the "new torque setting" goes, the only guys doing this (or should be doing this) are the fella's looking for a little more performance without spendy studs or o-ringing the head/deck.
Yep, that's true. Absolutely no idea how much stretch there is - well, outside of the guarantee that it's more than the factory 90 will give them. Let me explain what they DO know though - the added bolt TENSION!
I'll say it again: the tension created in the bolt as it tightens produces friction. As the bolt is turned, the bolt's tension increases. This increase in tension produces an increase in friction. This friction is what we measure with our torque-wrenches. Thus, friction is proportionally equal to tension. No, it's not 100% accurate and nobody said it was. It is however acurate enough to apparently hold some pretty high hp/tq numbers previously not attained with the factory torque-turn method. There's a whole lot of people that can confirm this - me included. I blew my factory head gasket at ~40psi and some timing. It takes a whole lot more timing and psi to blow the head gasket when the bolts have been torqued to the limit - the limit that was found by torquing the bolts because torque turn doesn't "measure" tension. Are there failures? You better believe it. My head gasket failed too - and it was torque-turned at the factory.
Yap. They're called ARP head studs and they come with a TORQUE SPEC.
Who want's to do that? Over-torquing the head bolts is way cheaper and way less work.
@OP - Amidst all our bantering, there's the answer to your question. A re-torque is a good thing! And jtamulonis has an excellent point - this HAS to be done cold! I forgot to mention that earlier.
...when you know the thread pitch (1.5 mm) for example, about .060", you can calculate the stretch per degree of rotation. one rotation = .060", 1/2 rotation =.030", 1/4 rotation=.015", 1/8 rotation=.007". 1/4 rotation= 90 degrees so I would guess cummins wants .015" stretch on the head bolts....
For the third time, how does a stretch gauge tell me how much further I have to turn a used head bolt in order to gain the proper tension? Torque turn method is near-perfect so I want all my bolts less than 1% difference in clamping force for my 600hp 12valve. Remember, you said yourself you can turn it to the 90 spec but it will have less tension than a bolt that's not stretched...
I'll say it again: the tension created in the bolt as it tightens produces friction. As the bolt is turned, the bolt's tension increases. This increase in tension produces an increase in friction. This friction is what we measure with our torque-wrenches. Thus, friction is proportionally equal to tension. No, it's not 100% accurate and nobody said it was. It is however acurate enough to apparently hold some pretty high hp/tq numbers previously not attained with the factory torque-turn method. There's a whole lot of people that can confirm this - me included. I blew my factory head gasket at ~40psi and some timing. It takes a whole lot more timing and psi to blow the head gasket when the bolts have been torqued to the limit - the limit that was found by torquing the bolts because torque turn doesn't "measure" tension. Are there failures? You better believe it. My head gasket failed too - and it was torque-turned at the factory.
@OP - Amidst all our bantering, there's the answer to your question. A re-torque is a good thing! And jtamulonis has an excellent point - this HAS to be done cold! I forgot to mention that earlier.
Last edited by 9812vram; 01-13-2013 at 11:09 PM. Reason: SP
Thread
Thread Starter
Forum
Replies
Last Post
aksparkey01
Performance and Accessories 2nd gen only
6
11-25-2007 02:51 PM