Hi ALL, the BEST TECH HERE was from SpeedTrigger, MY HATS OFF !

I can "feel" bolt stretch AND use this way for the calibration, the torque value, I use with my Snap-On Dail-a-Beam, the first "pull" with a strong arm.

The ROD BOLT torque is a "stretch" method, the ONLY one to use.
The common .0006" increase, a temporary increase in length, is found.

THE SAME METHOD as SpeedTrigger suggested as the correct procedure.

NOW add my "arm feeling for stretch" method WITH the education of a "bench" fixture, breaking several bolts, from "feel" of bolt stretch AND you will have fifty years of education in an HOUR.

Lance

 

Or just go ARP so that you can use a torque wrench instead.

 

Well, if it's stupid but it works...then it must not be that stupid

 

The problem with torque spec bolts is that if the threads aren't perfectly clean and the proper lubricant used; you end up with improper stretch anyways because the wrench will "click" from resistance to motion/galling/binding, etc.
E.P.L.(extreme pressure lubricant) works great under the head of the GM fasteners and as others have stated; a sharpie mark on the bolt head and eye balling the 90/70 degrees has worked great for me. As Lance stated; you can "feel" the fastener do it's thing at the point where your 70 or 90 degrees ends up at.
Not exactly rocket science.....important YES but there are plenty of other discrepancies that the head gasket has to deal with

 

This is a stress/Strain curve of a steel bar. The Y-axis is the Stress - the stretching force applied to the bar, expressed in KSI (1-KSI = 1000-PSI). The X axis is the Strain - the amount the bar actually stretches as a percent of the original length.

There are two types of strain: Elastic and Plastic

Elastic strain means that you stretch the bar, but it can still spring back to its original length. Plastic strain means the bar is permanently longer.

See that first part of the graph, where the curve is a steep diagonal line? That is all elastic strain. A certain amount of force stretches the bar a certain amount. When you are torquing a bolt, you are on this diagonal line. As you tighten the bolt and feel the increased torque, you are actually stretching the bolt elasticly. The more you try to turn it, the more torque is required to turn it. This is what you have with almost all bolts. When you loosen the bolt, it will go back to its original length.

Torque to yield bolts are different. See how the curve in the above picture stops rising at 72 ksi and just goes straight across? This is the yield strength of the bar. In this case, you are stretching the bar, but there is no additional force required to stretch it. This is what happens with TTY bolts. You ever notice how you get to a certain point on a TTY bolt, and the bolt just keeps turning while you're just putting constant pressure on it? In this case, you have strained the bolt plasticly. it will not return to its original length. You can verify this yourself by measuring TTY bolts after removal and comparing them with brand new ones.

So, if you followed the above, there is NO WAY you can use a torque value to properly tighten a TTY bolt. if you prefer to use torque over angular strain, then you must go to a different (traditional) type of bolt that only uses elastic strain to provide clamping force.

Hope that helps

 

I worked where we had really expensive precision equipment. We all had to annually get certified on procedures for assembling various parts of the gear, the guy who taught the class was our in-house calibration shop supervisor..

To prove to people how important torque specs were he had a machine with a bolt connected inside the box to a strain gauge setup that measured torque. The machine had a panel where you could set a target torque and have someone put a wrench on the bolt and try and torque it to spec, he would have every body in the class tighten the bolt with the goal being to hit the displayed torque by feel.. In like 8 years he had exactly 1 guy that could hit a torque spec blind.. It was the guy in his shop that calibrated all the plants torque wrenches.. Site had about 5000 employees that got certified each year.. Most folks were a minimum of 20 or more lbs off.. Was always hilarious to watch , guys would bet on it and it made the class more interesting..

 

And this is why you spend the time and money to use the right tools and techniques to do it right the first time.

You're not building a tractor motor with low compression and low rpm's. You're building a performance motor and you want it to last. Otherwise you wouldn't be here, you be reading JohnDeer.com.

How many times do we read here that a brand new motor blew up? or didn't make oil pressure? or failed after 200 miles? Why does that happen? This isn't magic, or rocket science. OK, well maybe it is a little bit of rocket science. It happens because people cut corners, don't follow procedures, and just wing it. That LS3 is a precision machine. Much more so than your daddy's 1969 SBC. What worked before doesn't work now.

I'v lost track of how many engines I'v worked on over the last 40 years. But when I bought a used LS1 for this race car, I bought another book, read a lot of web sites, bought some new tools, and learned some new techniques. So far, that engine has been rock solid and has won a lot of races with nothing more than routine maintenance. I'm not the sharpest crayon in the box. But I can read and follow directions.

 

Just to play devils advocate here. If we were to take a good sampling of stock, unstretched bolts, and experiment to find which torque consistently moved the bolt into its plastic transition range...then couldn't we agree on a torque value, or at least one that will be effective?

Every stress/strain curve has a "flattening" section or even a slight dip in the curve at the plastic transition.

The next argument is that not all of the bolts will have that same torque value, and I agree...but then not all bolts should have the same degree value as well. What GM is doing with the degree value is ensuring the bolt ends up SOMEWHERE in that plastic region, before necking. It just gets you somewhere between the plastic transition, and the ultimate tensile stress point.

I believe there is certainly a torque value (assuming proper lubrication and clean threads) that would correspond to the degree value and also land you in that same region.

 

right. the stress/strain relationship varies quite a bit for various alloys and also depends on their state of temper. steels are usually characterised by yield stress, tensile/ultimate stress and elongation that give an approximate idea of that relationship. for 8740 steel it is 670 mpa/855 mpa/22% when normalised. that means you can strech it 22% before it fractures. when its heattreated to e.g 1200 mpa tensile, yield and tensile move closer together and elongation might be 10% (out of memory).

https://www.recreationalflying.com/t...ss_strain4.gif

so its clear why you would stress a bolt "to yield" (meaning beyond yield stress): you can stress it more. if you have a tty bolt the method of tightening is not implied and vice versa. e.g. on prv engines the headbolts were quoted in torque until around 20 years ago and then they switched to angle. but everybody new what the figures were, it was the same bolts and they were not tty.

its important to know if a bolt is tty, because then you should not reuse it. well, you coud if you had a way to torque it to the correct length. above it will travel from sy on the blue line down when losened. from that point you can retorque it to sy again, but thats kind of hard to do.

(btw, the minimum above is not always there, the line can just be convex.)

 

Yes, you could, but in that case, you would only be torquing the bolt TO it's yield point. Once you reach yield, the bolt continues to stretch with no additional force. Meaning you could rotate the bolt an extra, say 1.5 rotations and feel no additional torque or measure it with a wrench.
True... For the most part. And if you miss your target rotation by a few degrees, no big deal, because you are on the yield "plateau" somewhere. There are stress/strain curves that don't have a plateau

You'd either be AT the yield point, or past the plateau and into the strain hardening section of the S/S curve. BTW, I like the chick in your sig pic, just wish I could see her better...
Someone understands a bit of metallurgy! You're right, you heat treat the bolts for strength, you lose both elongation and strain hardening capacity past the yield point (tensile to yield ratio). BUT you GAIN a tremendous amount of elastic strain capacity along the modulus (assuming a classic elastic/plastic curve like the one I posted). So you don't have to torque to yield in that case, you can simply torque to a point high enough on the modulus to have the bolt act like a spring. ARP bolts for example. But then, how much of the torque is friction, etc, so you get into the preload lubes like has already been discussed.
Also correct. TTY implies the bolt has been permanently stretched. Again, the ARP bolts are specifically NOT TTY, as evidenced by the fact that they CAN be reused. Bit of a sidebar, but I work with materials that we deliberately strain harden - We strain the bar past its yield strength and hold it there for a period of time and then stress relieve it. That bar when retested has a much higher yield strength than the original test. And a compressed T/Y ratio similar to heat treating (in most fastener cases, quenching and then tempering). TTY bolts are similarly strain hardened via installation into the engine, though to a lesser degree. They won't yield the same as they did as originally supplied, and you'll have less reserve strength before it breaks - compression of the T/Y ratio.

That convex curve you mentioned, we commonly call a round-house curve. If you have something with that shape, you don't get a yield plateau, and you'd just have to use torque. There is nothing wrong with that type of curve. In fact, newer research is coming out showing it can be beneficial in concrete reinforcement during earthquakes, but that's an entirely different can of sauce.

 

Hi ALL, GREAT "Tech" LS-1 members should be proud.

The SBC engine, engines of the past, where built 3X the needed strength.
The current engine, LS included, is ONLY built 1x + the needed strength.
The LS engine is designed to be assembled ONCE.

There are NO Hands in this work, it is done by a Machine at high speed.
These machines "read" TTY bolts within a large range of distance.
They do make automated Torque Meters, VERY COSTLY with need of repeat calibration.
The TTL method requires NO repeat calibration and uses a position encoder for report.

Race engines use Studs.

Lance

 

good thread, backed up by engineering

It is very interesting/educational watching the steel alloy S/S trace being drawn in real time in the lab!

 

Hi Joe, yes IT IS TIME THAT LS-1 TECH "stands -up"

There is the BEST THREAD, the "paper trace" that is the BEST.

This does NOT include the REPORT of Mike Costin about bolt stretch.

WE ARE making progress !

After fifty YEARS of Race ENGINE "tech" this information NEEDS to be "shared".


Lance

 

its important to keep in mind, that up to 90% of final torque gets consumed by friction and more than half of that by the head burnishing its seat. if you then consider that the friction coefficient of fasteners can vary between 0.05 and 0.25 it becomes clear that if you tighten to a torque value, the stretch inaccuracy can be huge.

btw, an intersting experiment you can make. thread three holes into aluminum. take a grade 8 screw and find out at which torque the thread yields. put it in the second hole and torque it to 90% of that value. its fine. now put a nut and washer on the screw, thread it into the third hole and tighten the nut to those 90%. what did just happen? you ripped out the thread. just some food for thought.

 

Whoa! Good point! Same holes, three circumstances, all three encountered in the same situation in a build....

 

I....think my mind is blown.

 

It's very interesting watching it. It used to be you only captured the yield strength and then removed the strain gauge. We moved to video so we could get the rest of the curve. What's really driving it is seismic research. Imagine a building swaying 3-4% in an earthquake and what that would do to the structure - especially the concrete. Now imagine being able to just repair some concrete and move back in. The demands on the steel are tremendous.

But the info is still applicable to bolts. We are after all still straining steel under a load, just different means of applying force.

@Dian, good point about aluminum threads. Part of the issue with the northstar. It badly needed steel inserts. Though admittedly, that was only ONE of its shortcomings.

@pantera, easy to see why in a race engine you'd prefer studs. Greatly reduced frictional side of the equation and you only stretch the stud, you don't stretch and twist simultaneously. Northstar would have done much better with studs from the factory.

 

It was always my understanding that the LS is actually a more robust design than the SBC. I have actually heard from a few sources that the aluminum LS blocks are as strong as your typical iron SBC, and that iron LS blocks are ridiculously stronger than your typical SBC. After all, such features as a deep skirted block and six bolt mains certainly don't make for a WEAKER block.

 

These engines are stupid simple and reliable. I used to be all **** but now I have put several together under the influence Ha Ha.

Black mark through head bolts after torquing to measure out degrees more or less. Never had a problem. I don't buy new balancer bolts anymore just impact them on. Oh the horror!

Re use head gaskets, and all other gaskets. Every vehicle I have has an ls engine.

On my TT camaro I take care when I work on it but it get's stretched.

 

Hi MATT, you "TWIST" what I stated.

I too AGREE that the LS engine series IS STRONGER, items learned from Fords 427 "Side Oilier" design.

The SBC 265 made 110 HP with the more modern 350, an easy 500 HP with modifications.
Thus do the MATH
WHEN YOU READ the GMPP block ratings for HP, THEIR report by GM, the TRUTH is written.

Lance