So if you're rebuilding a customers engine, on a regular basis, you need to put some kind of marking or serial number on each bolt so that you can track it through out its life cause its probably not going to stay with the same rod, and on the same side of the rod for that matter. So how do you mark them? One of those Radio Shack electric etchers or what?

 

In the circle track engines the bottom end bolts get replaced every season any way and are put in the trash. On street builds they mostly come back for more HP and most bolts are reused. The majority of my street engines,according to the owner's are good enough, and I don't get many back to freshen or upgrade. It's money and they think they are fast enough.The engines are sold with the cars after a few years.

Engraving don't seem like a good idea.Ya got undercut and a place to start a crack.

 

The best way is to measure each bolt before removal to determine the stretch on it as they were machined. Then you can verify it was properly torqued before machining. I generally disassemble, clean, retorque & then check the rod dimensions before starting assembly. Your motor is only as durable as the worst part.

On the eagles, I also noticed near 80ft/lb using the stretch method. This worried me & is likely due to the machining of the bolt face of the rod or the thread tolerance.

The Lunati rods that I used consistently required about 6ft/lb over to get the required stretch. Your torque wrench may vary.

It seems that the rod bolt/nut setup on old style stock rods were more effected (distorted) by altered torque values than the late model & aftermarket cap screw type fasteners.

Stretch is more consistent & can not be all that bad considering most new design motors use torque to yield fasteners which rely more on physical stretch than torque values. I now measure the stretch on all my builds.

True many engines have been built in a gravel driveway with a $20 torque wrench, but I do not know of a paying customer that would tolerate that level of detail on their engine.

For what its worth......Fewer shortcuts generally result in fewer regrets.

 

Gravel drive--- Isn't that hard on the skin and bones?

 

All fasteners torque values are based on stretch, it is pre-determined value that give the fastener its stretch without the need to actually measure it.

Torque to yield fasteners are junk, they are made out of cheap chinese material and once stretched, cannot be re-used.

 

In a perfect world you're right... you would simply torque to spec and the bolt would always achieve a load necessary to holding the joint together in service. Just doesn't happen that way in the real world though.

Torque is simply a measure of resistance to friction and friction is a sumbitch when it comes to any kind of simulation. Just too many variables at work there.

If I didn't want to use the "stretch" method for a bolt then my next step would be torque to angle. Much more accurate than trying to measure resistance to friction at higher thresholds.

"Junk" is a harsh word doncha think?

Hey, they do their job at a reduced cost to the manufacturers. Reducing costs is a good way of keeping jobs here in the good ol' USA.

On another note,
I am one of those customers who happily pays to have things assembled correctly. If an engine builder wasn't using rod bolt stretch, then he wouldn't get my business.

-Mindgame

 

Something interesting can be gleaned from this discussion. There is often more than one acceptable way to achieve a result. And no matter what, good enough is good enough. I am not involved in high end race motors but I don't doubt the theoretical superiority of the stretch method for those applications. I am involved in hi-po street and street strip motors and the old method seems to work fine there. I think the reason is that we tend to use parts that are overkill for this kind of use, so even when put together "incorrectly" there is plenty of margin for error.

I don't care what method is used to accomplish a task as long the result is acceptable.

Rich

 

I agree, Rich. Sutures, staples, or medical superglue are all ways of closing the skin. Each has its plusses and minuses, I'm sure. Getting clamping load on a bolted joint is somewhat similar.

As has been pointed out, stretching the bolt is what produces the clamping load. Measuring stretch is the most accurate method of determining clamping load if you can't put a tiny load cell under the bolt head or nut. It's also slow and the bolts have to be properly prepared with centers or ground ends.

Torque-angle is probably the next best because you are keeping the thread friction in a lower range, and stretcing the bolt a certain amount with the angle. On a 20 pitch thread, 36 degrees is .005, so you can probably come closer to the desired stretch than with torque alone.

With torque alone, the condition of both the male and female threads effects the stretch/clamping big time. Unfortunately the bolt manufacturers like ARP don't thread the rods the bolts go into. If nuts are used, they may buy them, so the only real checking they can do is with go & no-go thread gages. Also, good rod bolts have a rolled thread, but some rods may have a cut female thread. They really need to be cycled under load a few times to become friendly (or intimate might be more appropriate) with each other.

Torque-to-yield (TTY) is another horse altogether. Stress-strain curves for different grade bolts are very different. They all look like horsepower curves, or aerodynamic lift curves or tire slip angle curves, but like engines, wings and tires, there are many shape variations. The highest strength bolts have a nearly straight-line curve with a sharp drop after peak and then failure, while grade 5 or lower bolts have a much more gradual rounded peak and a long delay before failure. IMO, TTY works on the lower strength bolts (which are probably at least Gr.5) because you can just barely yield the bolt and still stay on the "front side" of the stress-strain (or "power") curve. You are then getting about the max out of the bolt safely.

FWIW, I disagree that all TTY bolts are Chinese junk. Unfortunately, nearly 40% of the world's steel comes from China. Much of it is high quality, some isn't, but that's the same no matter where you buy it. You need to choose your steel suppliers carefully if you are a bolt manufacturer or an OEM.

Of course you need to replace TTYs after you disassemble them, That's part of the design. OEM engines are meant to run for a long time before rod bolts are removed. Most modern engines never get torn apart to repair them. If that is needed, wouldn't new bolts, which haven't seen hundreds of millions of load cycles be a good idea anyway?

My $.02

 

You're right Rich, we all tolerate different levels of quality in our own work as well as in our products.

Lonnie's also right, it's all about minimizing risk. Good point and the one I subscribe to.

I'm sure there are a number of guys on this forum in the same boat as myself with their "real life" jobs where just about the only acceptable level of performance is that of perfection. Makes you kinda **** about things. Well... it does once you've been burned enough times by the "little things".

In my mind, 9-second strip/street daily drivers and endurance "racers" are one and the same.

-Mindgame

 

I observed a "bolt stretch" install today with Eagle H-beam rods equipped with ARP 2000 bolts. The bolts had been torqued into the threaded rods a number of times using the ARP lube. A dial indicator stretch gage with points on either end was used. Repeatability was within .001 no matter how many times the gage was removed and reinstalled.

Desired stretch was .0055-.0059 or effectively 5-1/2 to 6 thousandths. All 16 of the rod bolts eventually got there, but the interesting part was the variation in torque required to achieve the results.

ARP suggested 43 lb-ft. with their lube and multiple cyclings. Actual values varied from about 37 to about 55 lb-ft. to get the stretch. The torque wrench used was a high-end Snap-On so readings are very repeatable. The bolts were pulled up in multiple steps to get the correct stretch without going over.

Interestingly (to me at least) was that torque values needed ranged about -6 to + 12 of the suggested value, or -14% to +26%. Ouch!

Because these were threaded rods, not bolts and nuts, my take is that the threads in the rods cause the variation. Everything from tapped hole size, to straightness to heat treating distortion could cause the problem. Those variations make sense to me.

This procedure is not quick nor is it easy, but for a engine designed to build lots of power and rpm, IMO it's worth the time.

 

Hey Guys, let's talk about the compression of the rod. I would think that the rod big end would take a certain shape when the rod bolts are tightened.

So what really brought on this thread was my worry that I slightly distorted my rods by going to 80ft-lb instead of 60ft-lb. (I went to 80ft-lb due to trying to get the proper stretch).

It all boils down to me, (the customer), needing to make sure that I tell the machine shop that this rod with bolts needs to go to .005" stretch which is 80ft-lb. If I don't tell the machine shop then they will probably go to the torque spec of 60ft-lb.

So do you think there is significant rod distortion? Does the rod cap take a different shape between bolt clamp 60ft-lb (.003" stretch) and bolt clamp 80ft-lb (.005" stretch)?

There's probably a bunch of home rebuilders here at camaroz28.com. That's my case: I rebild a few of my hobbly/race engines each year and thus keep track of rod bolt health and such. I do not measure initial lenght of the rod bolts but what I do is note the torque needed for proper stretch and if the bolt follows that it's good....if it goes past that it's junk....(have not found a bad rod bolt yet in 10 years but I keep an eye on em).

Karl