I am also rebuilding a 930 - 3.3 liter engine.

I just got off the phone with ARP tech support about the installation proceedure for their 6005 rod bolts.

They advise to press the rod bolts into the rod, use the ARP assy lube on the bolt and bottom of the nut (NO LOCTITE), then torque down the bolts to 35 ft-lb, loosen and redo 3 to 5 times. If done this many times they guarantee that they will be stretched to the proper length of 0.0120"

Has anyone ever used or heard of this proceedure?

Very informative thread!!

I'll be doing this myself soon!
Great reading!

The only reason I can think of for torquing/ then un-torquing is to make the threads on the nut and studs more uniform, hence giving a more repeatable torque gage reading.

The problem with torque method is that 85% of the torque goes into overcoming friction at the thread and at the nut face. So if you're torqueing a fastener to 45 ft-lbs, only about 7 ft-lbs goes into stretching the stud. If your wrench is off by 5%, then that means that you are torqueing error is +/- 2.25 ft-lbs. The error is huge in comparison to the torque that goes into stretching.

Long story short, if it's critical, use the measuring technique. It's the much, much more precise than the torque method.

Yes, commonly recommended procedure with aftermarket rod bolts to burnish the threads. That said, we have found it still requires far more torque to get the required stretch.....

Cheers

Pauter ARP bolt stretch?
 

I have a set of Pauter rods. So the bolt is torqued, not the nut (the bolt threads into the rod cap).

The bolts are ARP, 3/8" in diameter. About 1.75" overall, which is a lot shorter than rod bolts for stock Porsche rods. Pauter says to torque to 50 lbs/ft using motor oil as a lubricant.

The information I got from Pauter, when asked about stretch, is that it should be between 4 and 5 thousandths of an inch. Because these are shorter than the bolts we are used to, I figure what I did for Racewares on my stock rods doesn't mean anything.

I have a Summit stretch gauge. Inexpensive, and works great even if I have to interpolate beyond the 0.001 markings. Also has lots of spring tension, more than the digital gauge I thought I could substitute. Helps a lot with consistent readings.

Anyone know if 0.004-0.005 sounds like the proper stretch for these?

Walt

p.s. - a buddy had a Pauter rod bolt break while I was racing him at Miller last fall. He says that for a while Pauter was not using ARP bolts. I was happy to find that my Pauters were not from that batch! These high strength steel rod bolts simply should not fail, as other things ought to go bad first. I had Racewares hold (though bent) while the rod cap itself bent due to a bearing failure. Piston hitting valves was the immediate symptom.

Walt, what's the material of the bolts? And do you know to what percentage of yield they target at4-5 mils? As someone mentioned earlier, a common target is 60% of yield strength.

Alfonso

I don't know. The bolts came with the Pauters, so I only got Pauter's information, which was a bit sketchy (torque value only) if one wants to look deeper. The bolts say "ARP", "2000", and "-02".

I'll take a look at ARP's website. They may have some info there.

Walt

ARP's website is full of information generally. It would appear that my bolts are this alloy:

19. What exactly is ARP2000 and how does it compare to 8740 and 4340?

ARP2000 is a heavily alloyed martensitic quench and temper steel, initially developed for use in steam power plants. As such it has excellent stability at high temperatures. But most important, ARP research discovered that in addition to temperature stability it has excellent notch toughness in the higher strength ranges and is alloyed to be tempered to Rc44/47. 8740 and 4340 can be tempered to the same hardness. But, the tempering temperature would yield material in the “temper brittle zone” (between 500° and 700°F), producing significant notch sensitivity. ARP2000 is tempered above that temperature range and has a strength between 200,000 and 220,000 psi.


Elsewhere, ARP say that ARP2000 has a yield strength of 180,000 psi, and a tensile strength of 215-200,000 psi.

ARP also states that fasteners which have permanently stretched more than 0.001" are permanently compromised and should be discarded. But it seems to me that stretch ought to be compared to length, as that amount of stretch in a short bolt would mean more than in a much longer one, would it not?

Walt,

You can use the formula D = (p/a) x L / E

where
D or delta is the stretch, .0045 (average of .004 to .005)
(p/a) is the loading on the bolt, in psi
L is the length of the bolt, 1.75 inches
E is the modulus of elasticity, 30,000,000 psi

You can back calculate the (p/a) is equal to 77,142 psi. This is about 42% of the 180,000 psi yield strength.

I would say that if that is what they recommend, that should be good. If you wanted to load the bolt to 60% or 108,000 psi, then the stretch would be .006".

If you are really worried about .004-005 not being enough, contact ARP and let them know your concern and what would be the consequences of loading the bolts further. The formula I posted above is based on well accepted theory. There is nothing like experience, though, and the folks at ARP should know their stuff. There might be a reason why they choose to load it 42%. As you probably know, you have to consider the whole system, not just one aspect. Putting additional load on the bolt may not hurt the bolt at all, but my introduce unwanted stresses into the rod and bearings. Hope this helps.

Pauters torqued and stretched.
 

After trying hard to outthink things, I finally called ARP.

My issue was this: Pauter said torque their 3/8" ARP bolts to 50-55 lbs/ft in the sheet that came with the rods. When I asked them about stretch, they said between 4 and 5 thousandths of an inch. But when I went to ARP's website, for what appear to be the same bolts they said torque at 45 lbs/ft or stretch 5.5-6 thousandths.

And using the formula kindly provided here, 60% of yield on a 1.5" bolt would be a stretch of 5.4 thousandths. But that has some loose ends. I never measured the distance between the base of the bolt head and the first thread in the cap. That distance, or some slightly longer distance depending on how many threads down one ought to go, would be the part of the bolt which actually gets stretched. And it will be less than 1.5" by a significant amount.

Not unexpectedly, the ARP guy said go with what Pauter says.

So I did. Here is how things worked out (with thanks to Chris Bennett, who sent me the spread sheet I modified for my own purposes).

http://forums.pelicanparts.com/uploa...1274303849.jpg

Most just hit the lower spec at 65 lbs/ft, one took 70, and two 75 lbs/ft. I added a column at 55 lbs/ft just for fun.

I used Torco assembly lube (same stuff I put on the bearings), and didn't do the burnishing bit. Perhaps had I done that (and maybe used motor oil, as Pauter recommends), it wouldn't have taken so much extra torque. But the good part about the stretch gauge is that you don't have to worry about the variables, including the accuracy of one's torque wrench.

I think the suggestion from early in this discussion to at least start with torque, so that any defective fasteners will be discovered, is a good one.

Walt,

I think you have a good plan. I agree with your statement, measuring does takes the friction factor out.

Friction is the main factor in the torque method and there are too many unknown variables to consistently preload a fastener using that method. This may not matter when you have several fasteners as you do in a pipe flange, but on a reciprocating component, the more consistent and accurate you are, the better for the reliability and life of the engine.

Good luck.

I stumbled across this topic when searching for the type/brand of stretch gauges people have used. I luckily found Walt's mention of the Pauter rods and smaller ARP bolts, which is exactly what I have.

What I found is that a stretch gauge using a dial indicator with 0.001" increments is not sufficient for these short ARP2000 bolts that require a 0.0055" stretch spec. Using my 0.001 dial indicator seemed suspect to me because I would be relying on the indicator to accurately tell me i'm 1/2 way between 0.005 and 0.006 on the dial. My dial is a 100 increment dial, so the sweep of the needle is only 5 to 6 increments in this case.

So I decided to try the 0.001" indicator to see how much torque is needed to get to 0.0055" stretch. Came out to about 60 ft-lbs, which is not too bad. But I still didn't like the resolution/accuracy of my indicator. So I took out the metric indicator that is a 10mm gauge with 0.01mm (0.00039") increments. Upon remeasuring with the higher resolution indicator, I got 49 ft-lbs of torque on the same bolt, same rod.

However, my metric indicator seemed like it was acting funky. Maybe it was the cold workshop...... Seemed to work better when I took it in the house house and things warmed up. I opened up and checked it out (wild how simple, yet precise these are inside) and everything seems in order. Granted its a Baker economy indicator ($29) so i'm not claiming its a world standard Brown & Sharpe or Mitutoyo. Nonetheless I will buy another indicator and recheck my work with my 0.0001" accuracy micrometer.

I guess my point is:

1: Thanks to everyone, and also Walt for sharing his experiences with the same Pauter rods and bolts.
2. For these ARP2000 bolts that have a much shorter stretch spec than the OEM-style bolts, I feel you must use a micrometer or dial indicator that has at least 0.0005 increment/accuracy. A measuring instrument with only 0.001" resolution isn't enough in my opinion.