djl

Just got to wondering about this. When you install chassis or suspension bolts do you lubricate the threads? Seems like a given level of torque would drive a lubricated bolt in further and compress (bushings, for instance) more than if if were not lubricated.

Octania

While that is true across the board, I think it is only crucial for things like our rod bolts, main caps, head bolts, and the like, which are fairly picky about proper torque. The suspension large bolts are a surprisingly low torque as I recall- like 45 ft-lbs or so, and rely on the self locking nut for certain retention.

I feel that being able to remove the bolt in the future is more critical, so the entire bolt gets greased. Won't rust in place in the bushing center sleeve that way.

Yes, you should shoot for a bit less reading on your torque meter if using a lube on the threads, assuming the goal is a given amount of clamping force produced by the fastener. How much less? Maybe google up some studies on the matter. I would think it's on the order of 10-20%. Do you add torque for when using rusty unlubricated fasteners? Frankly, in the real world, most things get hit with the air tool until it sounds tight and call 'er done.

I doubt that, if using a torque instrument, you will notice any difference ever between securing these bolts to 45 dry or 40 lubed.

joe_padavano

The measurement that matters in any bolted joint is the tension in the fastener. Unfortunately, this is difficult to measure directly. On critical fasteners in aerospace applications, we actually use strain gauges built into the bolts or load-sensing washers to get a direct tension measurement. Using torque as an indicator of tension is easy but notoriously inaccurate. Dry fasteners have a much higher variability, as rust and dirt in the threads can cause the torque to be much higher for a given tension load. As Chris points out, there are only a handful of really critical fasteners on older cars, and those typically have a torque that is specified with the threads lubricated in motor oil to reduce the variability. Racers go even further with tools like rod bolt stretch gauges.

Older cars are significantly overdesigned, so this variability in fastener torque usually is not an issue. Newer cars can't afford the weight hit, due to CAFE requirements, so they resort to tricks like torque-angle meters for head bolts to verify that the torque is correct. A bolt is actually a pretty well calibrated micrometer, and since stretching the bolt in tension is just a spring with a fixed spring constant, once a predetermined torque is reached, turning the bolt a given angle produces a pretty repeatable tension load.

djl

OK - Interesting stuff. I was afraid it might be a stupid question, but I guess it's not. It just doesn't have a lot of bearing (pun intended) on real world application.

Koda

Joe gave a good answer, as always. Your axial force (we call the guys who have to check it the ******* force guys) is what matters. At my company, we will sometimes put a bolt in with a machined flat, highly polished head, and use ultrasound to measure the stretching of the bolt under tension, which gives it the clamping force you need. Toyota bolts are generally fine thread bolts with a little dried red threadlocker on them. They back out well because of the fine threads, and the real trick is to use a bolt just long enough so there aren't many threads outside the nut to get corroded and fight on the way out.

We do track torques via nutrunner computers, and will do angle checks, mainly to make sure that it is not a guy hitting a bolt twice instead of all the bolts in his pattern. Serious torque charts have both torques listed, wet and dry.

In my opinion, put it to the right torque listed, wet or dry, the 20% extra torque should not affect anything in a chassis environment. I do put a dab of grease on my lug studs, and will torque to the listed amount. I do not think that is enough to warp a rotor.

joe_padavano

So you guys really do use that statistical process control stuff. Cool. Too bad GM didn't listen to Demming.

83hurstguy

These guys have pretty much said it all... one thing to note is that certain type of lubricants can burnish the threads, effectively increasing stretch every time the fastener is torqued. Joe Mondello talked about doing a rod bolt stretch study with same torque value, and mentioned that when using the ARP moly lube, the bolts start to stretch more after a few torque cycles.

Granted, most people aren't cycling fasteners on a car like that in any application except for the lug nuts. I have used some liquid lubricants on them before in rusty passenger car applications and never seen an issue. On the hot rods, I typically change to ARP studs with McGard nuts (and they don't see corrosion), so no lube is required. I always have used anti-seize on the rear control arm bolts, those things are a bastid to remove as it is, and I've never noticed any issues with an overtorque from application of the anti-seize.

BlackGold

The first rule of SPC is that you can't use SPC if your process isn't under control. Maybe that's why GM didn't use it.

joe_padavano

Two points for that one.

Koda

I work in the Assembly department of production engineering of one of the companies, mostly in chassis. We have two styles of tightening. First one is temp tightening, which is done with a pneumatic impact or cordless electric impact (the latter of which is becoming more common.) This is known as the "get it in there" processes. The guys on the engine lift process put in the entire front powertrain assembly from the bottom, raise it up till the little grabber guide cones lock the McPherson struts into the fenderwells, then they shoot two bolts apiece to hold the H frame in. (engine sub frame.) They then drop that lifter and move on to the next car.

Meanwhile, that powertrain isn't going anywhere fast, but it needs proper torquing. We normally use one brand of torque nutrunner computer, and it has a program that is looking for a number of shots. Let's say, our engine guy needs to shoot those 4 bolts, so that nutrunner wants 4 torques that will have the proper curve showing that it went from the temp torque to final, and we can track that with those curves. The nutrunner will not know which bolt is which, but it will know which order any problem happened, and the standardized work will show which bolt it is.

Two things happen then. One, the nutrunner computer sends the data to the network which then logs every torque for every car and links it to the vin for records if they are ever needed, but the line never sees that. The computer also sends a OK or NG (no good) to the control panel for that equipment, if all the torques are in target or not. If they are, the car moves on down the line. If they aren't, the panel starts a ruckus, and someone will either quickly retap the threads and try again, or click off with a manual torque wrench that has its load cell tied into the line, or they will key it off and log it to be fixed in confirmation.

We do run CPK statistics on nutrunners when installed or changed.

Hopefully that wasn't boring.

BangScreech4-4-2

I thought the first rule of SPC is: You do not talk about SPC.

joe_padavano

Actually, quite interesting. Thanks for the info. In the space business, we do something similar, but the stuff we build is onesie-twosie, so there is no way an automated system like that is affordable. All torquing is done by hand, data entered into a log book manually, and stamped off by a separate QC engineer. The books are retained as required for the duration of the mission. We also use torque stripe, where a paint stripe is applied to each critical fastener after torquing. This not only verifies that the fastener has been torqued, but also provides some level of visual indication if a fastener backs off.

Octania

Nerdfest!

[in a good way]
Joe, Koda, I for one appreciate your perspectives.

At Fiasco motors, seat and ABS motors and such, we had the practice of marking case bolts with a paint pen after torqueing to indicate it had been torqued.

Later, problems arose again. Loose case bolts. Why? Turmover in the line staff... new folks were taught to put a paint dot on the bolts, but not WHY that paint dot is there. Soon the torqueing got skipped but the paint dot did not. *sigh*

I tend to put paint marks on fasteners on say the latest project car [78 vette] not to emulate factory markings but to indicate to myself that the task is done, and done right.

joe_padavano

When we do torque striping, we actually us a paint pen with thick paint. The stripe goes from the fastener onto the part being held (and if it's a nut, the stripe also runs up onto the exposed bolt threads). Any movement of the fastener is immediately obvious. Of course, this is special low-outgassing paint for space use. You don't want to know what it costs.

Koda

We don't paint mark fasteners on the cars going out the door, but every piece of equipment with bolts has ITS bolts paint marked in the way Joe describes.

Seff

Joe: Wow, no joke on your LinkedIn about working with rockets. Am I the only one that sees the fun in you being a Rocket Scientist?

As for the technology, somehow it makes sense for very low-tech solutions to be used on space-faring vehicles - if they have to be fixed "up there", it's with low-tech tools, I assume?