I was thinking that you would need a hell of a lot of heat to get Inconel to elongate enough such that the strain would produce the needed tension. You'd have to do them all at once, and control the cooling.

How does Inconel's elastic modulus compare to the coefficient of thermal expansion?

So how does this bolt heating process work? You heat the bolt to cause a change in length based on the coefficient of thermal expansion, then install and snug (when hot) at which point the cooling reduces the length and induces the preload?

One nuance in our case (besides the difficulties of actually trying to heat the bolts as I don't know how we'd do that) is that when our two parts are mated up, they are not contacting - they are spaced apart by the soft iron seal that acts as the gasket. The act of torquing brings the upper piece (the target) down and the knife edges in the lower surface of the target actually "dig in" to the iron. To control the amount of penetration, the target then comes into contact with the carriage (that it is bolted onto) and at that point any additional torque is just preloading the bolts and not inducing any additional penetration of the knife edges into the iron.

In our case, I'm not sure how well the bolt heating method would work.

We're currently leaning towards an electric torque wrench. Adjusting the torque is the biggest challenge...

That sounds like way too much elongation for heating Inconel. How long are the bolts?

The bolts need a hollow hole drilled lengthwise in the center. The bolt heater is slipped in and electricty applied. After a calulated time, the bolt will reach full temperature and elongate. You then snug the bolts, basically "hand tight", then turn off the juice and pull the heaters.

Flieger may be right, Inconel may not stretch as much as you need for this application. Someone from a bolt heating company will need to look at your application.

Turbine Maintenance Tools & Industrial Supplies - Welcome to Hesse Equipment Company

Talk to these guys. I think they will be able to run the calculations.

OK, now I see. Your preload of the bolts is not really important? That is, there is not going to be significant pressure or loading on the bolts, they are just there to maintain seal integrity?

Essentially this is correct. The bolts are 1.25-7UNC x 21.625 lg Inconel 718. Operationally, it is important that the knife edges penetrate the soft iron seal in a controlled manner (i.e. the preload on the bolts must be evenly distributed during the 5 torque cycles), but ultimately the proof is in the seal performance. External loads in service are negligible. Eight 1.25" bolts at 900 ft-lbs are generating quite a bit of clamping force, so this thing's not going anywhere. :)

have you tried to get a hold of Jim Sims?

also, one more crazy idea -- any way to design a "clutch" that would give (yield) at the proper torque?

Well, it appears that companies make clutches like that for this exact reason, but I would need 5 different clutches and the ability to change them out (or adjust them) remotely. I have space limitations that would also probably preclude using them (or anything else much larger than a socket).

Contact one of the clutch makers and see if they will make you a multiclutch setup? Ala PDK?

So if you had one of those multi-spindle deals like they use on automotive assembly lines then you could just torque them to full tension in one go.

Mike -

What about custom shear-head bolts? Picture a bolt with 5 stacked shear heads, each of which is calibrated to shear off at each of your specified torque values. Smallest head on top (for the 200 ft/lb) progressing to the largest which shears at the 900 ft/lb value.

Because each head is slightly larger than the one above it, you should have no problem getting the wrong torque value, it would be immediately obvious if any bolt had not yet been torqued, everything is pre-calibrated, and it's relatively simple.

Only downsides I see are (a) bolts would probably be custom and therefore spendy - but probably irrelevant compared to the overall container cost; and (b) the driver would have to change out sockets for each torquing round.

Interesting project!
Jim

The shear head bolt idea is cool, but I don't think that would work too well for us. We really can't change sockets in the cell - in fact we use the same size hex on almost everything so we can use the same sockets. The sockets are essentially permanently attached to the wrenches. If we were to drop a socket (or anything else), it could conceivably shut the facility down.

Any loose pieces created during the torquing/installation would be undesirable. We dropped a washer (a big one for one of these 1.25" bolts" and spent hours looking with cameras to ensure it wasn't going to come back to haunt us. We're actually looking at a design change to eliminate the washers. Heck, we dropped a bottle of Snoop during an ill-fated attempt to troubleshoot a leak in a Hiltap fitting and never found the bottle!!

Probably a dumb question but...

Why is this being done inside the chamber? Are you placing something radioactive inside and then closing it up?
If not, do the assembly and sealing/torquing outside in a clean room and then transfer the whole thing inside the chamber.

But then they'd need a new assembly every time they do a test.

I thought it may have been a new assembly every time.

The OP said once complete this thing gets... "inserted ~30 feet into a core vessel on a 200,000 lb cart so access is impossible"

http://forums.pelicanparts.com/6606029-post33.html

You need only attach this simple device in the appropriate manner and your problem is solved.

http://forums.pelicanparts.com/uploa...1331168409.jpg

ok, new crazy idea...

is it feasible to eliminate bolts from the design?

some sort of press fit with a "slip cover" over it to wedge every thing together (i.e. unwrap the inclined plane that is a bolt...)

Just wondering what apparatus you use to determine the torque that need to be applied?

Also what degree of accuracy is needed?

Skidmore makes some very robust and accurate tension measuring devices, this will allow you to know how many rotations from snug you need to turn a bolt to reach design tension... Often this is read as torque on a wrench when the bolt is in motion..But you know this already...

You can put witness marks on the fixed part, and another witness mark or series of witness marks on the chuck and you can know with a high degree of accuracy how much you have turned the bolt (nut)

You can probably tell I am a big fan of using ganged hydraulic jacks to tension fasteners. We have tensioned cable saddle bolts 300 feet in the air on a bridge tower.... and we had more access than you appear to have...

Interesting problem...and I will definitely be thinking of solutions

Umm...I hate to say it but that lower right bolt is missing a washer. What did I win?;)

Washers integrated into the bolts seem better.
I like rwebbs idea of not using bolts at all, but rather another repeatable clamping system.

The chamber is designed to contain high pressure gasses and be reusable(?)
Not an engineer, but:
-perhaps the chamber is too thin and warping during heating?
-perhaps the tourque proceedures around the hollow circle are too rapid?
-perhaps more bolts are needed to maintain contacting surfaces?
-perhaps a secondary intergrated seal along the perimeter?

You said that currently you were attempting to control the torque output of the wrench by varying the air pressure input, but due to unquantified pressure drops in the lines between the air supply and the wrench and degradation of the air pressure/torque curve of the wrench itself due to being operated without lubrication you were not confident in the results.

Why not measure the pressure at the wrench head? Could you get a solid film lubricant coating applied to the wrench bearing/wear surfaces to slow down wear or negate the lack of lubrication?