Bolt Heating Procedure 2

[CrazyJ]

We're preparing to assemble two bolted joints, which are drilled for bolt heaters (induction, or other). The manufacturer is unable/unwilling to give us procedures to assemble them, other than to tell us the required stretch to target.

My initial thought is that all the bolts should be snug-tight (to take up joint slack), measurement made of the initial length with a micrometer/depth gauge, then heated, and nut rotated the amount required. They've given us the rotation angle, which is larger than just the rotation angle x pitch /360, so I assume this is to account for deflection of threads, etc.

Does anyone have experience in this procedure that could tell me if the above is correct? I don't think that countless iterations are particularly good for the properties of the bolt, and would prefer to minimize wasted time.

 

[DLite30]

Are you sure they're drilled specifically for a heating element?

Sounds to me like you have a bolt that's drilled for to accept a depth micrometer to measure bolt stretch. And that you have a rotation angle, and yes the rotation angle will be more than you calculate due to slight change in dimension of the nut/bolt threads during tightening.

Sounds like you already have all the information you need.

However, there is an equation in Machinery's Handbook you could use for the heating method, if you know the material properties.

T= S/(E * e) + To
S=bolt tensile stress target value (psi)
E=bolt modulus of elasticity (psi)
e=bolt coefficient of linear expansion (in/in/F)
To=normal operating temperature of the bolt (F)
T=targeting heating temperature (F)

 

[metengr]

Your description of obtaining the desired bolt stretch is very similar to how our large steam turbine horizontal joint shell studs are installed. The center section of each stud is drilled to accomodate a rod heater. The studs are measured in temperature and when reached, the nuts are tightened. As the stud cools, the necessary stretch is obtained.

 

[unclesyd]

You might want to look at the products from Superbolt to replace the method you are proposing. We have almost completely changed to Supernuts on our three turbines and the larger split case horizontal pumps.
The biggest problem we had was the installation procedure require that the nuts be seated using a hammer wrench.
We had several near hits when a missed strike broke the hammer handle sending the head flying.

http://superbolt.com/index.php

 

[byrdj]

This is a vary common method used for Steam turbine bolting

the studs probally have a hole drilled from the top to center of the lower threaded section, then a reduced hole through. this allows you to install a rod with a balled into the stud, then measure the relative amount of rod sticking out with respect to the top of the bolt. this should be done with studs cold and NO tightening of the nuts. this is a relative reading and most times a tool is made to hold a dial indicator to get this relative measurement. need to verify that the reading is repeatable.

then the bolt is snugged up, this is real critical and depends if this is a compressed gasket joint or if additional force is need to closed a warped jump. if so then then all the bolts are used to close the joint, then ONE AT A TIME is loosened, then snugged (about 2% of yeild torque), then marked for rotation.

For an 8 thread per inch stud (0.125 pitch), the rotation for stretch is calculated at 0.090/revoulution to account for thread defromation. (this is proven by over 100 years of turbine assembling)

the bolt is then heated quickly as possible and the nut turn, usually with minimun torque (knock wrench) to the nut roation marks

I have found if properly performed, the "turn of the nut" is sufficient to ensure proper stretch.

after all bolts have been strecched, the bolts are allowed to cool back to original temp (can take a day with forced ventalation) then the rod is installed and the relative reading taken, and the delta from the as found is the stretch. the desired stretch may require adjustment to either increase or decrease nut rotation

for turbine bolts, the stud active lenght is measured and the desired stretch varies from 0.0015 to 0.00075"/inch, depending upon the stud material and operating temp of the loacation.

again, for the meassurement to match the turn of the nut, the intial condition of the joint and nut preload are of the utmost importantance.

for

 

[CrazyJ]

In fact, the studs are drilled exactly as you described, which are installed into the crown of a hydro runner with nuts to couple to a mainshaft. The next coupling up the shaft is bolt/nut and has a hole straight through, as the step is not necessary since it's not a blind hole.

These are 3mm and 4mm thread pitch bolts, so I presume that the 0.090" per rev is very close for the 3mm pitch bolts, but may need to be modified for the 4mm pitch bolts.

The manufacturer has supplied a rotation value, so that should be helpful.

As the joints will remain at or near room temp (or colder) when in service, I suppose that is why the stretch values are less than I would expect from a steam turbine bolting stretch.

The cool down time is interesting. I never pondered how long that would take. Plus iteration time if they don't stretch to the target value. This could take a while.

 

[byrdj]

the cool down and iteration does present a scheduling concern for turbine reassembly. googling, trying to find some pics of the bolt extensionometer and bolt heating resulted in find of a lot of vendors offering hydrualic tensioning devices now.

hopefully the tolarance of your measured strech will not require an iteration. and the ability to repeat measurements doesn't cause a wild chase

the key though for turn of the nut is closing the joint and the intial snugup. sometimes the manufactor provides that pretorqe