Remaining threaded of bolt flange 2

[dursosono]

What the minimum threaded have to remain on outer side of bolt-nut for studded flange? In which code/standard this requirement have been stated?

 

[racookpe1978]

The "requirement" is a physical, technical issue of the stretch and the ultimate developed strength of the material in the last few fractions of an inch (or millimeters) in the stud itself. (The nut will almost never fail - it is the stud (or bolt) that fails.)

1-1/2 thread to 2 threads extending past the nut (or threaded flange) is needed to develop full strength of the bolt. Extending more than 3 or 4 threads adds nothing to the failure point.

As a rule of thumb, if you have a problem bolt (and the real problem might be a supervisor or inspector or mechanic or project manager who doesn't want to fix the problem correctly) estimate the reduced yield strength by percentage: Add 3 threads to the physical nut length - this is how many threads the stud should be engaged across. Figure the percent of the actual engaged length to the required length. That is (roughly) how much less strength is actually resent.

Then go fix the project manager. 8<)

 

[doct9960]

dursosono,

For process piping, paragraph 335.2.3 of ASME B31.3 states...
"Bolts should extend completely through their nuts. Any which fail to do so are considered acceptably engaged if the lack of complete engagement is not more than one thread."

I don't have the power piping code ASME B31.1 at hand, but I guess it also had the same requirements as B31.3

 

[racookpe1978]

Doct9960: Thank you for the reference.

 

[CodeRef]

racookpe1978,

1-1/2 thread to 2 threads extending past the nut (or threaded flange) is needed to develop full strength of the bolt.

Could you please expand on this? Particularly, I would like to know the reference document if available.

I have discovered that there are many different opinions regarding thread engagement when using Nuts vs Flanged Assy's with tapped holes.
As an example, see the following paragraphs I parsed from ASME B16.34.

6.2.2 Flanged Ends.....................................
When required, valve end flanges may be furnished
with tapped holes for engaging flange bolting. Thread
engagement in a flange assembly with tapped holes shall
provide full effective thread engagement, not including
the chamfered thread, for a length at least equal to the
nominal diameter of the bolt thread. For additional considerations,
see para. 6.4.3.


6.4.3 Additional Considerations. Bolting or threading
in excess of the minimum requirements of this Standard
may be required because of, for example, valve design,
special gasket compression requirements, special specified
service conditions, or operation at high temperatures
where differences in the creep characteristics
between body and bolting materials could compromise
joint sealing capability. Since these factors vary widely,
these requirements must be accounted for by individual
manufacturers.

With the above stated, what would be your opinion of a Flanged Assy. where thread engagement was "of a length at least equal to the nominal diameter of the bolt thread" and the additional considerations of 6.4.3 are satisfied BUT the thread engagement does not extend through the tapped hole?

 

[doct9960]

Okay, I've got ASME B31.1 with me. The requirement is different with ASME B31.3. ASME B31.1 requires that "bolts and bolt studs shall extend completely through the nuts" (no exceptions like that in ASME B31.3). See 108.5.1(A)

UG-13(a) of ASME Section VIII, Div.1 requires that nuts engage the threads for the full depth of the nut. In my opinion, this is the minimum requirement on all vessel nozzles with bolted joints regardless of whether an ASME B31.3 piping is connected to it.

Also para. 8.2.1 of ASME PCC-1 (Guidelines for Pressure Boundary Bolted Flange Joint Assembly) has this to say...
"Corrosion of excess threads can hinder joint disassembly. A practice that facilitates joint disassembly is to fully engage the nut on one end (no bolt projections beyond the nut) so that all excess threads are located on the opposite end; the excess threads should not project more than 13 mm (1/2") beyond the nut, unless required for the use of hydraulic bolt tensioners."

 

[CodeRef]

Okay, so the engagement of NUTS is settled, now, how about discussing thread engagement for tapped holes that are provided for studs?

ASME Section I (High Pressure Boilers) = PG-39.4
ASME Section VIII-1 (Unfired Vessls) = UG-43(g)
ASME B16.34 Valves (Flanged) = Paragraph 6.2.2

Each of these references above state that thread engagement would be adequate for a length at least equal to the nominal diameter of the bolt thread.
Thus, full body/tapped hole engagement is not a criteria for acceptance.

Does anyone disagree?

 

[racookpe1978]

1. Nuts (by standardized and MILSPEC/MS/NAS catalog sizes) are slightly shorter than the bolt diameter. Bolt heads (being universely forged with the bolt shaft) are 82% - 90% the length of their mating nut since they are inherently stronger that a nut - which has cut threads. (As strength requirements increase (the higher the grade of the nut & bolt), the longer the nut usually is.

When a bolt or threaded rod extends through the nut by 1-1/2 to 2 threads, the length from bottom of nut to top of threaded end ends up right at the full diameter of the original threaded shaft.

Remember, you have got to design for the field to assemble, take apart, and re-assemble these fasteners - and the field mechanics and millwrights will not have access to a manual or ASME spec. for them, a conservative and easy-to-remember thumb rule is safer as a design criteria than a chapter and verse in a book they can't find. And don't want to read.

2. Tapped holes are a little different: In a forged or machined part like a turbine cassing or pump or valve, it is very, very likely that many different depths are going to need to be drilled and tapped through - all of the same diameter stud and the same nut and washer. What will differ between these tapped holes is the depth of the metal being drilled into.

For manufactoring and repair, a drilled-through speeds up fabrication: coolant runs through, all holes can be drilled straight through (manually or cnc) and all threads can be tapped right through the metal into "air".

If you specify many different hole depths, each time the chips and coolant need to be vacuumed out, wiped out or soaked out, blown clear, etc. That's hard, slow, and expensive work - and often still leaves residue in the tapped hole. Tapped threads can't get a grip in the bottom of the hole anyway - unlike a through hole where the tap can be run all the way to good threads all the way through. So through holes (hole that are apparently way too long for their purpose, are often cheaper and better than closed-end tapped dirt collectors (er, holes).

So, there needs to be some uniform criteria (thumb rule) that allows for varying depth tapped holes: Hence the requirement that a tapped hole must be engaged at least one full diameter of the stud. Easy to remember, easy to apply and check.

And the one full diameter for tapped holes, as you see above, is compatible in form, fit, and function with the "2 threads past the nut" thumb rule I used above for nuts.

 

[CodeRef]

racookpe1978,
Thanks for the detailed explanation.
I particularly like the way you addressed the field assembly thumbrule and how "through bolt hole" fabrication is economically based.
As an inservice inspector, it is important to recognize the fundamentals behind the construction and assembly of threaded connections for tapped holes and nuts.

I was involved in a situation where another inspector was of the opinion that full "through body" engagement was required and a non-conformance was issued based upon the examiner not recognizing this condition for the tapped hole. Of course, engineering performed analysis and accepted the condition upon evaluation.
The question remained as to whether every similar attachment located throughout the plant required evaluation when studs did not extend through the full body of the attachment.
The resulting conclusion was that when lack of full thread engagement is identified in ANY case, an engineering evaluation will need to be conducted.
I guess being overly conservative has it's advantages.