Tighten nut on clamp T-bolt "snug": What the heck is "snug". 4

[WKTaylor]

We use the term "sung" for installing parts "tight-but-not-too-tight". Most of us have a "feel" for what snug is.

HOWEVER, recently, several of us got together to define "tighten-snug" for a 0.250-28 plain nut. None of us had a clear definition for the mechanic who wants to now what "snug-torque" is... or what a snug fit of parts is.

How do You define the term "snug" as related to torque and mating-part fit???

Regards, Wil Taylor

Trust - But Verify!

We believe to be true what we prefer to be true.

For those who believe, no proof is required; for those who cannot believe, no proof is possible.

Unfortunately, in science what You 'believe' is irrelevant – "Orion"

 

[MikeHalloran]

This has been discussed in the Structural forums, where 'snug-tight' means, if I recall correctly, a 250# ironworker hanging on the end of a spud wrench. I don't find that definition appealing, but in the structural steel world, that's what it is.

For me, with a little experience in AGE and automobiles, it means 'press your thumb on the end of the wrench', which equates to about 20# x the wrench length. My Dad said something that calculated out to 'a little less than yield', after threatening to cut all my wrenches in half. Later, I developed a 'feel' that would put it at a little less than that, i.e. just as the joint 'tightens up'. Now, I'd put it as about half of the yield torque, with Never-Seez on the threads, and maybe just a film on the nut face.





Mike Halloran
Pembroke Pines, FL, USA

 

[rb1957]

remember that picture "who decides how much to tighten a bolt ?" ... the analyst or the mechanic ...

Quando Omni Flunkus Moritati

 

[CoryPad]

According to Guide to Design Criteria for Bolted and Riveted Joints by Kulak, Fisher and Struik:

The snug condition was defined as the point at which the wrench started to impact. This occurred when the turning of the nut was resisted by friction between the face of the nut and the surface of the steel. Snug-tightening the bolts induces small clamping forces in the bolts. In general, at the snug-tight condition the bolt clamping forces can vary considerably because elongations are still within the elastic range.

For bolts equal to or greater than about ¾ in. dia., snug position provided by an impact wrench is approximately equal to the tightness attained by the full effort of a man using an ordinary spud wrench. For longer or larger diameter bolts, the force produced by this snug load will be less than that for the “standard” case, and for shorter or smaller diameter bolts it will be more. These differences are accommodated in the specification by prescribing the same definition of snug tight for all cases but varying the degree of rotation required beyond snug for different situations.

According to Specification for Structural Joints Using High-Strength Bolts by the Research Council on Structural Connections:

Snug-Tightened Joint. A joint in which the bolts have been installed in accordance with Section 8.1. Snug tight is the condition that exists when all of the plies in a connection have been pulled into firm contact by the bolts in the joint and all of the bolts in the joint have been tightened sufficiently to prevent the removal of the nuts without the use of a wrench.

8.1 Commentary: The snug tightened condition is typically achieved with a few impacts of an impact wrench, application of an electric torque wrench until the wrench begins to slow or the full effort of a worker on an ordinary spud wrench. More than one cycle through the bolt pattern may be required to achieve the snug-tightened joint.

 

[MikeHalloran]

Oh, wait.
There is a spec for t-bolt nuts: 50 in-lb., no more.


Mike Halloran
Pembroke Pines, FL, USA

 

[tbuelna]

In the aerospace world, I have never seen plain nuts used to secure the T-bolts on band clamps. The typical procedure for installing the clamp would be to first measure the running torque of the lock nut to verify that its locking feature is still within spec. And then the lock nut is torqued to a value above the measured running torque.

I would never use subjective terms like "snug" when describing a critical assembly process. Such written process descriptions should always be specific, and should never leave anything open to interpretation.

 

[MikeHalloran]

Allow me to clarify my experience with 1/4-28 T-Bolt clamps.

In the marine exhaust industry, we use two types; V-band and flat (hose) clamps, all sourced from Voss or similar outfits, all with stainless bolts and all-metal self-locking nuts (the kind that look castellated), usually with some kind of anti-gall coating on the nuts.

If you don't put Never-Seez on the bolts, they'll gall before you ever get them tight. With the Never-Seez, by the time you get them run down to start tightening the joint, most of the prevailing torque is gone anyway.

The clamps with flat bands are used on silicone hose; going tighter than 50 in-lb cuts the hose, which is hideously expensive, and doesn't make the hose seal better.

The clamps with V-bands are used to secure beveled flange pairs, which are sealed with water glass, and manually brought into secure contact before the clamps are tightened. Using the clamps to move the (stiff) pipes just distorts the clamps and guarantees that the joint will leak. 50 in-lb is enough to securely tighten the clamp on a properly made up joint, and more torque doesn't help an improperly made up joint.

Your mileage may vary.
Don't believe me; call Voss (216) 771-7655.
Also see vossind.com



Mike Halloran
Pembroke Pines, FL, USA

 

[GregLocock]

Snug is way short of fully tensioned, in my workshop. It'd be hard enough to grip, but not necessarily developing much tension. So for example, you'd snug the cylinder head bolts up before doing the tightening pattern.

Indefinable, therefore unacceptable, yet obvious.

You guys seem to be going for more torque than that.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[WKTaylor]

All..

In this particular situation I have a very unusual band-clamp installation. See attached "scrubbed drawing".

The T-Bolt [A286 130-KSI HT] is secured with an MS9360-10 [0.250-28 UNJF-3A] plain hex nut, A286 silver plated. The Bleed Air Duct [Inconel-625, 3"OD X 0.035-WT] can experience -65F to +820F + 45-PSI thermal/pressure loads/strain

Per the drawing NOTE, the T-Bolt nut is to be installed as follows...

TORQUE REQMT: DO NOT TORQUE TRUNNION NUT. TIGHTEN NUT FOR SNUG FIT. Note: after "tightening to a snug-fit", the nut is then lock-wired to the clamp loop [holes drilled-thru the hex-corners].

Unfortunately these instructions were "lost in translation", when the MM was written: we now are experiencing crushed ducts due to application of "nominal" torque for a 1/4-28 nut [65--85-in#].

Here is what I wrote to define "snug"; however it was not considered acceptable to the MM owner because they are committed to have torque values listed and cannot allow not amorphous torque procedures like this [military] in their MM.

b. Install xxxxxxx duct support clamps (6) on bracket
(1) with bolts (7), washer (8), and nuts (9). Torque
mounting nuts (9) 65 to 85 inch-pounds. Install T-Bolt
Nut (6a) on threads and run it down until it is “snug
and lockwire the nut to adjacent clamp loop per NASM33540.

SNUG Nut (6a) as follows.
Use a screw-driver with a deep-socket bit or a short
handle box wrench to tighten Nut (6a) "snug" as follows.
Run the nut down on the T-bolt threads. As soon as You
feel the plain Nut transition from smooth run-on torque
to a noticeable torque-rise, then stop: the nut is
SNUG! Inspect each clamp installation to ensure the
duct wall under the clamp is round and smooth (does NOT
have a “pinched/wrinkled/wavy” appearance). IF Duct wall
is distorted, then loosen Nut and gently "re-snug" it.

Here is how they ended-up word-smithing para c [approximately]...

b. Install xxxxxxx duct support clamps (6) on bracket
(1) with bolts (7), washer (8), and nuts (9). Torque
mounting nuts (9) 65 to 85 inch-pounds. Torque T-Bolt
Nut (6a) 5--15 inch-pounds and lockwire the nut to
adjacent clamp loop per NASM33540. Inspect each clamp
installation to ensure the duct wall under the clamp is
round and smooth. IF Duct wall is distorted, then loosen
Nut (6a) and gently re-torque to low end of range.

Comments?

Regards, Wil Taylor

Trust - But Verify!

We believe to be true what we prefer to be true.

For those who believe, no proof is required; for those who cannot believe, no proof is possible.

Unfortunately, in science what You 'believe' is irrelevant – "Orion"

 

[MikeHalloran]

I should note that hot FRP becomes plastic, so we made 'crush sleeves' to be bonded inside FRP muffler nipples, to resist the stress applied by the hose clamps.

If the duct you're trying to retain is made of, er, foil, maybe it needs internal reinforcement at the clamp plane.

The redacted drawing suggests that maybe you're using the clamps to press a round thin duct against, e.g. a rib or stringer or other not strongly curved object. If the tube's entire OD is not supported either internally or externally, then it's liable to crushing at even moderate clamp loads.

Even if there is some kind of saddle not shown, the stress situation on the OD of the tube is not uniform. You draw the clamp leaves nicely curved, but when you apply tension, they try real hard to straighten out, again leaving the tube liable to distortion. You could specify a full circle clamp, and equip it with an ear to be attached to the structure. Talk to Voss about options; this is not their first rodeo.




Mike Halloran
Pembroke Pines, FL, USA

 

[drawoh]

wktaylor,

I do not think your concept of "snug" is meaningful. I have been trying to work out how to install screws finger tight, and rotate them some specified angle, to my required clamping force. For most smaller screws, the transition from finger tight to the yield stress of the screw is only a few degrees. I rarely use screws over 3/8" or 10mm diameter.

For this to work, I need an extremely precise concept of "finger tight". I think your concept of "snug" is even more adventurous. Can't you estimate the force you want, and work out the screw torque? For my calculations, I assume an infinitely rigid joint. In your case, the joint is flexible, and screw turns from finger tight might work.

--
JHG

 

[verymadmac]

I would suggest a slight variation on what drawoh said

One could approach it form a strain point of view, something along the lines of "tighten till the clamp is no longer loose to the touch, then apply 1 to 2 more turns of the nut".

 

[MikeHalloran]

Given that you're using lockwire on plain drilled nuts, a torque spec may be a perfectly reasonable and repeatable way to go, but ordinary torque wrenches are not the correct tools for 'snug' torque ranges.

Get a 'torque watch' in a couple of ranges, fitted with a 1/4" square drive adapter, and use them to establish a range that is acceptable to you.

Bring your wallet...



Mike Halloran
Pembroke Pines, FL, USA

 

[drawoh]

I would suggest a slight variation on what drawoh said

One could approach it form a strain point of view, something along the lines of "tighten till the clamp is no longer loose to the touch, then apply 1 to 2 more turns of the nut".

According to my spreadsheet, an M5X0.8 A2-70 stainless steel screw clamping through 35mm, reaches its yield point at about a 37[°][ ]turn. I am assuming that the joint is very much more rigid than the screw.

Question: If you, MikeHalloran, GregLocock, wktaylor and I each install a screw finger tight, what will the variation be in degrees?

My strategy for accurate torquing of screws within yield strength is hopeless. If I had longer, higher strength screws, clamping something with significant and known flexibility, my system could work.

If I trust everybody to systematically use each screw once only, a full turn of the screw keeps things well below the ultimate stress of an austenitic stainless steel screw. This does not solve the OP's problem.

--
JHG

 

[MikeHalloran]

From Wil's last post, I infer that the bleed air duct is very thin and easily distorted, which suggests that the torque increase per degree will be rather gentle, which in turn suggests that 'turn of nut' might be viable, provided once again that 'snug' can be defined in some reproducible way.

So, let's back up a little.
'NOT snug' could be defined as 'loose enough so the parts can be wiggled a little'. So for purposes of this discussion, 'snug' could be locally defined as "just tight enough to the parts don't wiggle". ... and the design team, privy to the redacted details about the duct, could take it from there.


I personally would try to redesign the assembly so that a one-piece clamp just tightens around the duct, and then the clamp is secured to the aircraft by means of ears built into the clamp. I.e., I think the trunnion style clamp and the implied saddle that I hope is there and their nonidealities conspire to make it too easy to deform the duct while attempting to secure the clamp to the duct. Or just weld ears to the duct, making them long enough and flexible enough to limit heat transfer and duct distortion to the airframe.






Mike Halloran
Pembroke Pines, FL, USA

 

[GregLocock]

There's a huge issue with grease/dirt on the threads if you are attempting to define a torque. As an example we had one design where a 30mm approx nut had to be torqued to approximately 1.4 Nm (about what you could apply by hand between thumb and finger) which was easy to do as a one off, because you could feel if there was any problem. But, we needed one every 30 seconds so we had a machine. It was hopeless.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[tbuelna]

wktaylor-

Thanks for the detailed description of your problem. Makes it much easier to suggest a possible solution.

I think MikeHalloran's suggestion for tightening the nut using an angle value rather than a torque value is the way to go. You can place a .009" thick feeler gauge between the mating faces of the nut and clamp lug, and then finger tighten the nut until you feel some sliding resistance on the feeler gauge. This .009" axial clearance is equal to 1/4 turn of the .250-28 nut. So all that is left is to tighten the nut another 1/4 turn for contact, plus some small additional amount (maybe another 1/8 turn) to ensure the nut is firmly seated.

This should produce the "snug" condition you desire, it only requires a wrench and a feeler gauge, it is a fairly simple procedure for the mechanic to follow, and it eliminates variables like thread frictions.

Hope that helps.
Terry

 

[dik]

I would think with the aerospace or aeroplane industry that snug would be defined in the specifications for a fastener. With construction and structural steel, there are so many fasteners and a relatively 'crude' environment for installation, that a rough identification of tightening has to be understood... Snug is considered to be the maximum effort on a 'spud' wrench... a standard tightening wrench that increases in length depending on the fastener size. This is a similar effort as when an impact wrench first starts to impact. It is a rough measure of tension that will prevent a nut from loosening by itself. It is out of place with the aerospace industry...

Dik

 

[tbuelna]

dik-

The modern aerospace industry requires specific, well-defined procedures for everything it does, including design, manufacturing, procurement, testing, assembly, maintenance, etc. Thus, subjective terms like "snug" are normally not acceptable in written requirements or procedures. Requirements and procedures are always written so that there is nothing left open to interpretation, and also that there are clearly defined methods of validating the procedure was performed correctly.

 

[MikeHalloran]

The modern auto industry takes a slightly different tack, because it can't enforce the procedures it publishes, but still has most global governments and all the trial lawyers in the world peeking over its shoulder.

Its response comprises things like fluid couplings that snap together with an audible click, with primary retention provided by a garter spring and sealing provided separately by an o-ring, and with secondary retention provided by a spring clip, all assembled without tools or torque specs. They're also lightweight enough to deserve investigation by the aero crowd. I'm not sure that aero alloys would do well with the extreme forming required to make the ends, but it seems worth a look.

Have you looked at an auto OEM radiator hose recently? No more worm-drive clamps, no expensive t-bolt clamps, but more likely a spring clamp that requires a tool to open, and springs into place when released, with applied forces controlled by design and dimension, not procedure. I think I've even seen shrink-tube in that application.

I know that redesign or replacement of the t-bolt clamp is not within the purview of the original question, but maybe it deserves a sticky note somewhere.


Mike Halloran
Pembroke Pines, FL, USA