Allowed angular slop on UNC thread?

[Jabberwocky]

Kind of a weird question, but does anybody have any info on the allowable or resultant angular mis-match on a standard UNC thread?

I realize the threadform standard is more concerned with diameter fits, and that my question is heavily dependent on the length of the screw - let's consider an arbitrarily long bolt with a nut that's got two full turns of pitch. What is the maximum angle that this interface will still mate?

My particular question is on a 1/4-20 UNC (2B) but if there is a more general source it'd be great.

 

[BiPolarMoment]

I don't have a specific answer but since the assembly angle effectively changes the "perceived" pitch of the screw at the interface I'd wager it's "not much". Unless you will accept a cross-threaded part in which case it's a bit more.

If you're trying to ensure that two things mate more consistently perhaps using a locating feature like a male/female taper on the parts is more appropriate.

 

[moon161]

Depends how many threads will be engaged on the mating part.
Assuming the thread profile is correct, you can look at how much linear difference (delta) there is between 2 fit conditions:

1) The thread flanks facing the head of the bolt are in contact.
2) The thread flanks facing the end of the bolt are in contact.

The lead of the thread is 1/pitch. If you know how many turns you want to engage , delta give you the allowable deviation from specified lead (dl=lead specified - lead deviated). delta = n*dl , or delta/n = dl

The lead comes directly from the angle and nominal diameter-
l=pi*d*tan theta, and dl = 1/pi*d( 1/theta spec. - 1/theta deviated)

so:
delta*pi*d/n = (1/theta spec. - 1/theta dev.)

Back to work.

 

[moon161]

oops:
The lead comes directly from the angle and nominal diameter-
l=pi*d*tan theta, and dl = 1/pi*d( 1/tan theta spec. - 1/tan theta deviated)

so:
delta*pi*d/n = (1/tan theta spec. - 1/tan theta dev.)

 

[unclesyd]

Aside from the above information there are other contributing factors to axial fit. The class of thread will play a role in the makeup clearance. Another contributing factor is the percent of thread used Also the type of thread will play a roll.
Here is set of tables from Engineered Parts that gives you quite a bit of information on small diameter fasteners. The differences are not as pronounced in small fasteners as they are in 1/2" and above.
We used the percent of thread on as an aid in the makeup of SS fasteners to the point where you had axial play even with a fully threaded nut.
In the larger diameters, 3" and above there are other factors o consider like relaxation after tightening and what is called the "Virgin Bolt Phenomena" which actually affect the integrity of the bolt connection.

http://www.engineeredpartsinc.com/

 

[Tmoose]

"What is the maximum angle that this interface will still mate?

Mate,like, can still be assembled?
I'd probably buy some reasonably high quality parts ( a few long socket head capscrews and grade 8 nuts) machine the nut to the 0.1 inch thickness and measure the angle of wiggle with light pressure applied every 45 degrees. I'm not sure I'd expect them to assemble all that well when offered up at maximum anhgle.

 

[Cockroach]

I haven't seen a direct answer to the question. Attached is the geometry for the thread required, thus answering the inquiry.

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada

 

[Cockroach]

The attachment.

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada

 

[Jabberwocky]

Thanks all for the input - as usual the problem is a bit more complicated than the standards allow since we are creating our own male thread that needs to fit within the 1/4-20 female thread.

It appears that even at the same diameters, if you start to deviate from the standard 30 degree threadform then your slop goes up considerably, which can actually be beneficial since that lets your screw get started when the angle isn't perfect (in lieu of other mating geometry) And I don't even want to get into the material question since we're using (relatively) soft metal which I know can be be forced to engage just by deforming slightly (and permanently (it's not a bug, it's a feature!))

 

[Cockroach]

I am really puzzled as to why you would want to deviate flank angle from 60 degree inclusive. Are you grinding your own cutting tool on a grinder?

Typically a titanium carbide insert or other tool steel alternative is bought from a supply house and placed in the holder. The insert has a standard thread form and the better ones are shaped to remove the rag edge from flanks of threading.

Essentially changing the angles and pushing the envelope of acceptability relative to the standard is bastardizing on a thread. I would question interchangability with future designs should you wish to continue. On the other hand you may be wishing to accept this as a means of controlling others from copying a particular design or using yours for interchangability with another product line. Either way, there are better ways on handling protection on a product line.

So I'm unsure of why you want flank angle deviation even if it is a result of very poor quality machining. Just my position as a PE.

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada

 

[unclesyd]

You don't have to worry about alignment if you use the Mathread. These work at up to 15° angle of attack. Check out the video.

http://www.engineeredpartsinc.com/mathead.htm

 

[Jabberwocky]

@Cockroach - the shortest answer to your question is that the human body does not contain any Unified series threaded holes (nor Metric, last I checked).

Neat find on the Mathreads, it's interesting that their lead-in round threads don't even have to taper up - although their thread cutter for those first turns has got to look odd.