Socket head cap screw strength 2

[horstr1]

I have an application in development that uses 1.375-12 socket head cap screws. The calculated maximum load for the beast puts a stress of about 119 ksi on each one of these. (The estimated load may be high, I plan to strain gauge the beast and find out what the real load is.) Since nominal yield of a grade 8 is 120 ksi, I've been trying to research high-strength bolts. I've come across a couple of places now that seem to be saying that SHCS's are normally higher strength than grade 8. I'm getting confused.

This is a really critical application, and this bolted assembly is one of the most critical parts of it. Of course, we're going to prototype and test the beast very thoroughly, but I want to be testing my other components without the distraction of these bolts turning out to be the weak link.

So is it true that these SHCS's are already high-strength? Anyone have definite data on the strength? Do I need to specify a certain type?

I'm also going to have to make some custom shoulder screws that will go in some of the same holes and see the same loading. I'm thinking 17-4PH looks good for those, does anyone have any rocks to throw at that idea?

 

[TVP]

The ultimate tensile strength of Grade 8 fasteners is 150 ksi minimum according to SAE J429 or ASTM A354. Socket Head Cap Screws have UTS > 170 ksi according to ASTM A574 for sizes above ½ inch. Here is a quick reference on the subject:

http://www.fastenal.com/content/feds/pdf/Mechanical Properties of Inch Fasteners.pdf

Regarding the shoulder screws, do you mean a standard screw/bolt such as those produced by UNBRAKO, Holo-Krome, etc.? If so, why use a PH stainless steel instead of the same steel used for SHCS (medium carbon alloy steel, quenched and tempered to > 170 ksi UTS)?

http://www.unbrako.com/standards.htm

 

[horstr1]

Thanks for the references, TVP, that's exactly what I needed. The BOM (which I recently inherited) called the screw out as a grade 8, which was throwing me off. I'll make sure to specify ASTM A574 so I'll know what I'm getting.

What was bothering me was that, although the ultimate strength for grade 8 is 150 ksi, its yield is 120 ksi and I'm just a sneeze away from that. Am I wrong thinking if it yields, it's failed? Won't it at least be losing some of the preload?

The shoulder bolt is a custom design; the diameter and height of the cylindrical body are very specific, and the head is truncated to just enough to hold a spring down. The internal hex goes down inside the body of the bolt. (The original design was a SHCS through a sleeve, but I had to put in a taller spring and ran out of headroom.) But the same ASTM A574 material makes sense for these too. No need to get unduly exotic.

Thanks again, you've been very helpful.

 

[dinjin]

If this is a critical application, how are you tightening
these screws?

 

[MintJulep]

What happens when the "good-n-strong" screw gets replaced with Chinese crap by an uninformed maintainer?

 

[horstr1]

dinjin, it'll probably be an air impact most of the time. Why do you ask?

MintJulep, that's a good point. It'd be on the news if this thing failed. But half of the bolts are the shoulder screws I designed, for which there are no Brand X substitutes. Having them designed with some safety factor, as long as the other half are semi-decent bolts it should still hold.

 

[MintJulep]

Air impact gun and critical fastener do not belong together.

If there are no Brand X replacements then one of three things will happen (in order of likelihood):

1. Missing bolts will simply not be replaced.
2. The closest thing from Home Depot will be used.
3. The closest thing from Home Depot will be modified on a lathe.

As a general rule it's a bad idea to design critical things that rely on "special" fasteners.

 

[hydtools]

For critical applications 'some safety factor' and 'semi-decent bolts' are not specifications in which I would place any faith.

Do not use an air impact gun. Agree with MintJulep.

Ted

 

[dvd]

Holo-krome makes some reliable fasteners. Call them and talk to a fastener engineer.

http://www.holo-krome.com/index.html

 

[dinjin]

Thanks to Mint Julep for answering why I asked about the tightening method. Are you going to strain guage the bolts
before and then after applying the load? Recommended preload
is about 75 percent of the Yield Stress. You can use a higher value of preload depending on how you control or use a better tightening method.

 

[horstr1]

Thanks to all for the thoughtful inputs.

As engineers, we always have constraints, and they usually preclude doing what we would ideally choose. If I had my druthers, I'd use more bolts or bigger ones or both. But I've got space constraints, and I've got the available space packed full of the biggest bolts that will fit. The only variable left to work with is the strength of the bolts.

On the tightening method, I'm just being realistic. I can specify how this gets put together initially, but out in the field they're going to use whatever tools they have.

I could use load-indicating bolts to verify the preload, but in operation this is all inside a heavy steel housing. Directly measuring the load on the bolts in operation will not be practical.

If it turns out the maximum load is significantly less than the theoretical, I'll have a lot less to be concerned about. And I have very good places to strain gage the whole unit and a very good strain gauge guy to set that up. I can effectively make the whole thing a big load cell, and calibrate it with well-quantified known loads. I will know with high confidence what my loading is before the total design is finalized.

Ultimately there will be extensive testing of the final product as specified by industry standards, and additional tests for our own comfort.

 

[ricklts]

Horstr1,

My misspent youth was 25 years in the bolting service and tooling business. I have seen several applications similar in concept to what I think you are proposing. At the very least, I would suggest that you specify for field installation and maintenance the following:

1. Use hardened washers under the head of your SHCS's to reduce friction and increase bearing area (make sure there is no more than 0.63" clearance in the hole),

2. Ensure that the SHCS does not bottom out in a tapped hole,

3. Service dependent, specify the best lubricant you can use, but because these are 12UNF threads, too much can be a detriment,

4. If nothing else, specify that a calibrated torque tool - manual, pneumatic, hydraulic, or electric be used, and that some means of limiting tool-induced angularity/side loads is employed.

5. Make yourself a test block, with umpty whatever 1.375-12 tapped holes, and use this to develop your installation procedures/torque values.

hope this rambling helps. good luck

 

[Duwe6]

Agree -- you have to put an Installation Torque value in the literature shipping with your part. Yes, the mechanic may ignore the torque value, but "shame on him". He was given the correct information.

 

[ornerynorsk]

Going to a metric class 12.9 ANSI/ASME B18.3.1M bumps you up to 176,000+ psi. If your bolts are stressed to 119,000, that is still not a comfortable safety margin.

C300 Maraging steel will put you into some higher cost with custom making the hardware, but the strength is phenomenal.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.