About the Author/JOE GREENSLADE

Joe Greenslade is President of Greenslade & Company, Inc. in Fort Worth, Texas, a supplierof fastener inspection products, inspection software, and ISO 17025 (A2LA) accredited calibra-tion services. He also provides a variety of consulting services including serving as ExpertWitness in fastener related litigations.

Joe has been in the fastener industry in a variety of capacities since 1970. He has written over220 fastener technology related articles and has spoken to many fastener industry organizationsthroughout the United States. Joe has been awarded 12 U.S. Patents for fastener inspectiondevices.

Joe is an active member of the American Society of Mechanical Engineers (ASME), theAmerican Society for Testing and Materials (ASTM), the National Fastener Distributors Associa-tion (NFDA), and a member of the Board of Directors of the American Association for LaboratoryAccreditation (A2LA). He is a former member of the Industrial Fasteners Institute (IFI). He was amember of the Fastener Quality Act Task Force representing the fastener industry. Joe alsoserves on the Screw Thread Improvement Task Force working with government and industry toimprove aerospace related thread specifications, and thread gaging and calibration procedures.

Here Is What a Hydrogen EmbrittlementFailure Really Looks Like

T he picture of the broken zinc plated socket headcap screw in this article is a great example ofwhat a typical hydrogen embrittlement failure looks like.I thought this is would be a good example to share withreaders because the discussion of hydrogenembrittlement comes up so often, but most people inindustry do not know what a real hydrogenembrittlement failure looks like.

There are several classic in-dicators of a real hydrogenembrittlement failure. If any ofthese factors are missing, thenthe failure must be attributableto something other than hydro-gen embrittlement. These fivecharacteristics are:

C36, the greater the chance that the part will experi-ence a failure attributable to hydrogen embrittlement.Inch socket head cap screws, L-9 bolts, metric prop-erty class 12.9 screws and bolts, and spring steelwashers and roll pins are the common fasteners whichare most susceptible to hydrogen embrittlement fail-ures because of their high hardness.

3. The failed parts must be electro-plated. The cre-ation of the hydrogen in the metal fastener is causedby the cleaning processes where acids are used andnot by the actual application of the plating. I have heardof many cases where non-electroplated parts were sus-pected of hydrogen em-brittlement, but I am not awareof any that were confirmed to have failed from hydro-gen embrittlement.

4. The appearance must be that of an intergranularfailure. Look closely at the surface of the broken areasin the photograph in this article. The surface of thefailure looks relatively smooth with a texture that lookslike the surface of emery cloth. If you look at it undermagnification, you see that the surface has a crystal-line appearance with many sharp faces or facets. Somedescribe the appearance as being similar to brokenrock candy.

A bolt or screw failure caused by excessive torqueor tensile stresses produces a failure surface that haspeaks and valleys in it which is referred to as dim-pling. This is also technically referred to as a ductilefailure as opposed to a brittle failure which is charac-teristic of a hydrogen embrittlement failure. If you lookat the failure surface of a bolt or screw failure surfacethat has peaks and valleys and/or a swirling appear-ance. The failure is highly unlikely to be attributable tohydrogen embrittlement.

5. The failure location is either where the fastenershead connects to the body or in the threads within twothread pitches above where the bolts thread engagesthe mating thread.

In the case of head to shank failures, it frequentlylooks like someone used an ice cream scooper to scoopthe shank out of the underside of the heads, as is thecase in the pictured part.

1. The failure must be a DELAYED failure. The de-lay is generally from one to 24 hours after installation.If the failure occurs during installation, it is definitelyNOT caused by hydrogen embrittlement.

If the delay is a week or later after installation, thecause is probably stress corrosion and not hydrogenembrittlement. The failure modes and metallurgicalappearances are identical in stress corrosion failuresand hydrogen embrittlement failures. The primary indi-cation of stress corrosion is that the delay is longerthan 24 to 48 hours after installation. More details onstress corrosion will be dealt with in a later article.

2. The fasteners must be hardened to at leastRockwell C37. Unhardened fasteners never suffer fromhydrogen embrittlement. Fasteners that have a hard-ness of Rockwell C36 or less are extremely unlikely toever suffer from hydrogen embrittlement.

The greater the fasteners hardness above Rockwell

JoeCross-Out

JoeInserted Textgrade 9 bolts

JoeNoteThe use of "L-9" (Trademark of Brighton-Best) was used in error. The term "grade 9" bolts is what was intended by the author.

If the hydrogen location is somewhere other thanone of these two locations, then hydrogen embrittle-ment is probably not the culprit. Hydrogen embrittlement

failure always occurs where the stress concentrationsare the greatest. If the bolt is absolutely straight whentightened, the failures will generally be where the fas-tener head joins the body. If there is any lack ofstraightness or bending action in the application, thefailure will most likely fail in the thread just above thethread engagement.

When investigating reported fastener failures, keepin mind that for the failure to be attributable to hydro-gen embrittlement, ALL FIVE of the characteristicslisted above must be present. If any one of the aboveare not associated with the failure, it is extremely un-likely that hydrogen embrittlement is the culprit.

The best insurance against ever experiencing a hy-

drogen embrittlement failure is to avoid the factors thatmake hydrogen embrittlement failures possible by do-ing the following:

1. Do not electro-plate inch socket head cap screws,L-9 bolts, or metric property class 12.9 bolts or screws.

2. If customers insist on using electro-plated boltsand screws, suggest they consider using a Grade 8 orproperty class 10.9 part of a slightly larger diameterinstead of using the socket head cap screw or prop-erty class 12.9 part.

3. If the customer insists on using an electro-platedsocket head cap screw or property class 12.9 part,specify that the parts must be baked at 375-400 de-grees F within one hour after plating for at least fourhours at temperature.

Also, conduct one of the recognized hydrogenembrittlement tests on every lot of parts to provideyou the opportunity to catch hydrogen embrittlementbefore it is exhibited in the users application.

4. If you must electro-plate any type of tapping screw,specify to the heat treater that the core hardness ofthe screws must not exceed Rockwell C36.

If you will adhere to the preceding suggestions, youwill greatly improve your odds of never having to referto this article to see if your customers failure lookslike the picture in this article.

For more information on the subject or any otherfastener quality assurance, gaging, or calibration mat-ter, contact the author at 800-435-2647 by phone or byemail at greensladeandcompany@sbcglobal.net.

First threadabove

engagement Headto

Shank

Reprinted from Distributors Link Magazine, Spring 2005

/ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 1200 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile () /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org) /PDFXTrapped /Unknown

/Description >>> setdistillerparams> setpagedevice