'setting' of grey vs. ductile vast iron (bolt loosening) 2

[321GO]

Hello Guys,

what's happening:

** Clamping bolts remain tight clamping ductile iron, but come loose clamping grey cast iron **

How does this work?

The HB hardness value's for both materials are fairly identical(it seems).
Furthermore the E modulus is smaller for the grey cast iron(I know that E modulus on grey cast iron is somewhat BS).
Wouldn't the smaller E modulus even be favorable for the grey iron - when bolts set a certain amound, there is less clamping force reduction?

I have read that the flake form graphite has a more negative influence on the matrix in general, then spherical graphite. Is the difference in graphite form of importance here regarding this more pronounced 'setting effect' in grey cast iron?

Help greatly appreciated!

GCI
Hardness(HB)
180-250
E modulus(tensile)
103-118 GPa

DCI
Hardness(HB)
170-230
E modulus(tensile)
169GPa

 

[CoryPad]

Settling refers to small-scale force loss (relaxation) due to plastic deformation, creep, surface roughness, etc. This usually is < 10 % of initial preload. If this is enough for your joint to loosen, likely you need higher preload. I am not aware of a significant difference in settling between gray and ductile irons. Cast irons usually have high allowable surface pressure from fasteners compared with their regular tensile properties, so extra tightening should not be a problem.

 

[btrueblood]

The problem may be more with the brittle nature of the gray iron, specifically in how the ductile fastener threads interact with the body threads. If the first few threads of a fastener carry most of the shear load, then you must count on the material of the female threads approaching or even slightly exceeding yield. In some of our old grey iron castings, we would see the threads strip in a zipper fashion, each time they were torqued, i.e. the lead thread would shear off on each tightening sequence.

 

[EdStainless]

Check the compressive strength of the two materials.
Sounds like you need higher preload.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[EdStainless]

What I meant to add was that there may be a time factor due to the fracturing of the flakes. So either load slower or go back and re-load.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[racookpe1978]

Do you have clamping bolts (an independent bolt clamping two pieces of each metal), or clamping studs (studs that are screwed into threads cut into the two different kinds of metal)?

 

[321GO]

Guys,

let me explain somewhat better:

The part is a disc which is clamped with 10 bolts on it's inner circumference. The disc is subjected to a torque load. In use the bolts of the cast iron part would come loose were on the ductile iron part they would not.

But i must also note that the cast iron part had cracks originating from those bolt holes! Could those cracks cause the bolts to come loose? Could such cracks cause a reduction in the compression of the material being clamped? The cracks were somewhat 'open' though.

I'm kind of puzzled by this at the moment.

Thank you all again!

 

[321GO]

illustrating my problem..

 

[TVP]

321GO,

You have identified the problem-- the cracking of the grey iron bolt holes/bosses means that there is less clamp load than in the ductile iron parts. You need to determine why the grey iron part is cracking.

 

[321GO]

TVP,

probably due to the torque load and resulting tensile stress on the holes and the fact that tensile strength is only half that of the ductile.

Thanks to all who participated!

 

[racookpe1978]

OK - You have several things going on here.

The bosses are overstressed and cracking, and from that jpeg file, it appears that several bosses have cracked, not just one. Is that right? (Having several crack in the same place on each boss indicates that actual stresses are much greater than what can be resisted by the actual metal present. That is, if only one boss cracks oone only one part, that one bolt or that one boss or that one part is overloaded or had a flaw. Having several bosses crack on one part proves the basic design is overloading all of the bosses, it's just that not all of the bosses have failed yet on all of the parts.)

It appears the boss wall is not thick enough to withstand the compression force of the bolt pre-stressing plus the actual load of the part being carried.

The jpeg is not clear - what is the cross section of the boss? The chamfer (counterbore?) around the bolt hole might be a contributing stress riser that is "wedging" the boss apart as the bolt is tightened.

A long axial crack in the boss wall drawn as it is in the jpeg should not allow the boss to "shorten" very much. (A twisted or transverse crack would allow the boss to shorten under bolt pre-stress and would loosen the bolt enough to remove all stress. But I don't think an axial crack would do that. Am I right? Don't know. Haven't had to measure a boss after such a crack.)

So - get a cracked part and a "new" part. Measure - to the nearest thousandth - each boss on the new part and on the cracked ones. If the cracked bosses are 1/4 bolt thread pitch length shorter, then you know your problem.

 

[321GO]

"A long axial crack in the boss wall drawn as it is in the jpeg should not allow the boss to "shorten" very much. (A twisted or transverse crack would allow the boss to shorten under bolt pre-stress and would loosen the bolt enough to remove all stress. But I don't think an axial crack would do that."

That's exactly what confused me also, I mean how could such axial cracks cause relaxation??

But maybe because of the increased movement (due to the crack's) during loading and subsequent abrasion of the mating surfaces??


p.s. image is just indicative

 

[cloa]

Is there some allowance for the fact that cracks alter the elastic properties so that the effect elastic modulus is less?