Anybody experience bad quality fasteners from China? 2

[plasgears]

I heard over the years that there was bootleg stuff coming out of China. Sourcing acetal from China ruined my company, our supplier of molded gears, and our customers. Remember, quality companies qualify suppliers, particularly if they are in the orient.

 

[WKTaylor]

Using unverified/verifiable [bogus] materials/parts from a third-world country or unscrupulous distributor, is an obvious invitation to financial and physical disaster. Hopefully no-one was killed or injured due to Your companies miss-steps.

Bogus aerospace parts and materials were becoming a major issue in the late 70s and early 80s in the USA.

The USA Fastener Quality Act (~1988) mandates tracability for all materials and parts used on aircraft back to the raw material producer. At each stage of the process the civil penalties, not-to-mention criminal culpability, for knowingly selling or passing-on uncertifieable materials or parts are severe. The FAA can/will arrest everyone involved with bogus parts/materials in a supply-chain. In fact, any mechanics tasked with using or installing the material or parts in a certificated aerovehicle, can/will be arrested. This is why aircraft hardware is no-longer "sold by the pound" in scrap disposal, like it used to be. I remember my dad buying bags of unsorted fasteners and other parts (floor-droppings) from Lockheed and Douglass plants... alongside mechanics from aircraft shops all over So California, in the late 60's. Bargains were incredible... but there was no tracability what so ever.

The DoD has a mandatory "buy-USA" policy that stipulates aerosace materials and parts be traced back to US raw material manufacturers... with certain notible exceptions for European countries that have equivalent standards as the USA. Leads to same effective tracability as the FAA system.




Regards, Wil Taylor

 

[plasgears]

Thanks WK,
The disaster I described was automotive related.

We were under pressure by Ford to reduce annual prices by 10%, and we passed this on to our plastic gear supplier. He responded by going to China as a source of material/molded parts. This was an acetal application. Obviously, he did not follow QS9000 rules that demand qualification of a new supplier, so the seeds of disaster were sown.

Our first indication of a problem was when the Jaguar sales mgr frantically called daily to complain of infant mortality in our line of seat actuators. I knew where the problem most likely originated, and I reported my views to the boss. He ignored me.

The problem spread throughout all lines, domestic and foreign. Ford suffered, we suffered, we discharged a long time supplier, heads rolled, and that was the end of my career. The company was sold twice in four years. I think China was using 100% regrind or off spec matl.

Main message: DON'T SOURCE CHINA! - for anything.

 

[CoryPad]

Yes, I worked for a USA automobile manufacturer that obtained nuts (steel, heat treated, electroplated with zinc) that had geometry errors (threads not tapped perpendicular to the bearing surface) from a Chinese source.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[drawoh]

plasgears,

Y'know, this problem is not really the Chinese. At worst, they followed orders. If you order a supplier to drop prices 10%, you are going to get a lower quality product.

JHG

 

[Compositepro]

The Chinese will get much better in quality, just as the Japanese did. But You do need to be careful. The mindless policy of mandating 10% price reductions every year from suppliers is simply evil, and is guaranteed to eventually destroy the supplier and the purchaser.

 

[WKTaylor]

Folks...

A senior-grade fastener engineer at my company educated me about the ways of the fastener world with a simple case study.

A 1-off fastener was designed by a sub-unit of the company, based on a existing Inconel 718 bolt design... without consulting him first. A very reputable fastener company made a prototype run (~100) per the drawing. The production run was very economical and timely... however, ALL of the bolts failed to inspect/perform as anticipated [stabilizer attachment bolt with altered head design]. Tensile/shear strength, Fatigue, etc... where way-low based on the altered existing design. Everyone was upset and livid that a reputable company would make such trash. WHY... There were even "odd dimensional and NDI defects" in a couple of the parts. OUTRAGEOUS!!!!

This engineer (who designed the existing standard part) was finally called-in for a consult (so they could ultimately hammer the vendor into humility).

He determined that the drawing failed to callout any production, inspection, strength or durability requirements: It was a stand-alone drawing". Essentially the manufacturer had produced the parts EXACTLY as the engineers had described... on the drawing. Unfortunately the engineers failed to understand that the original fastener design (they had based theirs on) had an entire specification standard dedicated to it that explicitly stated all the critical requirements... and more... that would have made the 1-off bolt a very good part.

The Senior engineer summarized this lesson learned in what has become one of my principle "golden rules of engineering"...

"You should expect to get what You ask for. However, if you don't ask for very much, DON'T (ever) expect to get very much."

Another correlary "golden rule" is as follows.

"It is unwise to pay too much for something. However, it is MORE unwise to pay too little and expect the same; for you risk not getting what you truly need."

or as more crude aerospace parrallel saying...

"It is unwise to pay too much for something. However, it is MORE unwise to pay too little and expect the same; for you risk not getting what you need. However IF you persist on this approach, all you risk is the F*ing wings breaking off."


Regards, Wil Taylor

 

[KENAT]

wktaylor, a classic example of why my job exists. People make all kinds of assumptions but if it isn't in black and white in the drawing pack (including MBD & specs) then you haven't fully defined it. Then even if you define what you want you need adequate QA/QC/Inspection to make sure you're getting it.

Make sure you define what you need, and check you're getting it.

We're looking at outsourcing stuff to China & the like and manufacturing have found that very few older drawing packs meet the first requirement. We're going crazy trying to get extra 'check resource' to help correct this.

As to the second part, I'm not over optimistic.

KENAT, probably the least qualified checker you'll ever meet...

 

[drawoh]

wktaylor,

I think you are moving from cost engineering to reverse engineering. This is a separate way to screw up.

I doubt that manufacturers in the car industry have margins of 10%. A 10% price reduction means cost reductions somewhere. If the company is already lean, efficient, and marginally within specifications, then they will have to take risks, or give up the work.

JHG

 

[dicer]

Out sourcing to China.
Is it really worth the hassle? Just like most of the stuff we get from there now, the China companys don't have to care about quality. They know once the product is out the door they will never see or hear about it again. Who ever sells the thing will eat it if there is a problem with it.
Why? The shipping hassles, the time trying to communicating it etc. Isn't this part of the reason the 787 has not made a flight yet? All outsourced parts had to be remade?

 

[carnage1]

My favorite personal example is socket head cap screws. we were using the heads inside a countersink to hold alignment in a machine we had just built. the heads were within spec (barely) but the heads were not on the same centerline as the bolts. So needless to say the bolts would not fit. I spent the day chucking bolts into the lathe and shaving the heads till they were concentric because we couldn't wait for new ones.
The other notable one was a 48x48x1 piece of fancy (expensive) plastic. It was 48x48 but was a parallelogram and we could barely get the rectangle out of the middle to build the prototype we needed.

Luck is a difficult thing to verify and therefore should be tested often. - Me

 

[KENAT]

Socket head capscrews in a countersink?

Are you sure the problem was the screws and not the countersink, or counterbore if you meant that?

I often see counterbores too small to accomodate tolerance variation at worst case.

I'm all for blaming foreign fasteners if they are the problem but sometimes the mating part is the issue.

KENAT, probably the least qualified checker you'll ever meet...

 

[carnage1]

counterbores, and yes they were speced nonstandard, however any bolt that had been in spec even at the top edge of spec would have fit with .001 to .003 inch clearance. The heads were off center by 10 to 15 thousandths.
We first thought it was that the counterbores were not big enough as that engineer had a history of that. (he didn't like how oversized some of the stock counterbore tools looked and would often spec them tighter, so we ended up buying a lot of 32nd and 64th odd sizes for his projects.)

Luck is a difficult thing to verify and therefore should be tested often. - Me

 

[drawoh]

carnage1,

I do not recall seeing a specification guaranteeing the concentricity of bolt or screw heads. Your problem is that you are using the wrong tool for the job. If you want to locate something, use dowel pins or shoulder screws.

Did you make both holes do accurate location in two axes, or did you slot one of them?

Raw material, such as 48"x48"x1" plastic sheet, should not be regarded as precisely dimensioned unless the vendor says it is.

What has all this to do with China?

JHG

 

[carnage1]

these items all came from china, and yes there is a minimum concentrically for cap head screws that the engineer looked up ahead of time. As for the sheet it was parallelogramed by 4-5 inches as I recall and was cut from a larger sheet not cast to that size.
Lastly the 8 countersunk holes were all round, and all fit fine with better quality bolts, we just did not have enough of them on hand and expected when we ordered more that we would get a similar product.

Luck is a difficult thing to verify and therefore should be tested often. - Me

 

[evelrod]

Thoughts along this same line...
Back in the late 60's I needed a bunch of high quality fasteners...Very expensive. I decided instead to buy surplus aircraft fasteners from Douglas in Long Beach...They were having a sale on such. I assumed (I know, I know) I was getting quality NAS fasteners, I bought a thousand or so, but instead I ended up with a bunch of counterfeit hardware. Obviously this problem predates China....Fool me once, etc. I still use surplus stuff from Deering or other reputable suppliers...sill cheaper...but I inspect each bolt, nut, screw individually and on critical parts I take at least one to destruction. Still cheaper (by far!) than new fasteners. Even commercially available Grade 8 hardware is suspect and, I check them the same way.
Not too big a problem these days as I don't build for customers any longer...just my own race cars.

Rod

 

[drawoh]

evelrod,

How do you test an alleged grade[ ]8 bolt?

I was under the impression that most of the counterfeit grade[ ]8 bolts were grade[ ]8.2, which have the same tensile strength at room temperature.

JHG

 

[carnage1]

given the factor of safety in most designs i would think that an 8.2 would suffice in all but the most critical applications. It is still a lot better than a grade 3 with extra markings.

Luck is a difficult thing to verify and therefore should be tested often. - Me

 

[drawoh]

carnage1,

An 8.2 ought to be just as good as a grade[ ]8 at holding on a door bracket. For holding on a cylinder head or an exhaust manifold, I would be a lot less sure.

JHG

 

[evelrod]

To start, I don't bolt cylinder heads or exhaust manifolds, indeed, ANY engine part, with surplus or G8. Critical engine hardware is reserved for ARP, SPS... custom fasteners. I do use G8 and in some cases, NAS surplus in other mounting areas...safety straps for drive shafts, camber plates, roll bar trunnions, mounting plates for accessories, accumulators, fuel pumps and filters...that sort of thing. Nothing critically loaded.

How do you test a G8 fastener? That should be pretty intuitive. Torque it up till it breaks or strips out. Trust me, the ungraded crap with the extra head marks will die well before a known G8!!! Example...In 2004 I picked up batch of (presumably) G8 5/16-28 nuts properly marked and all that. Nicely anodized. Only problem is they would strip at anywhere from 144 in/lbs to 240 in/lbs. when the known fasteners would tq. to 300 in/lbs with no problems.

Rod