I sent off another back of junkyard steel to my buddy with access to the mass spectrometer. Here are the results, some of which are very interesting:

• Torsion bar from jimbob, [added: off a Ford pickup]:
  

C:0.60 | Mn:0.85 | P:0.030 | S:0.025 | Si:0.28 | Cu:0.01 | Cr:0.80 | Mo:0.02 | Ni:0.01 | Sn:0.01 | V:0.008 | Nb: -
That's 5160.

• Cheapo, made-in-India, Harbor Freight bastard file:
  

C:1.3 | Mn: 0.34 | P:0.015 | S:0.009 | Si:0.24 | Cu:0.01 | Cr:0.62 | Mo:0.005 | Ni:0.02 | Sn: 0.003 | V: - | Nb:0.010

• Huge old, American-made mill bastard file [added: Nicholson]:
  

C:1.28 | Mn:0.34 | P: 0.016 | S:0.015 | Si:0.15 | Cu: 0.02 | Cr:0.14 | Mo:0.005 | Ni: 0.02 | Sn: 0.002 | V: - | Nb: -

• Another huge old, American-made mill bastard file [added: Heller -- this may be a pretty old file]:
  

C:1.20 | Mn:0.25 | P:0.010 | S:0.020 | Si:0.12 | Cu:0.04 | Cr:0.03 | Mo: 0.004 | Ni:0.03 | Sn:0.018 | V: 0.005 | Nb: -

Note: when I get home tonight I'll post an update with the brands of the two American files; I can't recall now which was which, but one was a Nicholson. Did you notice that the steel in the cheapo HF file is extremely comparable to the old American files -- arguably even a little superior? Steel snobs, take note: just 'cuz it's cheap and made overseas doesn't necessarily mean it's crap.

• Railroad tie plate (surprise here for me; I was figuring on something like 1050):
  

C:0.19 | Mn:0.42 | P:0.005 | S:0.030 | Si::0.04 | Cu:0.24 | Cr:0.05 | Mo:0.009 | Ni:0.07 | Sn:0.011 | V: - | Nb: -

• Used leaf spring of somewhat indeterminate provenance, from the dumpster behind my local truck spring shop (taken with the manager's permission!):
  

C:0.57 | Mn:0.74 | P:0.010 | S:0.015 | Si:0.23 | Cu:0.25 | Cr:0.70 | Mo:0.02 | Ni:0.09 | Sn:0.008 | V:0.004 | Nb:0.070

Per the conventional wisdom about leaf springs, that's 5160. I was actually hoping for something else, just to help me once again make the point not to take the junkyard steel charts too seriously. But in this case the charts were right. The next one helps make my point, though.

• Big ol' truck coil spring taken from the same dumpster as the leaf spring, above:
  

C:0.58 | Mn:0.82 | P:0.018 | S:0.016 | Si:0.90 | Cu:0.01 | Cr:0.46 | Mo:0.01 | Ni:0.005 | Sn:0.002 | V:0.094 | Nb:0.007

What's that alloy? A common junkyard steel chart says truck coil springs should be 5160. But this isn't quite 5160 (too little chromium), it's not quite 6150 (too much carbon and not enough silicon), and it's not quite 9260 (too much chromium, not enough silicon) -- and it has almost 0.10% vanadium, which is enough to make me think it might not be an accident. I'm really not sure what it is, to be honest. Should make a good, tough blade that hardens well even in a pretty slow oil, though.

• Here are the results for three pieces of mystery steel sent to me by Brian Brazeal. He can tell you what they came from, if he likes; I have no idea:
  

Eye l: C:0.18 | Mn:0.60 | P:0.012 | S:0.029 | Si:0.015 | Cu: 0.53 | Cr:0.13 : Mo:0.014 | Ni:0.12 | Sn: 0.017 | V: - | Nb: -

Eye ll: C: 0.58 | Mn:0.85 | P:0.010 | S:0.023 | Si:0.77 | Cu: 0.01 | Cr: 0.50 | Mo: 0.005 Ni:0.01 | Sn: - | V:0.004 | Nb:0.004

That one looks very similar to my mystery coil spring - not quite 6150, not quite 5160, etc.

Eye lll: C:0.40 | Mn:0.86 | P:0.010 | S:0.034 | Si:0.30 | Cu:0.23 | Cr:0.88 | Mo:0.16 | Ni:0.08 | Sn:0.010 | V:0.003 | Nb:0.023

That's 4140.

Edited September 25, 200916 yr by MattBower
correcting typos; adding additional info

Thanks matt. The Torsion bar was from a Ford pickup

Thanks, Jim.

very good info..Thanks for posting..Im also supprised about the cheap file :)

Thanks! I wonder about the Harbour Freight ball pein's? Great stuff!

Me too! Maybe next time.

Just my 2 cts. The info is great however we must remember all coil springs are not from the same place. Nor any other parts. Have had to deal with differant specs from differant suppliers. One time was pattern burning 3/8 steel plate(japan) Turned out to be a 3 piece plate 3 1/8ths rerolled to 3/8. Torch wouldn't cut thru it. I say use the #as a baseline. May be on the money Might only be close.
Ken

Absolutely, Ken. That's one of the explicit reasons I do this. In having roughly 15 pieces of scrap analyzed so far, probably half have turned out not to be what the "conventional wisdom" says they "should be." Everything depends on the particular manufacturer's specs, what steel was cheap at the time, and probably a dozen other variables. Maybe some demolition bits are S-series steel, but the ones I have are 1040/45 -- and others are undoubtedly other sorts of steel. Not all torsion bars are 5160, not all files are 1.2%-1.3% carbon (or W2, or 1095, as the conventional wisdom often claims), and so on.

Thanks Matt, I hoped i wasen't going to start a riot with that one. I don't worry all that much about exact specs. Don't make knives or any thing that needs exact specs. Lots of tractor repairs (old) and art pieces. Gun stuff I buy to spec.
Ken

"Did you notice that the steel in the cheapo HF file is extremely comparable to the old American files -- arguably even a little superior? Steel snobs, take note: just 'cuz it's cheap and made overseas doesn't necessarily mean it's xxxx."

have you actually tried forging one? there may be a layer of nice high carbon steel on the outside but the inside is some kind of xxxx pot metal...at least mine were...they crumbled with the first strike of a hammer.

Thanks for the info.

Snake: It's a whole lot more expensive to make some kind of "layered" material or putting xxxx inside of good steel. If you can make good high carbon steel, it's a whole lot cheaper to just make the whole thing from it. Having worked with hundreds of tons of 1095 I know that if it is over-heated it can look like two materials. What appears to happen is this: The steel is slightly over-heated in the forge. When it's taken from the forge, the outside gets a chance to cool a little. The first blow splits the material and the over-heated interior just breaks into crystals looking almost like a "geode". When it cools down it does indeed look like there was something different in the core. Seems to be easier to happen with high carbon steel than with anything else I've experienced.

Dragon lair: Same for your "layered" material. I think what happens there is that in rolling the plate cools a little too much on the outside and separates from the hotter interior into what appears to be three layers.

Edited September 27, 200916 yr by nakedanvil

Grant, Wasn't an appearance thing. After burning a few parts out(burn thru one layer then the flame went sideways) Cleaned the part noticed two openings in the edges. Sharp blade split it wide open. Only happened that one time.
Ken

"Did you notice that the steel in the cheapo HF file is extremely comparable to the old American files -- arguably even a little superior? Steel snobs, take note: just 'cuz it's cheap and made overseas doesn't necessarily mean it's xxxx."

have you actually tried forging one? there may be a layer of nice high carbon steel on the outside but the inside is some kind of xxxx pot metal...at least mine were...they crumbled with the first strike of a hammer.

As a matter of fact, I have. Didn't have any problems with it. Got screaming hard in the quench (though knowing what I now know, I'd HT it differently). Sounds like you overheated yours. You can't treat high carbon like 1018.