Dealing with Decarb in multiple normalaztion cycle's, coarse spheroid annealed steels

According to verhoeven, a carbon atom can move 1mm in 18 min at 2100f. At 1700-1800f, it's about an hour, plus or minus a few minutes.
 
Willie71 said:
According to verhoeven, a carbon atom can move 1mm in 18 min at 2100f. At 1700-1800f, it's about an hour, plus or minus a few minutes.
Click to expand...

Thank you for pointing out those numbers. Let's be certain to mention that for carbon to move and migrate at those rates it must first have a place TO GO.
That is why I often use the de-carb line in stainless/carbon san-mai as a carbon migration example. The carbon is moving from a carbon Rich environment to a carbon deficient environment. When you have a homogenous piece of Steel the carbon has no Need to go anywhere.
If our carbon was leaving our steel at the rate mentioned then in just any normal forging situation we would not have any carbon left in our Steel.
 
2-3 passes on the SG with a 120X belt & ZAP, I'm right back to a nice, clean, flat, parallel blade. From there its just a matter of belt progression until I get the finish I'm looking for.
 
kuraki said:
That's a good point. When I say decarb, I'm referring to the soft layer of material that always exists after heat treat, usually .005" to .020" thick depending on material, temp, time, etc. I'm using satanite wash to reduce scaling, not necessarily decarb, although I imagine it does help with that.

If carbon moved out of the blade as fast as some think, forging wouldn't be possible.
Click to expand...

While in an extreme situation, there might be .005" of secarb, .020" would be nearly impossible unless the steel was placed in a high temperature oven for days. 020" is 500 microns.
 
Willie71 said:
According to verhoeven, a carbon atom can move 1mm in 18 min at 2100f. At 1700-1800f, it's about an hour, plus or minus a few minutes.
Click to expand...

Not meaning to dispute Verhoeven, but I think the argument you give is apples to oranges.

I don't have those charts available, but I would point out that moving a carbon atom through steel in carbon migration isn't what is going on with decarb. I was under the impression that carbon migration was slower than the specs you list. I am pretty sure decarb formation is exponentially slower.


At 1mm in 18 minutes, a 1/8" thick (3mm) 1095 blade would have little or no carbon left by the time it was done being forged (between 1800 and 2100F). HT would strip out the remainder. I believe most of us get fully hardened 1095 blades with almost no decarb layer thicker than a few thousandths.
 
Well truth be told Im an old Blacksmith so I don't think of Decarb in the same way as scale..I know Karl is right but im use to calling it decarb because it looks so much different than the scale Im use to dealing with..But regardless the lizard skin I got after three normalizations and a heat treat is wayyy thicker than I like..Most steels we deal with don't require so many thermal cycles but the stuff that comes spheroidal annealed like that 80crv2 (in stock removal especially) does..I learned that the hard way.:grumpy:
 
Different steels decarb at different rates. I've never seen it go more than a few thousandths of an inch. Might be more in an oxigen rich atmosphere.

Hoss
 
DevinT said:
Different steels decarb at different rates. I've never seen it go more than a few thousandths of an inch. Might be more in an oxigen rich atmosphere.

Hoss
Click to expand...

I get almost zero scale unless I am forging at high heat or welding Damascus.
I get virtually no scale during normalization Cycles. If you're getting lots of scale things need to be adjusted.
 
Karl B. Andersen said:
I get almost zero scale unless I am forging at high heat or welding Damascus.
I get virtually no scale during normalization Cycles. If you're getting lots of scale things need to be adjusted.
Click to expand...

I was about to say the same thing, Karl.

Mr. Ken Tucky... how long are you soaking at your normalizing temps?
 
Carbon does not travel through the decarb layer. Wayne Goddard had some of his early cable damascus tested, the pattern came from a layer of decarb around the wire. It was determined to be pure ferrite, the carbon never did equalize with forge welding and forging.

Hoss
 
Stacy E. Apelt - Bladesmith said:
While in an extreme situation, there might be .005" of secarb, .020" would be nearly impossible unless the steel was placed in a high temperature oven for days. 020" is 500 microns.
Click to expand...

The 1095 I have is hot rolled plate and it comes in at 45 RC. I anneal it before working with it by soaking at an hour (stacked blade blanks) at 1600F and then shutting the furnace down and letting it cool overnight. When I pull them out the next day they're usually still 250F or so, so they cooled at roughly 100 degrees an hour. In that state they're right around 20 RC. After rough machining or grinding, I normalize 3x, 1600, 1500, and 1400. I austentize at 1475 for 10 minutes, quench in Parks 50 and at that point expected to see 66 RC. I had to grind .020 to see 66 RC consistently throughout the blade. Surface ground, I didn't mic the thickness, so while I dialed .020 wheel wear etc maybe I took .018".

That's where I get the .020". At .010" the RC would swing from high 40s to low 60s. I didn't test again between .010" and .020" I just went another .010" to make sure since the hardness tester is on the other side of the plant from the surface grinder.

ETA: no wrap, no clay, no atmospheric control
 
The presence of a decarburation layer just says that in oxygen rich athmosphere the carbon moves more promptly out of the steel than in between the matrix...it is not surprising that moving by simple diffusion gradient in between iron atoms is less easy than jumping into CO or CO2.
But the carbon present in the bulk of the steel most surely prefers to diffuse toward the decarburated layer, and travelling through it toward the athmosphere, increasing the decarb layer thickness depending on the time/temp applied.
A well balanced (rich/neutral) forge athmosphere will exert little driving force for the carbon to migrate toward. Scaling (not decarb), if occours also act both as athmosphere insulating layer plus you have the carbon depleted layer flaking off more or less as quickly as the decarburation proceedes.
 
Thanks. Is that using electric or gas? Since you gave specific heats and soak times, I am ruling out coal. I'm just trying to figure out how you are getting that much decarb/scale. This is my regime for 80CRV2 in an electric kiln with (3) T/Cs.

After forging or before stock removal
1650F, 2 min soak, Pull out for as long as it takes to ramp down.
1500F, 2 min soak, Pull out for as long as it takes to ramp down.
1475F, 5 min soak, oil quench
Heat 2-3 times to dull red(still magnetic) in my LP forge

I get very little scale. Most of the thin shiny skin pops off during those last few sub-critical heats. And that might just be polymerization from the quench oil.

After grinding...
1250F for 5 mins of stress relieving (check for straightness)
1500F, 2 min soak, oil quench.

I believe that quenching on that last normalization cycle "locks it all in" so-to-speak and allows me to get away without the extended soak folks seem to think 80CRV2 rerquires.

Truth be told that for 1084, 1075, and 5160, I run the whole normalization cycles through my forge "by eye:eek::cool::thumbup:" and use the kiln for the final quench and tempering. 1095, W2, 80CRV2, 52100, L6 and O1 are done with tighter tolerances.
 
Its electric, a evenheat 18"...Ill take a pic of the next one..I really don't want to say "scale" I still want to say decard because what I see as scale is black oxide that flakes off and I get very little of that...What Im talking about is that very grainy, softer than hard layer that looks like lizard skin that you have to grind through to get to good steel..
God Im sounding like a greenhorn here :D
 
I get it... The orange peel stuff... lol. Can't say I have never took note of the difference between coarse spher. and simple HR. Perhaps 1095 has given me stranger result than others. Have you ever had the scale come of like a sock after the quench?
 
Karl B. Andersen said:
Thank you for pointing out those numbers. Let's be certain to mention that for carbon to move and migrate at those rates it must first have a place TO GO.
That is why I often use the de-carb line in stainless/carbon san-mai as a carbon migration example. The carbon is moving from a carbon Rich environment to a carbon deficient environment. When you have a homogenous piece of Steel the carbon has no Need to go anywhere.
If our carbon was leaving our steel at the rate mentioned then in just any normal forging situation we would not have any carbon left in our Steel.
Click to expand...


This is why the atmosphere in the forge is important. As you said, the carbon has to go somewhere, and having no where to go mitigates the problem.
 
kuraki said:
The 1095 I have is hot rolled plate and it comes in at 45 RC. I anneal it before working with it by soaking at an hour (stacked blade blanks) at 1600F and then shutting the furnace down and letting it cool overnight. When I pull them out the next day they're usually still 250F or so, so they cooled at roughly 100 degrees an hour. In that state they're right around 20 RC. After rough machining or grinding, I normalize 3x, 1600, 1500, and 1400. I austentize at 1475 for 10 minutes, quench in Parks 50 and at that point expected to see 66 RC. I had to grind .020 to see 66 RC consistently throughout the blade. Surface ground, I didn't mic the thickness, so while I dialed .020 wheel wear etc maybe I took .018".

That's where I get the .020". At .010" the RC would swing from high 40s to low 60s. I didn't test again between .010" and .020" I just went another .010" to make sure since the hardness tester is on the other side of the plant from the surface grinder.

ETA: no wrap, no clay, no atmospheric control
Click to expand...

I'm getting about 0.010" of decarb with normalizing and heat treat.
 
Willie71 said:
I'm getting about 0.010" of decarb with normalizing and heat treat.
Click to expand...

Sorry I just realized there could be some misinterpretation here. I was talking .020 total, which would be .010 per side.
 
You must log in or register to reply here.
Back
Top