First time normalizing 1095 was a catastrophe

OccamsBlade

Jim Dobbler
Joined
Jan 23, 2014
Messages
227
Like the title says, my first time normalizing didn't go so well. I've been making knives for a few years now on and off. I've mostly been using 01 and 1084. The first several knives I made with 1095 recently were problem free. However I started making bigger kitchen knives and wanted to try a few 7" kitchen knives (ground down to .02 - .03 on the edge). I figured I'd try to get the best out of my 1095 and did some research. What I found is that a lot of knifemakers suggest normalizing 1095 before hardening, even if you do stock removal. So I gave it a shot.

From the many hours of research I did on the subject I decided I'd do a normalization cycle in my Atlas forge at 1500, 1450, and 1400(x2). After putting all three of my blades through the first cycle at 1500 (trying to heat them evenly until there was no shadow) everything seemed fine. The only odd thing, that I've never had happen before, was when the knives cooled to black they would develop a bright red powdery substance on the surface. Like a fine rust but deeper red. So just wiped it off with my glove and went for the second cycle at 1450. After pulling those out I got the same red stuff, but also noticed some warping. Thinking about it now it could have been something I did. I maybe swung it around too much, or let it sit on a cold surface, or uneven heating, or all of the above.

I put the ones with bad warps in a vice between two steel plates while still warm with no success. Not simple warps but slight S shaped warps. I've never had warps nearly as bad as these. So I guess I've been lucky up until now, haha. What had me beside myself is that it happened during normalization. I thought it was supposed to fix issues with the steel, lol. Anyway, was wondering what the best procedure is at this stage for fixing warps. I was thinking of taking a torch to the spines where the bends are and try to straighten them with a gloved hand in a vise. I did try doing it cold in the vise but thought it might add even more unnecessary stress to the blades and stopped. It did correct it a bit, but with a lot of torquing involved. At this point the blade has hardened a bit, so fixing warps isn't as easy as before. I got slight warping during grinding, even though I quenched every pass or two depending on the grit.

Another odd thing I noticed with this particular 1095 stock was that when I ground the bevels a wood like grain pattern started emerging (image below). Probably not related to the warping or red stuff, but thought I'd point those out anyway. The 2 pieces of 1095 stock were purchased from Jantz. The first one, with no weird stuff going on, was a 3/32 piece of 1095. However, the 1/8 piece of 1095 that I made the current blades from did. Mostly that pattern.

Sorry about the lengthy post, but wanted to make sure I included all relevant info. On a side not I have an Evenheat on order. But wanted to get another batch or two done while waiting 5 to 6 weeks for it to arrive. I think I'll just wait before trying to do normalization cycles again:/

PKIs2Xe.jpg

NnGnp3s.jpg
 
I think you're doing grain refinement with your 1500/1450/1400 cycles. Normalization would be 1650 and then air cool (after which you need to grain refine), and is normally done for spheroidized steel with stock removal.

How are you cooling the steel between cycles?
 
As I understand it normalizing (or thermal cycling) is for grain refinement. And is required for spheroidized steels, but is common for knifemakers to normalize after grinding, pounding, drilling, etc. My blades were slightly warping as I ground them, so figured it would reliveve any stresses and set me up for an easier time during heat treat, and produce a better blade/grain structure at the very least. At least that's what I've read here on the forums and elsewhere. But of course it quickly gets muddled, contradictory, and all over the place as far as numbers and methods go.

Yes, I've seen numbers as high as 1650. I was going off of Nick Wheeler and Don Hanson's suggestions for normalization in other threads. Cycling anywhere from 3 to 7 times at around 1400 to 1500 seemd to be the most agreed upon temps. Wheeler does go as high 1600, and says it worked best for him, but says normalizing/thermal cycling at 1400 to 1500 works as well.

I just air cooled until black. On one or two I slowly swung them back and forth, slicing the air, making sure not to go side to side. Maybe that's what caused the warping.

What really has me baffled is that pattern in the steel. Since it seems like it's there to stay. And seems to get more pronounced the finer I go on grits.

Right now I'm just trying to get these warps out. On closer inspection it appears they're warped from spine to edge as well. Not sure even the most experienced knifemaker could get these dewonkified. lol.
 
What did you start with as a surface condition ? Differences in surface conditions can cause problems .A the surfaces should be the same .LL
 
Not sure about the surface texture, but the red "rust" would just be red oxides that can appear when cooling from these high temps.

Normalizing is done at the higher temps like 1600-1650 in order to get all the grain to a homogeneous size and equalize everything. Doing this will cause the grain to "blow up" in size so the subsequent reducing temps are to refine that grain after getting everything "normal" or equal.

Also, if you want to take any potential stresses out after grinding but the grain has already been set up beforehand, just take it up to about 1250 and let it cool to black as usual.

These heats can cause the blade to warp if it has any stresses built up, which is why you would do a 1250F stress relieving cycle post-grinding but pre-hardening, so you have a chance to correct the warp before hardening. Just take the blade back up to 1250 and bend it back to straight while it's at temp. Then do it again until the blade doesn't warp anymore as it's heated and cooled. To get the warps out you can set the blade on an anvil and use the face of a hammer or something to press down on any raised section while using your tongs to slightly lift the tang to help over-correct the warp if it requires it.

~Paul
My Youtube Channel
... (Just some older videos of some older knives I've made in the past)
 
cotdt cotdt Yeah, I'm going to try and contact Jantz and see if they have any idea what it might be. It almost looks like damascus.

mete mete The surface conditions of the knife are as close to even as I coould get them on either side. I fixed any small warps I found as I went along. And ground it to 220 grit before proceding.

C coldsteelburns I figured it was some sort of oxidation, just never seen such a bright red before. Hmm, so Nick Wheeler and Don Hanson are wrong about normalizing temps? Maybe I misunderstood what they were saying. This is all so confusing, haha. Here's the thread I was referencing https://www.bladeforums.com/threads/how-to-normalize-1095.346114/
That's just one source I looked at. My understanding is that going over hardening temp would cause grain growth? From what I understand it needs to be closer to the steels hardening temp for normalizing, so 1500 to 1400. Although Wheeler does say the ideal temp for his setup is 1600, then reduces the temps. But like I said, the more you look into it the more muddled it becomes.

And here's a quote from Stacy:

"Normalization:
Normalization is done by heating the steel to about 1500°F and holding long enough to become fully austenitic, approximately 5-10 minutes. It is then air cooled to ambient. This makes a moderately soft structure, with patches of uneven hardness. It works fairly well for tasks such as bending and simple working, but not always sufficient for drilling or grinding. The resultant structure is fine pearlite. This process is often repeated at descending temperatures as part of a HT regime. This is done to refine the grain and place all the alloying elements where they are best situated for the final hardening."

The times that I see temps at 1600 is usually for spheroidized steels.

Alright, so there's no more confusion on my part... I should do the stress relief of 1250 first from now on, before doing normalization cycles? Then after fixing any warps that may occur, start the normalization cycle at 1650 or 1600, and drop the temp incrementally after that? This is for 1095 steel.

I'm off to fix these warps and hopefully find out what this texture is.

Thanks for the help!
 
Last edited:
Nothing is wrong with the steel.

You are guessing at the temps, which is part of the problem. These thermal treatments are far better done in an oven. If using a forge, use a muffle and a thermocouple with a readout (PID or meter).

Normalization by heating 200-300F above the non-magnetic point, followed by slow cooling, will release internal stresses in the steel and return it to a more workable condition.

Grain refinement by dropping the temperature through the austenitic range and cooling after each step ( quenching is the norm) (eg. 1600,1500,1400) will reduce the grain size.

Sub-critical Annealing, by heating to 1250F and holding for a while ( one or two hours is best), then slow cooling to below 900F will reduce the hardness before sanding/filing/grinding. Obviously, you need a HT oven for this. You can do a so-so sub-critical anneal by heating to around 1250F ( stoll fully magnetic) holding for a few minutes if possible, then sow cooling to black.

The red color on your steel is just red iron oxide that formed on top of the scale from the procedure. It is of no concern.
The darker surface after the second lower heat cycle was aso scale. To remove scale, soak in vinegar or FC for a few hours to overnight.
The pattern is hard to tell in the photos, but you may have formed carbide separation in a wootz like pattern.
 
Unfortunately I wasn't able to fit any kind of muffle in my Atlas that would also fit my knives. I did use a PID though, and tried my best at adjusting the temps on the fly. I'm definitely going to hold off for the evenheat before trying anything like this again, haha.

Not to be a nuisance, but I'm just trying to get this straight in my head. So what you're saying is normalizing and grain refinement are the same, but completely different. Heating it to 1600-1650 and letting it cool slowly will normalize the blade/relieve any stresses. Then dropping the temp through lower cycles, and quenching in between, is where the grain refinement occurs. Okay, I think I got that part now, even though the numbers are still all over the place, HAha. But for obvious reasons, and my own sanity, I'll stick with your words.

Yeah, I read that about annealing, temps needs to be held for quite some time.

Yep, soaked them in vinegar over night.

My first thought was that the pattern was created when the steel was formed. And I may have brought it out while grinding. Although it started to appear at about the same depth on the other blades as well, even though I quenched after each pass or two to see if it would correct it. And the other 1095 I bought from Jantz didn't have the same issue. Seems like it was layered or laminated on this particular piece of stock. I looks kinda cool. If it is wootz, and doesn't effect performance, bonus for me!

At any rate, I contacted Jantz and sent them a pic of it. We'll see what they say.

Thanks!
 
I would suggest griding past the decarb before you jump to any conclusions about the steel. I bet it looks perfectly fine once you get down a bit. Also, grind to at least 220 grit before shooting photos of the surface. Those deep scratches don't allow much detail to show through. Grind it down and post new photos.
 
The soak overnight loosened up the decarb for the most part. I just finished up taking the rest off with a scotch brite belt. I'm at about .02 on the edge, and probably need to straighten them out a bit more before doing any more grinding. The edges are a bit wonky too.

That image I posted was at 220 and a scotch brite, just to peak at what it looked like.

I'll see if a video will work better, moving it around in the light, as it's hard to see at one angle.
 
l couldn't figure out how to embed the video so here's the link:

https://dai.ly/x70vzou

I don't think grinding any more will get past it. It seems to be in the middle of the bar. Looks fine on the surface, and started exposing that pattern the more material I take off on either side.
 
That is a very odd looking surface texture, and red dust? Never had that happen to me either.

About normalizing, there is often confusion about it. Some guys are using the wrong terms describing what they are doing. I have seen people use the term "normalizing" when they are just thermal cycling for grain refiment, and it is not correct. Look up normalizing in any steel book for 1095 and it will tell you that the normalizing heat for 1095 is 1600F. All of them say 1600F.

However, we often recommend 1650F for 1095 because certain vendors supply 1095 in a heavily annealed condition (very soft), and you need a good normalizing heat to break up the spheroidized carbides. Especially if the steel came from the New Jersey Steel Baron.

After normalizing is done, then "thermal cycle". And what you did with 1500, 1450, etc is thermal cycling. I would just recommend 1500 3x, for a few reasons I won't get into here.

Normalizing is mostly about the carbide, while thermal cycling is about grain refinement. 2 different processes for 2 different things.
 
I downlanded the pdf and going to give it a thorough read. Cuz now I'm obsessed and won't be able to sleep until I fully understand what's going on, haha.

On that link Willie provided Cashen includes grain refinement/cycling under the normalization procedure, even though it's for completely different purposes. Not trying to be a sperglord, just trying to get it straight in my head. And certainly making more sense now.

"Normalizing is used after forging the blade to even out all the chaos inflicted by the hammer. Industry specifies much higher heats for normalizing than many bladesmiths, 1600oF.-1700oF., and I always start out with a higher temperature to be certain that I put things into solution. At this first stage I am not so concerned about how fine the grain is but that they are all the same size, uneven grain size can be worse than larger grains, so using the high heat levels out the carbide/grain size and actually "normalizes" the inside of the steel. I then follow this heat with two or three more normalizings at subsequently lower heats to step down and refine the sizes of those constituents."

So when doing grain refinement I want to quench with Parks 50 between each heating? And just air cool for the normalizing part at 1600ish? Just to clarify.

I'm definitely learning a lot and really do appretiate the time you are all taking to help me out.

EDIT: After doing more research I plan on doing a normalization at 1650 and air cool to black. Then thermo cycling at 1550, 1500, 1450, and 1400. Air cooling to black in between. Looking at another thread about thermo cycling, and Cashen's site, seems air cooling is the way to go. Quenching in between seems like you'd be opening yourself up to unnecessary hardening and/or warping by cooling it too quickly. https://www.bladeforums.com/threads/thermal-cycling-with-a-forge.1395570/

I have them all pretty much straightened out now. Just need to wait for the evenheat I ordered a few days ago. So it's safe in assuming I can just go ahead with my new plan once I get the kiln? Or is there anything else I have to do to correct what I did before moving forward? My understanding is that by doing a normalization heat at 1650 will bring the grains back up. So it's just resetting everything I did essentially.
 
Last edited:
I downlanded the pdf and going to give it a thorough read. Cuz now I'm obsessed and won't be able to sleep until I fully understand what's going on, haha.

On that link Willie provided Cashen includes grain refinement/cycling under the normalization procedure, even though it's for completely different purposes. Not trying to be a sperglord, just trying to get it straight in my head. And certainly making more sense now.

"Normalizing is used after forging the blade to even out all the chaos inflicted by the hammer. Industry specifies much higher heats for normalizing than many bladesmiths, 1600oF.-1700oF., and I always start out with a higher temperature to be certain that I put things into solution. At this first stage I am not so concerned about how fine the grain is but that they are all the same size, uneven grain size can be worse than larger grains, so using the high heat levels out the carbide/grain size and actually "normalizes" the inside of the steel. I then follow this heat with two or three more normalizings at subsequently lower heats to step down and refine the sizes of those constituents."

So when doing grain refinement I want to quench with Parks 50 between each heating? And just air cool for the normalizing part at 1600ish? Just to clarify.

I'm definitely learning a lot and really do appretiate the time you are all taking to help me out.

EDIT: After doing more research I plan on doing a normalization at 1650 and air cool to black. Then thermo cycling at 1550, 1500, 1450, and 1400. Air cooling to black in between. Looking at another thread about thermo cycling, and Cashen's site, seems air cooling is the way to go. Quenching in between seems like you'd be opening yourself up to unnecessary hardening and/or warping by cooling it too quickly. https://www.bladeforums.com/threads/thermal-cycling-with-a-forge.1395570/

I have them all pretty much straightened out now. Just need to wait for the evenheat I ordered a few days ago. So it's safe in assuming I can just go ahead with my new plan once I get the kiln? Or is there anything else I have to do to correct what I did before moving forward? My understanding is that by doing a normalization heat at 1650 will bring the grains back up. So it's just resetting everything I did essentially.

Yes, the terms are used incorrectly quite often.

The reason we normalize is to correct the grain growth caused by forging at higher heats, or to break up large (course) spheroidized carbides caused from the annealing at the mill. Don’t confuse this with fine soheroidizing with a subcritical anneal (1200f for 2h.) Chuck at AKS only sells steel ready for heattreat, that don’t needthe normalizing and thermal cycling. Aldo’s steel depends on which batch or thickness it is. I just normalize all of his steel to be on the safe side.

Normalizing will cause grain growth, but they will even out in size. Then you refine the grain with lower heat, above a1, but below ac3. You now have very fine even grain with even alloy and carbide distribution.

Many of us used to use decreasing heats, such as 1500, 1450, 1400f, but Larrin (phd metallurgist) has informed us that three heats at 1450f will do the exact same thing. Two is probably enough too. That’s how I do it now, as I don’t have to wait for the oven to equalize at new temps. Air cooling is correct too. You have to read lengthy explanations as to why this is best, but it leaves you with more desirable structures for final heat treat to keep it simple.

The interchange of terms for similar processes is very confusing to say the least.
 
Haha, yeah. It's enough to want to bang your head against the wall. But thanks to all of the help I have a much better understanding now. I do plan on sitting down an reading that pdf. Just a quick glance at it, looks like it gets pretty deep.

Yeah, started looking at what AKS has for steel a couple weeks ago on someone's suggestion. Decided that's where I'm going to get my steel next. So for example if I get 1095 from AKS should I not concern myself with normalizing or grain refinement after grinding? My research has shown that everyone doesn't agree on that either, cuz reasons.
 
Haha, yeah. It's enough to want to bang your head against the wall. But thanks to all of the help I have a much better understanding now. I do plan on sitting down an reading that pdf. Just a quick glance at it, looks like it gets pretty deep.

Yeah, started looking at what AKS has for steel a couple weeks ago on someone's suggestion. Decided that's where I'm going to get my steel next. So for example if I get 1095 from AKS should I not concern myself with normalizing or grain refinement after grinding? My research has shown that everyone doesn't agree on that either, cuz reasons.

Aks steel can be heat treated as is. Unless you forge it, then you need to normalize and thermal cycle it,
 
Back
Top