Blotchy Etch in 1084

Jason Wilder

Well-Known Member
Blotchy Etch in 1084 (With Pics)

Since an etch will show what's happening inside the blade, I think this is more of a heat treat issue than poor etch.

My Observations: Very cloudy and splotchy etch on 1084. Possible bad heat treat?

My Process:
Last night I heat treated a fairly thin (too thin) 2" wide 1084 blade with the following process:
1. Blade was forged, then normalized three times at reduced temps (by eye)
2. Profiled and beveled
3. Heat home built kiln to 1200 and hold for 20 minutes, then ramp to 1600
a. Insert blade, wait for temp to come back to 1600 and hold for 10 minutes.​
b. Remove blade and cool in still air until black​
4. Ramp oven to 1499
a. Insert blade, wait for temp to come back to 1500 and hold for 10 minutes​
b. Remove blade and cool in still air until black​
5. Reinsert blade into 1499 oven, allow to come to temp, and hold for 5 minutes
a. Remove blade and quench in P50 for 3 seconds, remove to inspect, then back in oil until cool (Blade was moved up and down to agitate oil)​
b. I first noticed a very uneven color on the blade. Very cloudy, but i wasn't too concerned.​
6. Clean off oil, put blade between straightening jig during 425 temper. First temper was one hour, second temper was 2 hours
7. Grind blade to 400 grit, wash blade, and etch in FCL
a. After a quick etch I noticed that the areas that were previously hazy were still that way after etch (see 5b)​
b. After a longer 5 minute etch I noticed more of the blade was turning dark but the spots were still there​

This isn't the first time I've had this problem. It pretty much happens with every blade I heat treat and am wondering if it is common or if my process or even equipment is off somewhere.

Possible Solution: Should I raise the temp on my oven incase 1499 isn't really 1499? Should I let the blade soak longer?

The splotchy area is in the middle of the blade. It etches much lighter than the spine or edge. Does that mean that it didn't get as hard as the spine/edge?

Aside from the large lighter area after etch, there are also spots in the etch. It looks a bit freckled.

Any ideas or suggestions?

20120926_104447.jpg


20120926_104515.jpg
 
Last edited:
How long do you take to inspect your blade? I wounder if the thicker portion of the blade is acting like a heat sink, holding the temp too high when you are inspecting the blade. Just a shot in the dark.
 
If you mean after I pullit out of the oven, I don't take any time at all. It goes straight from the oven to the oil.

Forgot to say that the P50 is at room temp (about 75 degrees last night)
 
I ment during the quench cycle. I'm wondering if the 3 sec quench is not enough to cool the thicker mass of the blade. After you quench for 3sec you look at the blade to check for warp. If (Big if) the spine did not get cooled down enough to harden in 3 sec the air would could slow the cooling down enough to loose the hardness you want. I'm just a new guy at metal bashing so take my advice with a grain of salt.
 
Brad, that is a good point and has crossed my mind. The spine at its thickest point is 3/16 and the edge was around .010 as quenched. I know that's way too thin but I got grind happy with a new belt.

Next time I'll leave it in until a 7 count and see if that makes a difference.
 
Maybe I missed it but you didn't mention clay coating so I assume you were trying for what is often called and automatic hamon. That requires a shallow hardening steel which goes beyond the alloy chosen. It also needs a fine grain and your heat treating process, as you described it, will cause grain growth. The 1600° with a 10 minute soak was especially bad. This is well above what is needed to austinize the steel. The 1500° with a 10 minute soak was not as bad but still excessive. The 1499° austinizing temperature before quenching was not too bad if you did not soak any longer than to heat the steel evenly throughout. What has happened, whether you clay coated or not is that you grew the grain so large that the steel hardened all the way through, even in the thick sections, preventing the formation of pearetic steel in the spine and the martensetic steel in the edge. There was little, if anything, for the ferric chloride to differentially etch. If you break the blade, I'm certain that you will see very large grain.

If you want to salvage that blade you will need to go back and re-normalize it three time to reduce the grain size. You heat it to about 1475°, without soaking any longer than about 30-60 seconds, three times, allow the steel to cool to a black heat between each cycle. Or you could heat to 1500°, then 1450°, and finally to 1400° holding at temperature only long enough to heat the blade throughout. No more than one minute.

1084 is a simple steel and the carbides in it are mostly cementite so it doesn't require a long soak to and high temperature to get the carbides to release carbon into solution. If you want to heat it to 1500-1550° degrees only do so long enough to heat the blade through and then quench. I you want to soak the steel, heat it to 1450-1475° and hold at temperature for no more than 5 minutes then quench. Holding the steel in the quenchant for 3-4 seconds, especially from 1450° should beat the nose of the cooling curve. Do not temper the blade until it is warm enough to hold in your hand.

Doug
 
That's some good information, Doug. I had some blotchy etching going on myself a few weeks ago with 1095. Those two blades are awaiting a series of normalizations and a re-hardening in new quenchant. I also may have been going too long with the 5 minute soak at 1500.
 
I work 1095 alot but my HT procedures arent far off from yours.

I always do a 1600 normalization with a soak. At this point, similar to Dougs thoughts, I may be creating a large grain size but I'm worried about creating EVEN grain size with this heat not SMALL grain size. Although even at 1600 it's still a reduced temp heat from my forging heats.

I follow that with grain refinemant cycles at 1475 then 1450 then 1425 cooling to black in between.

Then I'll harden at 1425 for hamon or 1475 for fully hardening. In my understanding you should be fine hardening 1084 at 1500 because you dont have to worry about getting too much carbon into solution.

I run all these cycles in my Evenheat and always give a 10 minute soak just as you described. With a blade that's 1/4" on the spine it takes that long to get the blade fully up to temp in the oven so I consider 10 minutes really a zero soak. I've checked quite a few at 8 minutes and thicker blades will always still have a hint of shadow in the blade indicating it's not 100% at my target temp yet so I go with 10 minutes from when the oven gets back to temp after putting the blade in.

Contrary to Doug's opinion I dont think you've got a large grain size. It could potentially be even smaller with reduced temp refinement cycles from where your currently at with your HT but even my opinion is an assumtion that would have to be verified with testing.

I also think that 3 seconds in the Parks should fully harden your 3/16 1084. The danger of pulling your blade out of the quench at 3 seconds would be ending up with an auto-temper. That said though the flashpoint of Parks 50 is 275 so when you come out of the oil if it doesn't burst into flames your below the flashpoint so even an auto-temper shouldn't be an issue for your procedure IMHO.

My first guess for the cloudy etch would be

1. dirty blade going into the etch. You did say you washed it though so probably not this. I typically wash with Acetone then follow with Windex and finish with a wash of clean water in the sink then go into the etch right from the faucet without drying the blade or anything.

2. Decrab. If you ground the blade after hardening and before etching this shouldn't be an issue at least not a re-occuring problem.

3 Surface finish. It's quite likely that underneath your ground 400 finish is a pattern of rougher scratches that are blended over instead of removed completely. This scenario seems the most likely to me given your description. Handsanding out to 400 would give a clean etch if this is the case.

Pics may potentially help ALOT also.

As for spots in the etch. I've had this happen in the past and to prevent this I alway get my blade in the etch and gently swirl it around a little then lift it out and check that I'm getting a full even covereing of the ferric on my blade with no contaminates or air bubbles on the blade.

I'll be very interested to hear Doug's thoughts on this post !

Good luck with your etching.

-Josh
 
One thing brought up by Kevin Cashen, who has the facilities to do microscopic examination of the steel he heat treats, you can run into trouble soaking 1095 too long not only by growing the grain too large but by also putting too much carbon into solution and causing an increase in retained austinite.

Regardless of the steel, assuming that it's one of the simpler oil or water quenching steels, there is no need to soak for normalization other than to make sure that the inside and outside are at the same temperature. This does not take long. You are not looking to put carbon into solution. All you need to do is cause a complete phase change in the steel. All soaking will do is grow grain, especially when you get up to 1600° with a hypereutectic steel. If you want to test it then make a test blade, heat treat it as you have described and break it to examine the grain. Yes, forging is done at even higher temperature but that is the reason that you do a triple normalization, but you have to do that properly or all you are doing is spinning your wheels.

If you want to correspond with someone about this, I would recommend that you do so with Kevin Cashen. Look up one of his posts on this board and click on his name and that will open up a link to him. He has often offered to help people with the knife making issues. Just remember that he has a life and might take a few days to get back to you. He also very experienced with working with 1095. You note that you are an ABS apprentice so I assume that you will be wanting to test for at least your Journeymans stamp. Heat treating is one thing that you will have to get down right or you will not pass your performance test. If you are not testing your heat treating method by the making a test blade and attempting the edge retention and the 90° bend, you should be.

Also, I strongly believe that the heat treating process that Jason describes will result in grain growth, especially if he has a lot of steel that was deoxidised with silicon, even though aluminum will not totally offset grain growth.

Doug
 
Doug,

My understanding is the real issue with hypereutectiods is not the high heat and getting too much carbon into solution that can cause issues but the use of a slow cooling (annealing) that will cause issues of carbides in the grain boundries. Air cooling from heat avoids this problem of leftover carbon getting into trouble as I understand things. I totally agree that a soak at 1600 will likely result in grain growth if you already had a fine grained condition but also feel its becuase of the temperature far more-so than the time involved in a little soak like 10 minutes. Time will certainly have an effect but the Temp is the lion. Assuminimg grain growth though, I'm not even worried about grain size with this heat as I'll be setting up my grain size with with 3 more refining heats.

I do understand that a soak is not needed for normalizing our "simple" steels but also feel a short soak at a controlled temp doesn't hurt anything either other than a little more decarb than there would be with a true zero soak time. My main issue with the oven is you cant see the steel while it's heating so a true zero soak is all but impossible. This leads me to my 10 minute figure in my oven and enviornment, others mileage vary. It's far more important to me to get an even heat in the entire blade than to nail a zero soak and risk not being fully austinized.

With the OP's HT of normalizing and a quench I'd expect grain thats not ultra-fine but not large either.

While I haven't tested with a Mastersmith yet my test knife has been done for a long time but it's 5160 which is of course a different animal.

As usual I appreciate you sharing your thoughts !

Take care,

Josh
 
Let’s see if addressing this step by step may be able to clear up some of the questions both here and in other posts within this thread.

My Observations: Very cloudy and splotchy etch on 1084. Possible bad heat treat?

First let’s establish the expected outcome in order to ascertain our goals and how the current outcome is falling short of that. Is that a blade that is intended to have a decorative hardening line of sorts, and is the purpose of the etch to reveal that feature?

Assuming that is the case we will address the following:

1. Blade was forged, then normalized three times at reduced temps (by eye)

If you are following this up with full controlled normalization one cycle to get started is more than enough. The fewer times a blade sees heat, under uncontrolled (atmosphere and temperature) circumstances, the better; i.e. less opportunity for something bad to happen.

2. Profiled and beveled

I would seriously consider swapping steps 2. and 3. There is no need for a full normalization after simple grinding operations and normalizing, particularly with soaks, can be a very decarburizing operation. Grinding afterwards would allow you to remove decarb. If a post grinding heat treatment is indicated I would go with a simple stress relieving cycle of heating to 1200F and then air cooling.

3. Heat home built kiln to 1200 and hold for 20 minutes, then ramp to 1600
a. Insert blade, wait for temp to come back to 1600 and hold for 10 minutes.
b. Remove blade and cool in still air until black

First, if the kiln is home built double check the calibration, if you have not yet done so, to assure that the heating is even and accurate throughout. This is 1084 so you will have full solution at the lowest possible temperature and time of any steel common blademaking steel. A soak is beneficial to almost any steel but for 1084 it is not critical, I don’t think 10 minutes is too much at all considering time also needed for blade to fully assume temp. 1600F is a very good true normalizing temperature, the idea of which is to even out all internal components of the steel. Steels with carbon content greater than .85% benefit from the dissolving of the extra carbon and distributing it evenly throughout, but beware of slow cooling which will allow it to pool back up, so no slower than still air. 1084 will have little in the way of leftover carbon to deal with so it will quickly be dealt with at this temp and the grain size may very well increase, but that may even be a good thing on this first heat in order to homogenize it. Even grain size is of even more benefit than the size itself.

4. Ramp oven to 1499
a. Insert blade, wait for temp to come back to 1500 and hold for 10 minutes
b. Remove blade and cool in still air until black

Very good second thermal cycle to reduce grain size, it could even be followed with a third at 1450F. but be aware that the finer the grain size the shallower hardening the steel will be, thus moving your hardening line closer to the edge.

A note about soak times and grain growth:
Time equals temperature but not proportionally, temp always trumps time. The lower you go in temperature the longer your soaks can be. Almost all modern steels are what could be termed “fine grained” due to aluminum killing procedures, thus they will resist grain growth through most of the normal heat treating temperatures, almost regardless of time, but once the grain coarsening temperature is exceeded growth will be VERY rapid. Thus you can go from ASTM size 9 to ASTM size 3 in one heat if that heat is out of control. If, however your heat is tightly controlled and held under that given coarsening temperature the grains can be stable even in hours of soak times (this is especially true if you have an alloy steel). This is due to aluminum nitride particles contributing to grain boundary stabilization even beyond what carbide contributes. I am confident that upwards of 90% of all grain growth episodes that bladesmiths tell horror stories about were solely a matter of loss of temperature control, even if only for an instant.

5. Reinsert blade into 1499 oven, allow to come to temp, and hold for 5 minutes

If your goal is a hardening line, a temperature between 1450F and 1475F will give you more definition. Do be aware that as temp drops soak times can increase.

a. Remove blade and quench in P50 for 3 seconds, remove to inspect, then back in oil until cool (Blade was moved up and down to agitate oil)

3 seconds is way too short a count, you are flirting with upper bainite there, and even if you feel knives are for beating on things rather than cutting, lower bainite would be better, the upper feathery stuff sucks for just about everything. I am surprised you are not flashing the oil at that temp. There is no need to return to the oil if you get as close to Ms as possible, and the lower temp will be safer. If you have clay on the blade you will want a 10 count or better. If the blade is bare, a seven count will get you to just at the top of Ms and then you will be approximating a marquench, if that is your goal.

6. Clean off oil, put blade between straightening jig during 425 temper. First temper was one hour, second temper was 2 hours

What size blade is this? For a small cutter you could start out at 400F instead, for a larger chopper, go for it. Be aware the higher tempering temps will somewhat fade really vibrant hardening lines.

b. After a longer 5 minute etch I noticed more of the blade was turning dark but the spots were still there
This isn't the first time I've had this problem. It pretty much happens with every blade I heat treat and am wondering if it is common or if my process or even equipment is off somewhere.

For hardening lines FeCl is not the ideal etchant, if you nail the process the lines should pop all on their own without the need for etching, but something like apple cider vinegar will work better. The idea is to have a very weak etchant that will take a long time to attack the microstructures at the top of the line differentially, more aggressive reagents tend to attack more evenly.

Possible Solution: Should I raise the temp on my oven incase 1499 isn't really 1499? Should I let the blade soak longer?

Possibly longer but first let’s examine the answer to the next question more…

The splotchy area is in the middle of the blade. It etches much lighter than the spine or edge. Does that mean that it didn't get as hard as the spine/edge? Aside from the large lighter area after etch, there are also spots in the etch. It looks a bit freckled. Sorry I didn't get pictures, I know that would have helped. Any ideas or suggestions?

Without pictures it is very hard to say, but what you describe sound a lot like decarb. I have found several areas in you procedures that would result in decarb and allow it to survive to the etch. You don’t need more and a few microns of decarb to throw the etch off. Splotches and light “freckles” are classic decarb descriptions. Try to move all of your normalizing to before the grind so that you are removing any decarb it causes, and then watch your atmospheres during your hardening, especially with extended soaks in a kiln. Anti-scale compounds or foil envelopes will help.

Do not confuse scale with decarb, they are two different things but do result from many of the same factors. If you helps, I would be willing to say that most makers would see the same splotches if they dipped all of their blades in FeCl , decarb is very perniciously ubiquitous.

A final note:

Since an etch will show what's happening inside the blade, I think this is more of a heat treat issue than poor etch.

This is a very erroneous concept that unfortunately is perpetuated in places like knife magazines, by folks who want you to think they have a clue about what goes on inside of the steel but most often reveal that they don't. To actually know what is going on inside the blade requires a bit more processing than a dip in FeCl for the entire blade. Logic would dictate that in order to see inside the blade you need to actually reveal the inside of the blade. This is done by proper cross sectioning and the use of special reagents (none of which resemble FeCl) designed specifically to reveal particular microstructures. As you are experiencing, the slightest issue on the outside of the blade will throw a wrench in the concept of seeing into the blade from the outside. You may get an idea of zones of hardness or different microstructures in a very general way, but actual hardness, grain size or exact microstructure are not revealed by the FeCl etch despite all the deceptive marketing claiming otherwise, and a little decarb will rob you even of that.
 
Last edited:
Kevin, I am honored that you would take the time to help me resolve this issue and educate me on the subject.

I'll try to address your thoughts:

First let’s establish the expected outcome in order to ascertain our goals and how the current outcome is falling short of that. Is that a blade that is intended to have a decorative hardening line of sorts, and is the purpose of the etch to reveal that feature?

Actually, my goal in etching was not to create hamon type lines because I figured 1084 wasn't suited well enough for that. The etch has/had two purposes. One is to varify that the heat treat was even throughout the blade. The second is that I like to add a patina on all carbon blades that leave the shop for rust proofing purposes. I think I'll switch to gun blueing for this as it seems like it might give it a more even color.

I would seriously consider swapping steps 2. and 3. There is no need for a full normalization after simple grinding operations and normalizing, particularly with soaks, can be a very decarburizing operation. Grinding afterwards would allow you to remove decarb. If a post grinding heat treatment is indicated I would go with a simple stress relieving cycle of heating to 1200F and then air cooling.

I will do this on future blades. What if in the course of grinding I get the edge of the preHT'd blade blue hot. Will that cause me to have to renormalize or is it OK?

First, if the kiln is home built double check the calibration, if you have not yet done so, to assure that the heating is even and accurate throughout.

I have ordered 1450 and 1500 Tempisticks to check the temps. Should I also get a pyrometer as a backup?

3 seconds is way too short a count, you are flirting with upper bainite there, and even if you feel knives are for beating on things rather than cutting, lower bainite would be better, the upper feathery stuff sucks for just about everything. I am surprised you are not flashing the oil at that temp. There is no need to return to the oil if you get as close to Ms as possible, and the lower temp will be safer. If you have clay on the blade you will want a 10 count or better. If the blade is bare, a seven count will get you to just at the top of Ms and then you will be approximating a marquench, if that is your goal.

Noted

What size blade is this? For a small cutter you could start out at 400F instead, for a larger chopper, go for it. Be aware the higher tempering temps will somewhat fade really vibrant hardening lines.

The blade is 5.5" long

Without pictures it is very hard to say, but what you describe sound a lot like decarb. I have found several areas in you procedures that would result in decarb and allow it to survive to the etch. You don’t need more and a few microns of decarb to throw the etch off. Splotches and light “freckles” are classic decarb descriptions. Try to move all of your normalizing to before the grind so that you are removing any decarb it causes, and then watch your atmospheres during your hardening, especially with extended soaks in a kiln. Anti-scale compounds or foil envelopes will help.

Do not confuse scale with decarb, they are two different things but do result from many of the same factors. If you helps, I would be willing to say that most makers would see the same splotches if they dipped all of their blades in FeCl , decarb is very perniciously ubiquitous.

I think you hit the nail on the head. Here are some pictures. After hand sanding one side then flipping to the other I noticed some decarb in the area that didn't etch properly. It was fairly deep and wouldn't sand out with 400 grit so I'll either switch to a higher grit to take it out or just put a machine finish on this one as that is now what the customer requests.

Here are 1000 words:
20120926_104447.jpg


20120926_104515.jpg


Looks like decarb it is. Thanks for the help. Next time I'll try to remember what I should have done in the first place, which is forge it thick, grind it thick, HT, then grind it thin.

Thanks for everyone's help.
Jason
 
I will do this on future blades. What if in the course of grinding I get the edge of the preHT'd blade blue hot. Will that cause me to have to renormalize or is it OK?
Don't sweat it, any issues from heat treating on a grinder are in the realm of strain energy which is very easily wiped out by recovery in the temperatures of a simple stress relief.


I have ordered 1450 and 1500 Tempisticks to check the temps. Should I also get a pyrometer as a backup?

I wouldn't spend too much but use things like the sticks to help, a reliable pyrometer is pretty handy. Folks often wonder how I know my temps are on with my units, the answer is redundancy, at any given time I have no less than two thermocouples verifying each other and between type "k" switches and three simultaneous operating readouts I often have three.



I think you hit the nail on the head. Here are some pictures. After hand sanding one side then flipping to the other I noticed some decarb in the area that didn't etch properly. It was fairly deep and wouldn't sand out with 400 grit so I'll either switch to a higher grit to take it out or just put a machine finish on this one as that is now what the customer request...
...Looks like decarb it is...

You are that much ahead of the game because of the etch, most folks never detect the decarb unless they are etching damascus, on carbon steel it is unseen on a polished blade.
 
A pointer on pyrometers. Most that I looked at did not come with thermocouples (probes) that were for temperatures up in the ranges that we use. I think that a few of them had one's that would work up to about 1400-1600°. A good place to look would be where they sell pottery/glass kilns and supplies for them. The last time that I bought some pyrometers and thermocouples that exceeded a 2000° working range was actually on Ebay. Make sure that the display one the pyrometer will register in the highest temperature range that you think that you will use. Don't assume that just because the pyrometer will register up to a certain point that the thermocouple will operate in that range too unless the add says that it will.

Doug
 
Thanks Doug. I got a pretty good deal on ebay for one that comes with one of the small thermocouple. My plan is to cut the thermocouple off and wire it into the existing high temp thermocouple on the oven. I know this will give me only one thermocouple instead of the ideal two but I don't want to drill another hole into the oven just yet. It's a bit crowded in there.

Is there any reason I couldn't run two devices on one thermocouple? Would one interfere with the other?
 
Before you detach the existing thermocouple from the oven make sure that you want to disable whatever it controls. If all it does is display the temperature inside and not other regulator runs off it, all you will be doing is changing from one pyrometer to another. I would install a second thermocouple in the oven and run both. Ed Caffrey would be a good person to ask about doing this. He's battling some health problems now but he still posts rather regularly and I doubt that he mind answering you question is you send him a quick email.

As far as monitoring two devices goes, as long as the thermocouple has a plug that matches the socket on the pyrometer then you can switch the pyrometer between as many devices as you want. If you have a duel display you can even plug two thermocouples, each from a different device, into your pyrometer. You will just have to make sure which thermocouple the pyrometer is displaying.

Doug
 
I don't even plan to test for the ABS Journeyman's stamp. Besides the issues that I have with how they handle their rules it also requires time and money. I'm retired, so I have plenty of time but with a combination of factors associated with preparing for the test and testing, it's just not worth the investment to me just to have the bragging rights for having a stamp. If I had any intention of doing knifesmithing as a source of income, I'd suck up my problems on how they handle things and do what I needed to do to pass the test for my stamp. I have still made knives with the express purpose of testing them to destruction so that I could determine the quality of the blades. Especially if you don't have the resources to hire the testing or accumulate the equipment to do lab testing, performance and destructive testing is all that is left to you and you should be doing it to check your proceedures. You don't want to wait until you're taking your performance exam to do destructive testing for the first time.

Reading Kevin's reply did show me some things that I hadn't considered in my reply. One was the time required for the steel to come up to temperature in a kiln. You can't very well check for decalescence with the oven door closed and opening it to stick a magnet in it would release the heat. Decalescence and lack of magnetism are what I rely on with heating the steel in my forge. I'm also not clear why you are wanting to do this 1600° with a 10 minute soak and then do three more normalization cycles following it. This, as I understand what Kevin said, should be your first normalization cycle with two more a decreasing temperatures. There is no reason to do a stress relieving cycle then three normalization cycles. If you do your three normalization cycles before rough grinding then you would want to do one stress relieving cycle before hardening and tempering. If you do your rough grinding before you normalize three times, then do a single cycle when you finish forging.

Doug

Also, in hypereutectic steel you are going to have carbides form between the martensite crystals. If they didn't form, then this "left over" carbon would form graphite and weaken the steel. The problem that annealing from above critical temperature causes is that it gives time for these carbides to form clumps and increase brittleness. The problem with excess carbon being dissolved in the austinite, as Kevin has explained it, is that it can physically prevent the austinite from converting to martensite upon quenching and increase the amount of retained austinite present in the steel.

From the texts that I have hypereutectic steel is more prone to grain growth than eutectic or hypoeutectic steel. That's why hypoeutectic steel can be austinized at a little higher temperature than hypereutectic steel
 
Back
Top