Heat treating

D

db

Joined
Oct 3, 1998
Messages
3,762
I see quotes like this a lot
“It’s fact that a "one-off" or small batch heat treat will be more
consistant than a large batch. So, knife A has at least a more consistant if not flatly
better heat treat.”
Is it just a theory that large heat treat batches are some how inferior than a smaller one? Sure it sounds reasonable but in real life is there any facts to show it’s true? Is a small heat treatment batch done by XYZ better than a large one done by Bos?
 
It's not difficult to understand.

Temperature distribution in a large oven is not uniform from edge-to-edge.

For a large batch of blades, if the blades in the center are 'perfect', then the blades on the edges and in the corners are less than perfect - or vice versa.

Practically, it makes no difference if you're heat-treating a large batch of blades or baking a large number of loaves of bread.
 
gud4u that does sound simple, thanks. Your post makes me think of a few questions though. I thought heat raises, so would the temps be the same at the same level? Would an oven be hotter on the top than the bottom, or does it matter how the oven is vented? How much of a temp difference makes a difference 10 degrees, 5 degrees, 1 degree at the temps used for heat treating? Is there really enough difference in temp from the middle of a large batch than the edge? Is this temp difference enough to effect the heat treatment? Is there a bigger difference in the temp of the blades themselves, thin tips and thicker spines? I guess I’m assumeing because it’s a large batch it’s different than a small batch. I’m not even going to ask about salt baths, I don’t even know if they are used or if ovens are, or if even either one is used. It is easy for me to assume a small batch is better heat treated than a large one but I guess I really don't know if that is true or not.
 
Temperature distribution in a large oven is not uniform from edge-to-edge.
Click to expand...

Are you implying that it is uniform in a smaller oven (more than 1 piece treated at a time)?
 
db said:
ow much of a temp difference makes a difference 10 degrees, 5 degrees, 1 degree at the temps used for heat treating?
Click to expand...

Clark gives some information specific to 52100 on his webpage on salt pots, you can run a simple linear interpolation to see the effect per degree and then compare the effect the %carbon makes to hardness by looking at the graph kell_aa posted awhile ago to see the kind of effect you can expect per degree shift on hardness. However note that steel varies as well. Look at the carbon and chromium variances for 440C and think about a blade from one end to the other. When you run really small batches you can adjust for this by altering the tempering, Wilson does this for example as he HRC tests each blade. You can't expect production compaines to do the same and thus they will have more variance.

http://www.mvforge.com/salt pot info.html

The custom makers who are really serious about performance will individually test knives from each batch of steel (quenching media, etc.) to confirm they are all on target. However it isn't that uncommon to have production companies comment on problems with batches reaching the public both in terms of geometry and the steel. If you want specific information on ovens just contact a manufacturer and ask for specific details. Realize though this is only one small factor, you also have issues like the time at temperature, effect of preheating, how they are removed, how long before tempering, etc. .

I’m not even going to ask about salt baths, I don’t even know if they are used ....
Click to expand...

Several custom makers are using them, I have not heard of a production company doing so yet. They do solve many of the temperature problems with ovens. One of the major problems with heat treating aside from austenization is the tempering and they addess that as well. Otherwise lack of agitation is usually an issue, especially in air cooling. Just think about a blade surrounded by numerous other blades compared to ones on the outside which only have blades on one side of them. Now think about the temperature gradient across that essentially really coarse crystal structure.

No I didn't call Bos, the cost of a long distance call would be the price of a tool steel textbook and generally I don't ask serious questions about performance/steel on the phone from maker/manufacturers because it would be all private information and not repeatable thus useless from an academic point of view. In private since there is nothing to insure responsiblity people tend to say things they never would in public. I have heard lots of things about knives/makers in such conversations which are staggering if true, but they are just rumors, nothing more.

Swordforums is probably the best user interactive blade specific heat treating/materials resource now. Cashen specifically has a crazy amount of resources dedicated to actually understanding steel. He even bough an impact machine awhile back. He is also one of the few makers who will publically comment about the hype about testing / steels / heat treating. If you have not done so, scan through the Metallurgy sub-forum there, specifically for example look at the comments regarding the effects of forging, high cycle normalizing, multiple quenching, etc. . The list goes on, he has also wrote several very nice articles about the myths of modern blade making which are worth reading.

-Cliff
 
I think Camillus uses or has used salt pots for heat treating if true then that would be one. From your link Cliff Bos says he HRC tests alot of Bucks production heat treating, I assume he makes ajustments acordingly. I'm also guessing from your posts you don't know either if large batches are any better than small ones. Does anyone really know? I see it said alot but no one really says why or by how much. I guess I'm just no longer going to assume they are, or maybe I will.
 
IIRC (I don't know if this is still done, must be osha laws against it or something) Camillus used molten lead for their heat treating.
 
My knives that saw heat-treatments from Phil Wilson are better than when they were factory fresh. One to three points harder on the HRC scale. Only one of my Spydies was close enough for the heat-treatment to be an extravagance. The rest are all easier to sharpen and keep sharp.
 
All of the blades (several hundred) that Paul Bos has heat treated for me had Rockwell marks, some up to eleven marks. There are many ways to check hardness besides a Rockwell tester. Scatch blades or pieces of steel of a known hardness particularly in experienced hands come quite close. After belt sharpening I can come fairly close to a quadrant of wear resistance-hardness questimation. Having a Rockwell tester is nice was to teach yourself and reaffirm evaluations.

Phil Wilson constructed his own ceramic encased coil high temp kiln. He has the right equipment to heat treat high alloy steels, in particular early entries into the cutlery world by Crucible steel. Myself and many of my friends quit making fillet knives after we saw Phil's fillet knives; they were perfection. Phil has done a lot of work with Crucible products, he does it right...Take Care...Ed
 
db said:
...Does anyone really know?
Click to expand...

Yes db, there really are thermodynamic and statistical reasons why batch processing has larger variances than small scale / individual and why these difference will scale with the size of the batch. The reasons are many as were noted. There are also numerous direct examples of this on the forum both in regards to geometry and heat treating. The reported frequency of problems with S30V in production vs custom for example. I have personally seen S30V blades which could not even be sharpened as they would crack apart under the stone. This is not going to happen when you have guys hand crafting each blade.

The extent of the variance depends on the desire and willingness of the manufacturer to take steps to minimize it. I have seen much larger variances in performance from Ontario for example than Swamp Rat. This would be expected given the difference in blade output of the two companies plus the added QC steps taken by Swamp Rat such as the crack scanning they do as well as the extensive R&D into heat treating / steel in the first place.

As for making corrections, after Wilson HRC tests he adjusts the temper accordingly, do you really think a production line can do this for every blade, of course not, it would not be production if they did. They are working inhernetly with an averaging/smoothing process at a fundamental level. As you reduce the size of the batches the effect of this smoothing becomes smaller and smaller until it is removed with single piece manufacturing.

The easiest way to understand this would be to imagine for example two tree lots. They both have the same trees, the same soil and get the same general enviromental conditions as they are located right next to each other. However in one lot they are each taken care of individually. Any problem with a tree is immediately taken care of specifically. In the other lot anything which is done to one tree is identically done to all trees if they need it or not. The consequences should be obvious.

To clarify though, this doesn't mean all small scale is superior to all large scale. It just means that fundamentally it can be, just like there are fundamental differences in steels but in how much of these are realized depends on the effort of the maker/manufacturer. It differs significantly depending on how important performance/quanlity is compared to the monetary bottom line.

-Cliff
 
Let's work with the surprise principle: would you be more surprised if you found out that the variance in terms of cooking time/ingredient ratio/cookie quality is less from Nabisco using monitored pass-through ovens than your granny using her kitchen oven? Or would you be more surprised if you your granny's cookies had less variance?

Of course makers like Phil Wilson can hardly be equated to granny here. I can see Nabisco not being able to respond to changes as well as granny though. But in terms of company A processing 100 pieces per day and company B processing 1000 pieces a day not much can be gained from that alone. Prehaps company B can afford to spread out the cost of more precise equipment over the 1000 pieces whereas A cannot.
 
“
Cliff said..
To clarify though, this doesn't mean all small scale is superior to all large scale. It just means that fundamentally it can be, just like there are fundamental
differences in steels but in how much of these are realized depends on the effort of the maker/manufacturer. It differs significantly depending on how
important performance/quanlity is compared to the monetary bottom line.
I think that is probably the best answer to my question. It seems like a lot of people just assume a small batch, or even a single blade heat treatment is better than a large batch heat treatment. I don’t think I agree that is true. Sure in theory it is possible, but that doesn’t mean in real life or for all or even most cases it’s true. If a blade is done by Wilson sure I can believe its going to be a good/great heat treated knife. But if a small batch or even a single blade is heat treated by Dbflame inc. that doesn’t mean it is better or even as well done as a Bos heat treated blade. Even if DB flame used the same type of equipment and heating directions.
I do find this very interesting in Thom’s post
“
thombrogan..
My knives that saw heat-treatments from Phil Wilson are better than when they were factory fresh. One to three points harder on the HRC scale. Only one
of my Spydies was close enough for the heat-treatment to be an extravagance.
I have to wonder is the target HRC the difference, or is it the quality of the heat treatment? Or are they one in the same? And was the one that was close enough was it from a good batch or one of the good ones in a batch?
 
Good question, db. Unfortunately, the good'un was an S30V ATR blade, so that's a mystery not to be found out anytime soon.

Ed Schempp,

I completely agree with you that Phil is an expert at using the right steel for his customers' needs. Cliff and I are trying to corrupt him into trying M2. :D

kell_aa,

Please don't call Phil anyone's grandmother. That's just not right.
 
db said:
It seems like a lot of people just assume a small batch, or even a single blade heat treatment is better than a large batch heat treatment.
Click to expand...

In general people realize that you are speaking of probabilites when you make such a statement, such as the fishing is better in one location than another. It doesn't mean that every time you catch more fish there just that you are far more likely to and the results are more consistent.

I have to wonder is the target HRC the difference, or is it the quality of the heat treatment? Or are they one in the same?
Click to expand...

It takes a fair amount to move a knife significantly in hardness. In many steels shifting the hardness by such a small amount can move it from a toughness maximum to approach a minimum. Examine the torsional impact/strain curves and then consider the effect of just small shifts in hardness.

-Cliff
 
I know I probably shouldn’t but I’ll do it anyways.
“
Cliff said..
It takes a fair amount to move a knife significantly in hardness.
Is it enough difference in a large batch from the center of the batch to the edge of the batch to make any difference? Large batch,small batch. Does it make a difference if done by a quality heat treater? Ixsample is there a difference if Wilson does 1, 2, 50, 100, or what ever number that would be his largest batch?
 
thombrogan said:
...trying to corrupt him into trying M2.
Click to expand...

Forget M2, I want a multi-longitudinal laminate of Carpenter Maxmet/L6 with alternating micron thickness layers and the L6 in full bainite and the Maxmet at full secondary hardening. I'll settle for a small paring/shop knife, I am not hard to please.

db said:
Is it enough difference ...
Click to expand...

The differences in reported hardness reported by Thom and others on production knives is significant enough to have a large effect on toughness as noted, check the torsional data for specifics. What you see of course is the total variance, not simply the temperature gradient in the oven. You could compensate for that for example by adjusting which knives were in which location, as some knives would benefit from a lower soak (it generally rasies toughness at the cost of wear resistance). If you really want to know oven specifics just talk to a manufacturer as noted. This however doesn't tell you how the ovens are used or isn't the only factor. The true reflection is just the end variance seen in custom vs production.

-Cliff
 
db I did build my own furnace and learned a few things doing it. I wanted to be able to go up to 2200 with out worrying about element burn out and also wanted a uniform temp with in the furnace. To do this I used industrial elements and placed them on both sides and on the top. The bottom is a ceramic plate to even out the heat down there. I have 4 thermocouples inside. 3 just to monitor the temp and one is input to the controller. In spite of all this if I do a temp survey inside I find some places where there is a 100 F difference within 6 inches. That surprised me. 100 degrees can make a big difference in hardness.
Thermocouples can drift over time especially when they are operating in the 2000 range. To slow this down and to get longer life I went to a heavy wall Inconel sheath. This helps and now can get up to 100 hours on them. The other thing not realized is that hardness testing is not as precise as some think. The Rockwell certified test block is accurate to .5 pt. The machine on a good day is probably good to about the same .5 pt. So if all this adds the wrong way it is easy to be off a full pt. even with the best equipment. To compensate for all this I learned to use the center of the furnace where the temp is most even, overheat a little and let the temp average out over the length of the blade, check the temp with a fresh thermocouple inside the furnance every 5 hours or so and use the test block for reference every time a blade is hardness checked. The subzero cycle tends to even all this out so is a good quality control step. A too hard blade can be tempered a little hotter. This is better than tempering cooler to get a higher finished hardness because in doing so you can sacrifice toughness. I do most tempering in the kitchen convection oven. It is with in 20F in the center at 400F. Furnace location and temperature makes a difference, quench rate is critical, so the final result is going to vary. The trick is to make a blade that performs and then to do the same thing again.
As far as batch heat treating is concerned if you are a production heat treater for a knife company you would want to err on the side of lower hardness rather than higher. Knife companies do not want broken or chipped blades back. To be realistic IMHO you would have to work in a range of probably 2-3 pts Rockwell. It would not be feasible to heat treat blades one at a time and shoot for a range of 1 pt Rockwell. So if you aim at RC 60 on the high end to be safe some blades may come out at 57.
Paul Bos is a custom heat treater and does a hardness check on every blade he heat treats. He has been doing this for at least 20 years and I believe his work would be within 1 pt. He helped me get started heat treating, and he was generous with sharing his knowledge. I hope this helps to put some of this in perspective… PHIL
 
You must log in or register to reply here.
Back
Top