Preferred RC Hardness

[Marko Tsourkan]

A question to the knife makers.
What is a preferred hardness on your knives? It seems that 60-61RC is very common and 62 not too common. What about 62-63RC for steel that have excellent edge stability like 52100? Overkill?

Marko

 

[Delbert Ealy]

I do not use 52100 as my steel of choice, and thus I cannot answer your question directly for that steel, however with all steels hardness is a compromise between hardness and toughness. Going too far in one direction may result in less than satisfactory results.
I have often found that this question can be answered by test-to-destruction tests, it also helps give you a better undestanding of the steel properties at certian hardnesses.
Thanks,
Del

 

[PierreRodrigue]

A risk is there if RC is pushed, that you increase the likelyhood of edge chipping, and reduced toughness. Try a grind on a few scrap pieces of the steel you want to use, do a brass rod edge flex test, and check for chipping. It will give you a level of confidence with regards to how far to push your hardness.

 

[l r harner]

here is where i come on and bash the brass rod test as a test for hardness.
one it is mostly used as a test in the heat with a torch and quench in any old oil and see how the edge holds up (i like to think that wiht working with edges as extreme as the ones we do that we have a bit better way of HTing)
2
i can make a blade pass the rod test at 64rc and if i want i bet i coud make one fail at 58

the brass rod test was from what i can understand was mostly used to test razors but it was not a test of the HT it was a test of the grind and to make sure that it was ground to the proper thinness

in that use a razor would have to be not HTed and the blade woudl take a set or not tempered and maybe it would brake (but not at the edge)
if the brass rod test is used to see about the edge its more about if you have grossly missed the HT and even that can bee covered up by how the blade is ground (both if its too thick and unyeilding or too thin )

i have a 63rc blade that i can bend to 90 degrees (another of those fun tests) and it comes right back to straight (and its even cpm154 SS) no chipps cracks or any other problems

 

[PierreRodrigue]

I don't use a brass rod test for hardness testing. It is a way to get the edge to flex past its normal line, to deform it as it were. It isn't ment to replace a Rockwell tester, or any other meathod of determining hardness, just another thing a maker can use to see what the edge will do. Your right you can beat the test, sure, To call it a test might me the wrong term, but it will allow you to flex an edge past normal, and if there is chipping, it (the test) has helped you out, wouldn't you say? At63, with that much flex, wouldn't you be fairly confident in saying the edge will be fairly chip resistant. Whether you use a rod, pair of plyers, a vice or whatever, makers sill test edge flex, to check for chipping., or deformation. Then determine if an end user would go that extreme and decide if an adjustment to the HT is warented. Otherwise why bother to flex a given edge 90* if not to "test" it? Have I used a brass rod test sure, do I still no, not much, but I will if I am trying a new steel, or adjust a HT, I'll also bend it till it breaks or chips out, tested to distruction as it were.

 

[l r harner]

no worries i just see too many smiths fall back on that as proof that there HT is good

how bout this
make a knife and then sharpen it
then drag it sideways across the board like some might do to "clear" the board
does the edge chip or roll? but what if the edge is too obtuse and you dont see any problem? does this mean you nailed the HT ?
does any of that show much about HT or blade hardness

im not after any makers so dont think im tring to nail any one down its more about making sure we are all thinking about what we are doing after all we have seen many makers adn HTs that work great for the knives they make and many at different RCs

steel like knife shapes are all about trade offs in strength, sharpen-ability, reactivity and many other options

 

[Marko Tsourkan]

Thanks, guys. I can't say I understand brass rod test very well, but I will read up on it.

I heat treated a few knives to 63RC, so I think I will grind one and put it to test. If need be, I can draw it down by a point later. I was just curios, as my last one was 62RC and I beat c*** out of it without any adverse effects, so I wanted to see if at 63RC I will be able to do the same.

Many Japanese makers harden their blades to 63RC and Carter does it as well and as posted by a person who works for him, there is no issue with chipping. White steel has less of an edge stability of 52100, so that was another reason why I wanted to give 52100 a try at a higher hardness.

I will grind the blade to about .007 above the edge, so at this thinness if it is brittle, I should know right away. Standard cutting includes everything vegetables, cheese, meats, etc., push, pull cutting, down-cutting and down-chopping.

M

 

[l r harner]

Thanks, guys. I can't say I understand brass rod test very well, but I will read up on it.

I heat treated a few knives to 63RC, so I think I will grind one and put it to test. If need be, I can draw it down by a point later. I was just curios, as my last one was 62RC and I beat c*** out of it without any adverse effects, so I wanted to see if at 63RC I will be able to do the same.

Many Japanese makers harden their blades to 63RC and Carter does it as well and as posted by a person who works for him, there is no issue with chipping. White steel has less of an edge stability of 52100, so that was another reason why I wanted to give 52100 a try at a higher hardness.

I will grind the blade to about .007 above the edge, so at this thinness if it is brittle, I should know right away. Standard cutting includes everything vegetables, cheese, meats, etc., push, pull cutting, down-cutting and down-chopping.

M

thats the best testign as it will tell you if you can use that hardness it can also help you fine tune how thin you grind your blades
if you start hard and thin you can always sharpen the blade till it stops chipping/rolling the edge (it will get slightly thicker ever time you "fix" a chip) or temper down a bit and keep your edge thickness the same

my first big buy (over 200$) that i made in knife making was a kiln and then LN dewer
ugly knives that are well HTed are still great knives beautiful knives that are poorly Hted are little more then letter openers

 

[PierreRodrigue]

That is a good way to go Butch/Marko for sure, I didn't mean to sound so hard there, the house was a zoo this morinig, and I was typing while trying to bring order back!

At the end of the day, HT, grind, test, adjust, test again. Be sure of your meathod, test till your confident, and your work will hold up.

 

[Bill Burke]

Marko 52100 will hold up fine at 63 If you have refined the grain sufficently. I do this with multiple quenches after forging then normalize three times then sperodize aneal and grind. I then tyriple quench and triple temper then finish grind and polish. I still use a torch to harden and an oven to temper. However I would be willing to change my ways if someone can show me a way that works better.

 

[Marko Tsourkan]

Thanks!

 

[l r harner]

Marko 52100 will hold up fine at 63 If you have refined the grain sufficently. I do this with multiple quenches after forging then normalize three times then sperodize aneal and grind. I then tyriple quench and triple temper then finish grind and polish. I still use a torch to harden and an oven to temper. However I would be willing to change my ways if someone can show me a way that works better.

so why not use the kiln to heat for quench bill? least for the kitchen there is little to no need for a soft back or the ability to bend 180degrees many times (maybe on a single bevel to straighten after honing)

is it that you like the edge to have all the grain refinement ? why not make the whole blade just as refined? is there a benefit to having a spine thats not as refined as the edge ?


if you still wanted just a hard edge you could always just edge quench

i love the kiln for repeatability while im not questioning your heat color eye witha torch (ok maybe a little ) why not have a +/- of under 2 % and know for sure what your temp was (yes i know kilns can go bad too)

 

[Marko Tsourkan]

Evenheat kiln will have some variation in temperature (one zone and a temp probe at the top). You can figure what the likely temp of your piece is by some experimenting heat treating and measuring hardness, but you still might be off by 10-25 degrees at low temps and more at high.

Sugar Creek kiln is better this way, as a probe is on the side, just above the floor level, but it has two rows of coil instead of 4 of Evenheat (it will heat up more slowly), and at $200 less, it might not be a better deal over Evenheat.

In that sense, you don't get precise temperature measurement, unless you add additional sensors into the kiln or able to measure temp of heated piece with an infrared thermometer (industrial grade - very expensive, all others - inaccurate). To know at what temp you heat treat and temper, you have to do a bit of experimenting, testing for hardness and recording your findings. There will be some guess involved at all times.

Which is to say that it will be not that different in terms of accuracy than torch heat treating. Heat treating with torch (by eye) could be pretty accurate, as long as your eyes are not failing you and you test your work for hardness and cutting ability regularly, as I believe Bill does.

M

 

[l r harner]

also good to have a baffle for at least the sides of the blade to keep the heat even and not being directly "blasted" buy the coils when they cycle

the added mass also helps keep the temps for swinging wildly

and for the guys that temper in there home oven or toaster oven a brick or 2 will help leaps and bounds (keep the blade between the heated bricks )

 

[Bill Burke]

so why not use the kiln to heat for quench bill? least for the kitchen there is little to no need for a soft back or the ability to bend 180degrees many times (maybe on a single bevel to straighten after honing)

is it that you like the edge to have all the grain refinement ? why not make the whole blade just as refined? is there a benefit to having a spine thats not as refined as the edge ?
if you still wanted just a hard edge you could always just edge quench

i love the kiln for repeatability while im not questioning your heat color eye witha torch (ok maybe a little ) why not have a +/- of under 2 % and know for sure what your temp was (yes i know kilns can go bad too)

One of the things that helps with fine grain is how quickly you can heat the steel to austenitizing temps. An oven is not very fast. Salt or lead is pretty fast and I have been doing some playing with salt. a torch can be very slow or very fast or in between what ever I want. I can also dictate the size and shape of the hardend portion of the blade, From just the edge to the entire blade, much better than heating the whole blade and doing an edge quench. Lastly I have done it this way for so long and I get such good results that I am very reluctant to change things now. Not that I won't change but there will have to be some pretty strong evidence and definitive proof before I do.

 

[Marko Tsourkan]

What do you use for a baffle, Butch? I got two extra bricks with Sugar Creek, wonder if this is what they are for.

M

 

[jmforge]

Roman Landes swears that W2 can and should be left harder than "normal" even for field knives. I think he leaves his at around 62RC.

 

[Delbert Ealy]

Evenheat kiln will have some variation in temperature (one zone and a temp probe at the top). You can figure what the likely temp of your piece is by some experimenting heat treating and measuring hardness, but you still might be off by 10-25 degrees at low temps and more at high.

Sugar Creek kiln is better this way, as a probe is on the side, just above the floor level, but it has two rows of coil instead of 4 of Evenheat (it will heat up more slowly), and at $200 less, it might not be a better deal over Evenheat.

In that sense, you don't get precise temperature measurement, unless you add additional sensors into the kiln or able to measure temp of heated piece with an infrared thermometer (industrial grade - very expensive, all others - inaccurate). To know at what temp you heat treat and temper, you have to do a bit of experimenting, testing for hardness and recording your findings. There will be some guess involved at all times.

Which is to say that it will be not that different in terms of accuracy than torch heat treating. Heat treating with torch (by eye) could be pretty accurate, as long as your eyes are not failing you and you test your work for hardness and cutting ability regularly, as I believe Bill does.

M

My kiln was made by evenheat, but it is not a traditional knifemakers kiln. Mine is vertical 36 inches high and has 16 rows of coils. My controller(Omega) and thermocouple was not included and I put that system together myself. They are located in the middle of the kiln right where the blades go. I have tested my system with yet another thermometer and my system has an accuracy of 2 degrees even in the 1500f range.
I believe in constant testing and even test-to-destruction and I do that to insure quality control.

I strongly believe that making a blanket statement about the accuracy of all kilns over other methods of heating is inaccurate and misleading.

 

[Marko Tsourkan]

...

I strongly believe that making a blanket statement about the accuracy of all kilns over other methods of heating is inaccurate and misleading.

I was not referring to vertical kilns, or salt pod kilns, but horizontal kilns in general and from Evenheat and Sugar Creek in particular (have both), basically one temperature zone kilns. I would bet that one would get different readings if one installs two more temperature probes in this type of kilns. For the users, there will always some guess work involved when heat treating at higher temperatures (Evenheat kiln temp difference near the probe and the bottom where a knife is located is over 25 Degrees), so to know where you are, one has to do a variety of tests at different temperatures and then test for hardness (not with files, but with an actual hardness tester). Then temper, measure hardness, destruction test and grain analysis. After some trial and error, one can come establish the optimal heat treating temperatures.

What I do find misleading is when people state particular hardness on a finished knife without using a hardness tester. I accept that experience will help one to establish a range, but not an actual number.

M

 

[ajhuff]

In my previous life my boss showed me his Puma pocket knife one day and you could see the Rockwell indentation on the knife and they scribed the number right next to it. Very impressive to me. I've never seen that on any other knife. He told me it was an old knife and even Puma had stopped doing it. Shame, I thought it was pretty classy.

I still have problems with Rockwell reporting. I remember my standards were +/- 0.5 at best, example 50 HRc +/- 0.5 would be scribed on the side and the dial had 5 unit increments if I remember correctly, 90, 95, 100... etc. The machine didn't have the resolution to differentiate between 60 and 62 nor the tolerance to do so either. This was on my old Wilson unit. Maybe those fancy digital ones are a whole 'nother ball of wax.

-AJ

 

[SpikeC]

I have a set of Sorby chisels that have the dimple from the hardness tester, but the hardness is not uniform throughout the tool. The cutting edge had to be ground down before the actual hard metal was reached. Whether that was from initial grinding or burnout at the edges during heat treat is the question. The hardness tester only tests one point, is the point here, I think.

 

[jmforge]

Decarb frpm the heat treat most likely.

I have a set of Sorby chisels that have the dimple from the hardness tester, but the hardness is not uniform throughout the tool. The cutting edge had to be ground down before the actual hard metal was reached. Whether that was from initial grinding or burnout at the edges during heat treat is the question. The hardness tester only tests one point, is the point here, I think.

 

[Marko Tsourkan]

...

I still have problems with Rockwell reporting. I remember my standards were +/- 0.5 at best, example 50 HRc +/- 0.5 would be scribed on the side and the dial had 5 unit increments if I remember correctly, 90, 95, 100... etc. The machine didn't have the resolution to differentiate between 60 and 62 nor the tolerance to do so either. This was on my old Wilson unit. Maybe those fancy digital ones are a whole 'nother ball of wax.

-AJ

Wilsons analog or digital (mine is analog about 20 years old) if well calibrated, will get you pretty accurate readings - +/- .25 or less. Helps to have test blocks to check on accuracy periodically and to have replacement indenters, should the one in the tester wear out, as that will throw off your reading. I typically take 5 readings and average.

M