View Full Version : When to test Rockwell??

11-21-2010, 06:54 PM
So I think I have read in different threads that people are getting X amount of hardness out of the quench and X amount after their temper. Does everyone who has a tester test their blades after the quench. I thought it was best practice to get the blade into the tempering oven as soon as possible. I read in Goddard's book to never lay a blade on a cold anvil. That is exactly what you are doing when placing the blade on the platform when testing.

With that being said, I am pretty sure that I have witnessed a couple of hairline cracks in the tang before after testing the blade out of the quench. I don't know how deep they run, but is there a better solution to testing right out of the quench? Anybody have any opinions on this matter or experienced this?

Kevin R. Cashen
11-21-2010, 08:02 PM
A final reading after tempering is fine for telling your customer what the blades hardness is, but it really doesn't help much in refining or troubleshooting your heat treatment. Say your final hardness is 59HRC, that sounds great but is it 59 after tempering what was only 61HRC to begin with? I have had many people just starting out, without a full understanding of what goes on inside the steel, ask what difference does it make if the end hardness is the same, but it makes a huge difference depending on what phases or structures gave you that reading; tempered martensite is the ultimate in strength and abrasion resistance, while tempered pearlite is pretty much a waste of time.

On the topic of time before temper, people tend to jump to extremes on the concept. I have seen some claim that there is no need to temper right away at all and that it will be fine whenever you can get to it, which is patently false if you properly hardened your blade. On the other hand I have heard of people being so paranoid about the temper that they are probably tempering a lot of leftover austenite in a blade that is still too warm for conversion to be complete. A Rockwell test should only take about 2 or 3 minutes to do and if a blade is cracking in that amount of time there probably is not much one could do to prevent it. If a test is take too quickly without proper cooling you will also get false readings from that same unconverted austenite. The anvil of a Rockwell tester bears little resemblance to an anvil for forging on and would only come in contact with a rather small area of the blade.

If visible cracking occurs in such a short time after the quench it would be very likely that the blade already has micro-cracking from the quench itself, and the best safeguard would be to review the severity of the quench. If there were still concerns after this, a couple of measures could be taken to relieve the problem, one would be a short “snap” temper below 350F to stabilize things, the other would be to approximate martempering by interrupting the quench at 400F and allowing the blade to air cool to room temp. Both would require an allowance for the “as-quenched” hardness. i.e. if the steel should be 65HRC it may be a couple points lower due to the snap tempering, how much would depend on your specific temper. With a proper martempering you can expect from .5 to 2 points lower as quenched due to the auto-tempering effect of the air cool, this will not be lost hardness as some people mistakenly fear, but a head start on the actual temper.

I use true martempering processes on all my blades and will even rinse the blades in cold water to insure complete transformation before taking my as-quenched readings, my confidence that there is not so much as a micro crack on fully hardened stock done this way is backed up by extensive microscopic examination.

Doug Lester
11-21-2010, 08:34 PM
Actually most makers don't have the means to test for Rockwell hardness. The instruments for doing it are not cheap and the readings don't really mean much if the tester is not calibrated and accuracy verified with a known standard. Most of us have to rely on performance testing such as edge deflection or wire cutting. Many also do things like testing edge holding by repeated cutting of abrasive fibers, such as hemp or Manila rope, or 2X4 chopping follow by shaving hair.

I agree with Kevin that testing a blade after quenching and before tempering is of questionable value. It's a bit like saying that I was 5'4" before I grew to 5'8". Another problem could be measuring the fire scale instead of the steel. Unless you are trying to chase down a problem with the hardening/tempering proceedure, I don't see what you would gain from the knowlege of what the hardness was out of the quench.

Doug Lester

11-21-2010, 09:24 PM
Thanks for the replies guys. I do have a bench model tester and was asking out of experience. I will now admit, after looking more closely at my blade, that I do not think this one is a crack. I am pretty sure what I was seeing was a very light scribe when I was fitting bolster holes. I did think that I saw one on a previous blade though. Since I received the tester, I have been testing everything that I do, including test pieces out of the quench, blades out of the quench and blades after the temper. My thought is that the more testing that I do, the more I will know what is going on with my HT'ing.

Regarding the response to the size of the tester anvil, mine has the big plate (approx 4-5" diameter). It also came with a smaller one, and maybe I should be using that one. My thought, after recalling Goddard's book, was that it is very similar to laying it onto cold steel and a shock to the blade. My tester is in my basement, so the temp can be in the 50's. I just try and eliminate as many mishaps and mistakes as possible by researching as much as possible. In this hobby/profession though, there are so many different opinions that it is hard to know what to believe.

The steel that I was HT'ing tonight was damascus (1095 & 15N20). Quenching with McMaster 11 sec. I have read enough to know that Parks 50 is probably more desirable, but I was able to easily get the McMaster. Anyway, I have read that sometimes hardness readings on damsascus can be false due to the layers. What is the overall opinion on this.


Kevin R. Cashen
11-21-2010, 10:48 PM
Yes, I would use the smaller anvil, it will increase the accuracy of your readings in the sense that if the steel is not perfectly straight (even less than the eye may see) there could be flexing over the larger area under the major load. Also the tester needs to be kept in a controlled climate to be truly accurate, so I guess if you calibrated it in the 50 degrees range and always kept it within that range it may work. The major load should apply within 4 to 7 seconds and that is controlled by the hydraulic piston in the back. The speed of that piston is heavily effected by the temperature, so if all of you readings are not done in roughly the same temperature range the results could be all over the map.

I have several notebooks full of numbers as I log every Rockwell test done in my shop and I test just about everything I can at every stage of heat treatment, blades of course but also samples of a variety of materials both for myself and many others. The value of Rockwell is another topic that many take extreme positions on, some say that it can't tell you anything that a plethora of homespun tests can't, which is not at all true, while other believe that just having a Rockwell number is all you need to say you make great knives. It is but one test that very accurately measures but one property of steel, no improvised methods currently popular in knifemaking can really replace it for measuring penetrative hardness. But on the other hand it only measures penetrative hardness and can't tell you anything about the internal structure of the steel, and other properties we may wish to measure. I have seen many things under the microscope that are not good but gave rather nice Rockwell readings.