Kevin,
Being the frugal miser that I am, OK, poor guy with no cash to speak of, I end up buying Chinese (crap) quite often, surprisingly it does do fairly well, although it irks me to no end that the money I spent could have employed a hard working American. I'm not familiar with a China made Rockwell tester, mind sharing the particulars about where to find one, the cost and the the quality test blocks you use as well? After reading the "Quench Wars", I'm getting a tester, just to be on the safe side, and KNOW-not ASSUME what the hardness is of my steel.
While I applaud anybody’s efforts in pursuing truly accurate and effective means of testing, the Rockwell tester needs to be viewed in the same way as any test in a field where so many have an undying and oversimplified belief in single tell-all tests. There are no single tell-all tests, and the closer you get to general tell all type tests the less accurate and reliable they become. The most accurate tests measure one single aspect very precisely, but cannot provide an overall big picture and only through the combining of many little pictures can you get an accurate view of the whole.
First off, it cannot be repeated enough in order to help folks out with a widespread confusion about hardness testing, that there are different forms of what we call “hardness”; here I will focus on the two we most commonly encounter. The most common that knifemakers look for is “scratch, wear or abrasion hardness”, this is what we check with files (think of the very old Mohs Scale measuring one mineral scratching another). For this discussion I usually use the analogy of imagining a ball of Playdoh loaded with glass shards. If you pass a file over the surface of your doh ball it will skate off the glass and register things as VERY hard.
The next hardness we will consider is “penetrative hardness”, it more accurately describes the materials strength in not being pushed aside as an object is forced into it. For this visualize our same Playdoh ball, but take the handle off the file and shove the tang straight down into it. Of course the tang will simply push the glass aside and sink into the ball as if it were nothing but Playdoh.
Now we have two tests of “hardness” but one is saying the material is as hard as glass while the other is saying it is as soft as Playdoh?
Which one is correct? Well, they both are but only as long as we are well aware of what it is we are measuring, and combining both tests will give us a more complete picture all around.
Bladesmiths have come to rely on a few tests believed to be “tell all” with no real concept of what it is that those tests can measure, leading to way too much confusion. And in the meantime they have turned their backs on tests that can very accurately measure specific things because those tests did not tell them what they
wanted to hear. One of the greatest examples of this is the almost hackneyed “brass rod” test which relies on flexing the edge. Flexing anything made of steel is almost solely a function of the cross section and the only time the heat treatment (hardness or softness) comes into play is when the yield point is exceeded, i.e.- “snap” or bend. Since this “test” can only measure the mode of failure it really tells us next to nothing about scratch or penetrative hardness yet is believed by many to be a reasonable replacement for Rockwell, one of the most accurate measurements of a true hardness.
I only bring all of this up as a reminder that to rely on one test without realizing its limitations is no better than no test at all, and worse if it leads you in the wrong direction.
Files are a good quick way to check scratch hardness as long as you know everything else is in order, but a file cannot detect fine pearlite (the soft stuff made when the quench didn’t quite do its job) any more than it could detect Playdoh between piece of glass, I have very detailed proof of this for any doubters. On the other hand the powerful Rockwell test can also only tell you about one specific thing- the overall strength of the material as measured by penetrative hardness, it will indeed very quickly tell you if you have any fine pearlite as it pushes the hardened bits aside. But the Rockwell test can tell you nothing about the grain size, carbide size, and many other important features inside the steel, in fact in the case of grain growth the Rockwell readings will be even higher!
So, look into a Rockwell tester if you want more precise readings of penetrative hardness (and trust me you do want that), but also realize that that is all it will tell you about and is best used in combination with other tests for the total picture.
The unit I use is a common hardness tester that you can get right now at Production tool Supply for $999.95 (# NR50-900330), although I think I may have gotten mine from Enco. It is not an expensive brand name model but using a tester is like sighting in a rifle, if it will give you a nice tight grouping all you need to do is adjust your windage and elevation, if you know what I mean. The sighting in of these units is all in the test blocks, so even if you have a cheaper unit, if it reads consistently and is adjustable the test blocks is where it is as and the tester can only read as true as the blocks, so one can get better results with a cheap tester but excellent test blocks than a guy with a top of the line tester but cheap test blocks.
Good test blocks will start at around $75 and should be NIST certified to insure accuracy. Look for ones rated in the range you plan to work in the most, so 57-63 HRC would be good.
Always remember that these testers are designed to be very accurate and, like any precision instrument with that level of accuracy, can be easily affected if you are not careful with their use. I keep mine in a heated lab area in my shop because the hydraulics are effected by cold and will skew the readings, also continuous heating and cooling will create internal condensation and corrosion which is bad for any instrument. Use an old piece of clean steel to take several “disposable” readings off from at the start of a testing day to compress all the parts and squeeze the oils on the elevator, after this warm up the tester should be in working order. I use my expensive test blocks to calibrate the tester every so often but make up my own everyday test blocks to double check things (I am anal retentive to the point of neurosis in their creation however).
Test pieces need to be over 200X finish for truly accurate reading (I like 400X). Readings not taken on a flat level surface will not be trustworthy as the micro-crater created by the penetrator must be symmetrical on all sides, one side giving way will result in a false reading. The anvil and the test piece MUST be clean. If I ever hear that almost inaudible “crunch” sound of dust being squeezed under the major or minor loads I discard the reading and do it again after cleaning things off thoroughly.
One single reading is not worth much. I prefer at least 5 readings over an area with the resulting number being an average of these, any readings more than 1.5 HRC deviation from the rest should be redone to confirm.
As you can see I am really a geek and take all of my testing very seriously and insist on verifying all data, especially my own.
I will post answers to your other questions, but right now I am being summoned away from the computer with subtle reminders that it is Mothers day
.
