52100 (Aldo's) alternate heat treat= Epic Fail !!!!!!

OK, this information has been very helpful, but I still have a few questions.
My plan now is to use the below recipe, provided by Samuraistuart and derived from Mr. Cashen:

1. 1650F, cool to black, quench in water.
2. 1500F, cool to black, quench in water
3. 1400F, cool to black, quench in water
4. (if needed) 1250 for annealing and stress relieving.
5. Harden at 1475, soaking for 10-15 minutes, and quenching in medium speed oil that has been warmed to 130F.
6. Optional cryo treatment. I have yet to try any cold treatments. There is dry ice/acetone or liquid nitrogen. Personally, I don't see how 52100 can benefit from cryo, since it's Mf is 235F...but I may just have to throw it in the freezer)
7. To temper....350F for one hour, quench in water, clean off any scale/decarb and test edge. Probably too chippy. Back into oven at 400F for 1.5 hours, cool to room temp.

I'll skip Cryo for now, so my questions boil down to these:

1) How do I know if the stress relieving cycle at 1250 is needed?

2) Can I put all 6 blades in the Kiln to utilize a more production-line-style technique (Such as heating all 6 blades to 1650, removing one, cooling to black, quenching in water, moving on to next blade once internal temp is back at 1650, etc), or should I do one at a time, going through the entire process with each blade?

3) And finally, will stamping the blades have any adverse effect come heat treat? I can etch post heat treat but prefer the aesthetic of a stamp.

And that's it, thanks everyone, the info on here is invaluable.
 
Last edited:
I may be missing it, but there's one question that isn't quite put to bed yet:
Does quenching from such a high austenitizing temp 1650 cause "microfracturing" and if so, is it a problem?
 
I believe Rick mentioned earlier that quenching from 1650 had the potential to cause microfracturing, which is why you heat to 1650, cool to black, then quench.
 
elementfe said:
I may be missing it, but there's one question that isn't quite put to bed yet:
Does quenching from such a high austenitizing temp 1650 cause "microfracturing" and if so, is it a problem?
Click to expand...

I edited my post from earlier on. Kevin noted the concern with microfracturing if quenched from 1650f.
 
adinb114 said:
OK, this information has been very helpful, but I still have a few questions.
My plan now is to use the below recipe, provided by Samuraistuart and derived from Mr. Cashen:

1. 1650F, cool to black, quench in water.
2. 1500F, cool to black, quench in water
3. 1400F, cool to black, quench in water
4. (if needed) 1250 for annealing and stress relieving.
5. Harden at 1475, soaking for 10-15 minutes, and quenching in medium speed oil that has been warmed to 130F.
6. Optional cryo treatment. I have yet to try any cold treatments. There is dry ice/acetone or liquid nitrogen. Personally, I don't see how 52100 can benefit from cryo, since it's Mf is 235F...but I may just have to throw it in the freezer)
7. To temper....350F for one hour, quench in water, clean off any scale/decarb and test edge. Probably too chippy. Back into oven at 400F for 1.5 hours, cool to room temp.

I'll skip Cryo for now, so my questions boil down to these:

1) How do I know if the stress relieving cycle at 1250 is needed?

2) Can I put all 6 blades in the Kiln to utilize a more production-line-style technique (Such as heating all 6 blades to 1650, removing one, cooling to black, quenching in water, moving on to next blade once internal temp is back at 1650, etc), or should I do one at a time, going through the entire process with each blade?

3) And finally, will stamping the blades have any adverse effect come heat treat? I can etch post heat treat but prefer the aesthetic of a stamp.

And that's it, thanks everyone, the info on here is invaluable.
Click to expand...

I can't help you with the stamp. The cryo could help with retained austentite as it is hypereuctoid. From what I have read, it can result in 1 RC point increase in hardness. If well heat treated, the performance difference will be minimal, but there may be a very slight increase in performance.

I did 4 blades at once and the 6 coupons. I closed the door quickly in the last cycle between each piece of steel, and had a half pipe over the blades to minimize heat fluctuations.
 
I would dispense with the water quench after black, unless you do it on the last cycle. If I am not mistaken, the reason for the water quench is to avoid coarse pearlite formation during the slow cool to room temperature. If you are just bringing it back up to your next normalizing heat, as soon as it cools to magnetic(and forms a new structure to re-nucleate from), you can ramp back up for the next cycle. Only on the final cool down cycle do you want to speed things up from 850F or so down to room temp.
 
Any ideas on what plate quenching will do instead of a water quench in between heating cycles?
 
Karburized said:
Any ideas on what plate quenching will do instead of a water quench in between heating cycles?
Click to expand...
Heat up your plates for no reason that I can think of?:p Just messin'. But seriously, I don't see the need to quench in between normalizing cycles. Save yourself the time and energy.
 
HAHAH figured as much. I've been air cooling my blades, and at times i wish i had access to a lab.
 
Adinb114, you need to stress relieve if you've been forging, or doing other things that induce stress into the blade, like bending it extensively, or making it do long form divisions of imaginary numbers. ;-)
 
Ok, if anybody is still monitoring this thread, I'm back with more questions. :)

Using a friends Kiln, I attempted heat treating 7 similar 52100 blades with the Recipe posted here:
1650, cool to black, 1500, cool to black, 1475, cool to black, quench in room temp water. 1475, 10 min soak, quench in canola.

Unfortunately, this failed completely (a file easily bites into the edge of all seven blades), and I'm trying to determine the cause and avoid it for the next attempt.

Some specifics: For the first three thermal cycles, I placed all the blades in the kiln at once, removing them one by one after the kiln had reached temperature. None of them soaked longer then a minute or so at temp for these cycles.

I rushed the final cycle. Once the blades had soaked at temp for ten minutes, I removed them one by one and quenched them in canola. The quench tank only held a quart of canola, and my suspicion is that after the second or third blade, the oil was too damn hot to cool the blades fast enough to reach the target hardness. Furthermore, I think it's also likely that the last few blades may have been a bit below 1475 when they were quenched as the door to the kiln was repeatedly opened and closed and lost heat.

The next step will be to attempt the process again with a gallon of canola. I'll also do the final quenching cycle one blade at a time and see if my results differ.

Anyways, I suppose this is a learn from your mistakes but don't repeat them moment. That being said, if anyone here with more experience has some light to shed, it would be greatly appreciated.
 
I'm pretty positive that your file test "failed" because there is a layer of decarb you must get through before you reach hard steel. I remember being very frustrated with a knife I had made out of CruForge V. I thought I did something wrong with the heat treat, because my file dug into the edge. Turns out, there was a layer of decarb, and underneath the steel was so hard and so difficult to sand by hand. So check those blades again. Yes, one quart of oil MIGHT do it for one small blade, but you definitely need more for seven blades. I would say a few gallons minimum for that many. It is highly recommended to heat your canola to 120F-130F. If you are quenching in water between cycles, just make sure that the blade has gone black, and well below. The only reason I mentioned water quench during thermal cycling is so you can handle the piece easier, in case you sand off scale and what have you. Quenching in water after steel has gone below black heat does nothing to benefit the process......just helps speed things up for some folks. There is nothing wrong at all with going right back into the kiln after it has gone black, or below. Again, my guess is you just have a small layer of decarb, and that's why your file felt like it dug in. Another thing with kilns, keep in mind that a thermocouple might read 1475, but the atmosphere may take a little longer to fully reach temp. Make sure your soaks are thorough, especially with 52100. You really need to make sure this steel has reached it's target temps, and equalized throughout the entire blade. If you followed that "recipe", and it really is 52100, they should come out screaming hard after your decarb is delt with.
 
First off, I want to point out that I forge my 52100 blades from 1-3/4" round stock that I got from Ms Ray Kirk, so my process is geared toward that.

I haven't used Aldo's 52100 yet, so I have no personal experience with it.

One thing that jumps out at me from a lot of these posts, is the time spent in kilns. Unless you're coating the blade with something prior to all the trips in the kiln, you're going to have to do some grinding to get back to good steel.

With my Paragon, if I run a blade through all the thermal cycles and do a long soak at austenitizing temp, a file will bite into the surface of the as-quenched blade.

I had this happen with CruForgeV. It was low 50's on the ricasso and a file bit right in. I surface ground 0.010 off and re-tested the Rc and found it was 65/66.

Just something to think about. :)
 
Outstanding, thanks for the responses Nick and Stuart. I'll do some grinding when I get home and let y'all know the results.
 
Nick and Stu have it right. You get decarb with all that cycling. Some folks grind decarb off in between cycles(I'm lookin at you, Stu:p) I don't agree with that as it just exposes the good metal to further oxidation. I suggest leaving it on to act as a protective layer during the normalization process.
 
I heat treated a large batch of 50 knives in my salt pots using Aldo's 52100. I did all the thermal cycles mentioned above 1650, 1550, 1400, stress relieve, 1550 and mar quenched in low temp salts at 400. They all had a Rc of about 64. I tried it at first at 1475 with some test pieces and didn't even register 20 on my hardness tester. I had to take it up all the way to 1550 before I got decent hardness. I ran close to 15 test coupons before I had it dialed in.

Just because a temperature works well for someone else doesn't mean it will for you and the only way to be sure is to test a bunch of pieces and test them. Every thermocouple reads a different temperature than another. Usually not by much but the do. Also if someone has cut and stripped the wires on the thermocouple by as little as an 1/8" it can throw the temperature read by a 100 degrees. Every connection all the way if not tight will throw your readings off a little. Every kiln, forge, salt pot, etc will be a little different in temp even if they are all reading 1500. Depending on how old, connections, maintenance will determine how different they are. So when someone tells you to take it to say 1500 that is 1500 on their equipment not yours. You just have to keep testing till you get the results you want then stick with that until you change something, then start over.
 
Jimmy, that REALLY surprises me....that 1475 only got you 20. Something wrong there. BTW, the reason I go ahead and quench during thermal cycling is because I sand off the scale during every cycle. NOT necessary, folks. But because my work is done by hand, files and sandpaper, having a layer of decarb to get through really helps me when I have to leave an edge around .030 or so before heat treat. That decarb layer makes it nice and quick to sand through....and then I hit hard steel.....around .015 or there abouts....depending on the steel, temps, and time. So that's why I quench in water between thermal cycling and sand off the scale. To ADD to decarb. NOT recommended for most situations. Just to clarify........RIIIIIIICK!!! he he!!!!!! (I did read one's opinion on sanding decarb off. forgot who, when, and where. he said do it because of possible uneven scale buildup translating to warping during quenching......not sure about all that...but maybe?)
 
Last edited:
Any good controller has the ability to calibrate. All you have to do is calculate the boiling point of water at your elevation and then check and adjust the controller to that if necessary. The manual should have instructions. When people use industrial standards to recommend heat treating they sometimes leave out the standards of the entire process, calibration included, which go into developing a standardized process. Do that and 1500 is 1500. But it all hinges on getting the whole process to standard not just the recipe. Its kind of like needing to use calibrated test blocks on a hardness tester.

To skip the equations, http://www.csgnetwork.com/h2oboilcalc.html .
Just enter your elevation above sea level.

Or an even easier method it to make an ice bath with distilled water. Fill up a cup with ice, preferable crushed and add water. Stir it for ~ 10 minutes and then stick your thermocouple in the bath. It should read 0*C or 32*F. If its not, adjust the controller.

jimmyseymour said:
I heat treated a large batch of 50 knives in my salt pots using Aldo's 52100. I did all the thermal cycles mentioned above 1650, 1550, 1400, stress relieve, 1550 and mar quenched in low temp salts at 400. They all had a Rc of about 64. I tried it at first at 1475 with some test pieces and didn't even register 20 on my hardness tester. I had to take it up all the way to 1550 before I got decent hardness. I ran close to 15 test coupons before I had it dialed in.

Just because a temperature works well for someone else doesn't mean it will for you and the only way to be sure is to test a bunch of pieces and test them. Every thermocouple reads a different temperature than another. Usually not by much but the do. Also if someone has cut and stripped the wires on the thermocouple by as little as an 1/8" it can throw the temperature read by a 100 degrees. Every connection all the way if not tight will throw your readings off a little. Every kiln, forge, salt pot, etc will be a little different in temp even if they are all reading 1500. Depending on how old, connections, maintenance will determine how different they are. So when someone tells you to take it to say 1500 that is 1500 on their equipment not yours. You just have to keep testing till you get the results you want then stick with that until you change something, then start over.
Click to expand...
 
Last edited:
Matthew Lomas said:
Any good controller has the ability to calibrate. All you have to do is calculate the boiling point of water at your elevation and then check and adjust the controller to that if necessary. The manual should have instructions. When people use industrial standards to recommend heat treating they sometimes leave out the standards of the entire process, calibration included, which go into developing a standardized process. Do that and 1500 is 1500. But it all hinges on getting the whole process to standard not just the recipe. Its kind of like needing to use calibrated test blocks on a hardness tester.

To skip the equations, http://www.csgnetwork.com/h2oboilcalc.html .
Just enter your elevation above sea level.

Or an even easier method it to make an ice bath with distilled water. Fill up a cup with ice, preferable crushed and add water. Stir it for ~ 10 minutes and then stick your thermocouple in the bath. It should read 0*C or 32*F. If its not, adjust the controller.
Click to expand...

True, and a necessary step, but you still need a standardized tester to know if you have a bad thermocouple or not. K-type thermocouples are expendables. They go out. They can read as far as 20 degrees difference between thermocouples when only the thermocouple is changed. It was part of my job in the power plant to routinely change them when they go bad. If we got a bad batch and cycled the gas turbine off at night it was not uncommon for me to change out as many as 10 a night, that's when we upgraded to the quick disconnect thermocouples.

1475 did not work for me. Was that the procedure or the thermocouple giving me a reading 20 degrees higher than I actually was? 1455 should not harden 52100 very well. Every thermocouple is a gamble. Unless you have the proper equipment to test the thermocouple before using it and monitor it constantly their is no way to know when it goes bad. The only way to gauge it's accuracy is by testing the results you get. I have literally changed close to a thousand K type thermocouples in power plants. They go bad, they sometimes arrive bad, and each has a lifespan. So specific numbers are great to shoot for and a good guide, but because you have no way of knowing how accurate each thermocouple is, their is no way of knowing when someone tells you that 1475 is the magic number that it is for your set-up. It is for theirs and you should shoot for that number and adjust according to your results. Don't think a small change in number is someone giving you bad information, it's just that all thermocouples will read slightly different.

1475 didn't work for me. 1550 did. I bought my thermocouples off of Ebay. I walked my temperatures up until I got the results I wanted. Was my salt actually at 1550 I will never know unless I set up multiple controllers, a minimum of three, and average them will I ever even be close to accurate. A 75 degree misreading is entirely possible. And I no of no controller that can test the resistance of my thermocouple before I use it, and test it at different temperatures to see how much it is off. We have one at the power plant that is huge and test if they are in range before we use them. Sometimes our thermocouples never made it out of the warehouse. They failed the test before we even used them and sent them back to the manufacture.

Sorry to sound like Debbie downer but thermocouples are not infallible. But with proper diligence are still more accurate than the eye. I keep in stock a stock of ten. I fully expect to go through them all in several years. So I bought the ones with the quick disconnect so it's easy to change them out.

Here were the knives I heat treated.



The bit into bolts and cut them quite well.





I learned not to do this in my hydraulic press as the knife edge was so thick that it would shear the bolt about half way through and bit deeply into my bottom die causing me to replace it.



Still very fine grained.



edge rolled but did not chip at a Rc of 62 even though I lowered the Rc to about 59.



I'm not going to claim to anyone that 1550 is the magic number it's just what I settled on with my setup. I'm going to do another large batch soon and am using the same steel and will do more experiments and will not decide on based on a number but by results. And unless you have the exact same setup with the exact same thermocouple, in the exact same state of decay then you will have a different temp reading than me when you do yours. You can only tell if it was good enough by testing your blades and adjusting accordingly.
 
You must log in or register to reply here.
Back
Top