Quench Media

[Jerrod Miller]

I came across some pretty cool info (I love getting new books) and wanted to share some of the highlights with you all. I did a search and couldn't find a single thread with a ton of quenchant specs so I figured it would be cool to start one where we could all just accumulate our various data. I'm thinking cooling rates, flash points, speed comparisons for different manufacturers, etc.

 

Here is the cool stuff I wanted to start with. (I take no responsibility for the units here, they can in C and I wasn't going to re-create them just to get F.)

 

 

 

 

 

 

As you can see, it looks like with water, your temperature is hugely important, but with oil, not so much. Choosing the right oil has a much larger effect than the oil temperature.

 

Personally I'm hoping someone with a cooling curves for canola oil and engine oil (because it is asked about a lot due to the ease of access) can add them here too.

[owen bush]

what size materials are these for as they seem awful slow.

[Jerrod Miller]

Very good question. I was more looking at it from a relative standpoint; e.g. cool water is more than twice as fast as fast oil, and hot water has the wrong profile and generally sucks.

The charts say the source is Houghton, and based on what I find on Houghton's website (see link below, page 8) it looks like their standard test is a 12.5mm (about 0.5") inconel 600 probe.

http://www.houghtonheattreat.com/pdf/tech-tips/Houghton_On_Quenching.pdf

All the charts I posted and many more can be found at that link, though formatted a little differently.

[R.H.Graham]

Cool charts, thnaks for posting them!

 

But, i must say the water quench chart I find a bit off at least in my expirience.

May be because of the steel used in testing of course, but, I often waterquench in hot water using simple steels, in the 80c range, and certainly do not see a hardness down at 45rc, they still seem to me to be in the 65-66 range without any problem,

What I do see is a more pronounced shallow-hardening effect in the steel which I actually prefer, but in any case, I get full hardness at the edges of big stuff no problem, and on the smaller knives that are 1/8th or so at the thickest, they through harden completely and skate the 65rc test files anywhere no trouble.

 

I wonder when looking at heat-treating data, if we should perhaps consider the fact that the data is being provided by a company that sells heat-treating oils commercially. Personally I think the water-quench chart is quite far removed from actuall quench performance.

Not the first time I have seen this.

Edited June 20, 2014 by R.H.Graham

[Jerrod Miller]

Well, the water chart with hardness readings looks to be from a Jominy-End-Quench, which would give pretty different results from a full quench. I've seen several charts showing "how much worse" (slower) hot water is than cool/cold. We pre-heat our quench water for some alloys to take advantage of this slower quench to reduce warping/cracking (much better than trying to maintain a 20,000 gallon tank of polymer for the occasional quench).

 

I wouldn't worry about the source too much. Especially with Houghton, they make such a wide variety of products (and offer design services, including water quench systems) that it is pretty clear (to me at least) that their goal is to understand how everything works and then provide the right solution to each scenario. Also, These charts are found in the ASM Heat Treater's Guide (2nd Ed.) and they are not trying to sell anything (except the book).

[Alan Longmire]

Cool stuff, but I have always suspected the quench speed and depth of hardening with a knife-shaped piece of steel is very far removed from the reading given by a 12.5mm inconel probe or a Jominiy end-quench sample. Of course you know that very well indeed, but it is the big thing that adds that bit of voodoo to the way we do things. The charts do help by providing a (valuable!) baseline glimpse of what to expect for a given quench, i.e. the rate of potential heat transfer from a really good conductor (inconel) and a not-so-good conductor (Jominy end-quench), but that is all they are intended to do.

 

None of the books and charts seem to mention the effects on, say, a piece of 5160 that is 42 inches long by 1.5" wide by .25" at one end, tapering to a sharp point around 1.5mm thick, with double bevels and a fuller ground 2/3 of the way down the center of each side. We can determine that 5160 requires an oil quench due to its hardenability, and we know from experience that a point-down vertical quench with little to no agitation (assuming you can't get perfect laminar flow in the tank) helps with warping issues. All the charts do at that point is tell us that being 5160 it should through-harden nicely. It is fun to play with the numbers and try to figure out how deep you can get full hardness in the same blade made from multiply-normalized W-2 (say grain size = 14, and yes, I have seen it that fine ) and quenched in hot water. 1/16"? That means the edges are fully hard if you left 'em thin enough, and the core once the total thickness is greater than 1/8" starts to be measurable less hard, and that hardness from the edges coming in towards the center will have a nifty fishmouth shape if sectioned and etched...but if you left the edges at 1/8" when you quench, by the time you get 'em ground sharp you'll be left with a thin skin of fully hard steel backed by that softer core.

 

That said, if you ever see me thinking of trying to water-quench a longsword blade made from W-2 in water, please slap me upside the head!

 

edit: Also, please keep the information flowing, I enjoy it immensely despite the apparent tone of that rant in paragraph #2.

 

 

 

Edited June 20, 2014 by Alan Longmire

[Jerrod Miller]

Yeah, I definitely should have prefaced the charts with a "for comparative purposes only" disclaimer. It is nice to know that one quenchant is x times faster than another, but either may be too slow, too fast, or adequate for a blade. I really like the approach of try the slowest you have and get faster only as needed (i.e. if the first quenchant you used doesn't harden it then try something faster). Keeps stresses lower which helps avoid the heartache of quench cracks and warpage.

[R.H.Graham]

Allen, Be braver. W-2 is no harder to deal with than W-1, 1095, or 1086m in a waterquench.

The W stands for water eh

 

:0)

 

You won't get 14 grain in oil.

 

( duckin and grinnin)

[R.H.Graham]

Jerrod...

 

you are succumbing to the dark side. Your only hope is to make a sword....

[Jerrod Miller]

I'm not ready for that particular rabbit hole. I know what is going to happen when I take that plunge (pun intended), and I fear getting lost in it. Some day though...

[R.H.Graham]

Oh c'mon....