forward curve during marquenching

[Ben Potter]

Greetings,

 

I was testing my new HT set up (drum forge, turkey-burner oil tank) and ran into a problem. I was testing it on two blades one was 3/16 x 1 1/8 x 18 inches and the other was 1/8 x 2 x 15 inches both tapered from the spine to edge, and both developed forward curves and wavy edges. I didn't normalize either one( I know bad idea,my wife reminded me but I forgot) does any one know what causes this?

 

Other specs.

 

oil temp. 430 f

quench tank 6ft from forge

 

Thanks

Ben

 

P.S. Thanks for the picks of the drum forge Don, works great and with a really even heat.

[Jeremy Vaught]

Well, first off, I wanna raise my glass to a guy who has a wife that reminds him to normalize his steel. That's just cool!

 

The normalizing (or lack there of) was probably your biggest culprit, but something else that comes to mind is the fact that those are pretty big pieces. Did you get them heated evenly?

 

430*F oil? What kind of steel is it? That sounds aweful high to me, but I'm still green.

[Chris Meyer]

My understanding is that steel expands when it forms Martensite. Makers of Japanese style swords use this fact to get their swords to curve toward the spine when they quench a clay coated blade. Martensite forms on the edge of the sword, but not the back, so the expanding Martensite curves the blade back.

 

In your case, you have the opposite happening. The conclusion that I draw (perhaps incorrectly) is that the spine of your blade formed Martensite, but the edge did not. If that happened, the blade would curve forward and the edge would wrinkle. Why your edge was not hot enough to form Martensite is something you will have to figure out. Are you using 1095 by any chance? If so, you have almost no time to get it into the oil to beat the nose of the Pearlite curve. (In fact, with 1095 at the higher end of the carbon range, the nose on the chart is off the page.)

 

 

[Ben Potter]

Sorry about that I should have said the steel type, it is Admiral 1095. It was heated evenly.

 

The thing I didn't understand was the forward curve And the wavy edge. it is like the spine and edge cooled first and then the middle some how shrunk and curved the spine forward and crinkled the edge.

 

does the oil temp sound right (high? low?) the blade came out more or less plastic save that the thinner one cooled rather quickly.

 

-Ben

[Sam Salvati]

with 450F oil you are most likely not getting past the pearlite nose for 1095, so therefore are only mostly getting the forward curve as it comes down, then the slow formation of a small amount of martensite(?) is not enough to curve it back. What type of oil?

 

IMHO, ditch the 450F oil (waste of time unless you are using a faster medium like water first to get your steel past the pearlite nose, then slow cool from there), get some real quenching oil (houghto quench-k from houghton or park's #50) and use it regularly at no hotter than 120F, it is a fast oil designed for 10XX type steels that need a fast cooling rate to get past the pearlite nose and form martensite.

[Chris Meyer]

Sorry, I didn't catch the oil temp on the first read. Sam's correct, your oil should be much cooler. Heatbath is showing a working temp for their #50 oil of 50 to 120 degrees F. You need a fast oil, like the #50 to have any chance of getting all (or mostly all) Martensite out of 1095. You can get it from Darren Ellis if you are interested. I don't know were you buy the Houghton oil, but here's the company's link.

[Doug Lester]

I'll have to admit that I can't read those isothermal diagrams fer nutt'n. All the 10XX series of steels tend to be shallow hardening due to the fact that it basically only has the the carbon to retard the formation of body centered crystals as it cools from austinite, a face centered crystal. With your oil that hot probably just all the blade cooled slowly enough to make it back to ferrite, pearlite, and cementite. If you have any martensite present it is probably just at the thinnest part of the edge. Another thing that you might look at is your forge temperature and and your soak time. I was quenching 1095 in brine but I got my blade too hot and soaked too long and got the edge coming out of the quenchant looking like a potato chip. If you have the forge too hot, the edge of the blade will overheat while you soaking the spine area to get complete conversion to austinite. Try cutting back on the fuel if you're using a gas forge. If you're using a charcoal or coal forge, put the blade in the fire spine down and cut back on the air. I'd more the quench tank closer to the forge too. Your flare up shouldn't be that bad with the oil temp down to about 120 degrees. I keep mine close enough that I don't even have to turn my body to go from the forge to the quench tank.

 

Doug Lester

[jake cleland]

i don't know a lot about marquenching, but i'd guess it can be done on 1095; i believe the oil temp and time are pretty critical, though. have you checked the spine for hardness? my guess would be that the downward curve is because the spine did not harden (while it's true that differentially hardened blades curve up in water quenching, they go down in oil - no one knows why); either it didn't get fully up to temp, or it didn't get past the nose quick enough due to it's larger thermal mass. generally when i full quench 1095 the spine doesn't harden, but i normalise at least 3 times. the warping on the edge could be from compression from the spine, or most likely just from lack of normalisation.

and 6ft is way too far from your forge for 1095 - it has to get past the nose (900f) in under 2 seconds to form martensite.

Edited September 18, 2008 by jake cleland

[Ben Potter]

I've marquenched 1095 befor (not useing a drum forge, and smaller knives) with good results useing 450F instead of 430f. I snapped one of the blades to see how hard it was and what the grain looked like. it was hard all the way through, edge and spine. I'm guessing it is a normalizing and distance to quench tank problem.

[B. Norris]

The thing I didn't understand was the forward curve And the wavy edge. it is like the spine and edge cooled first and then the middle some how shrunk and curved the spine forward and crinkled the edge.

 

There was quite a long thread in the past about this phenomenon. That of blades curving tip up in water and tip down in oil. The consensus, if I recall correctly, was interesting. Basically, when the blade is quenched, the edge, spine, and interior all cool at different rates. The steel, as it cools passes through an intermediary phase, or phases, which occupy slightly different volumes. For example, water quenched blades end up with martensite at the edge and (I guess) perlite in the body and spine. The martensite occupies slightly more space than the perlite and this causes the blade to bend towards the spine. The oil quenched blades are thought to pass thru intermediate phases at different times than blades quenched in faster mediums. The theory is that during one of these phases the opposite happens, the spine of the blade occupies a larger volume than the edge, the whole thing goes tip down and stays that way. Here is the link to that post - I actually found it for once! The Nichols Report

 

~Bruce~

 

My link, above, did not work for me. If that is the case, you can search the Metallurgy and other enigmas area for a post titled "The Nichols Report," subtitled "Metallurgical analysis of oil/water quench," the post is by DFogg.

Edited September 23, 2008 by B. Norris

[Sam Salvati]

I snapped one of the blades to see how hard it was and what the grain looked like. it was hard all the way through, edge and spine.

 

 

Canyout ell the difference by eye between martensite and pearlite? If so you have a VERY good career in metallurgy ahead of you

[Ben Potter]

The link worked for me, (but I didn't find the pdf) Thanks for posting that, most helpful

[Chris Meyer]

I found the link to that Metallurgical Report.

[jhobson]

Suppose that austenite is easy to deform and that it is the most compact form of iron, but ferrite/cementite and martensite are not easy to deform, then you can explain the direction of curve in terms such that it will curve towards the last austenite to transform and away from the newly formed phase.

 

If you quench quickly to below martensite start then you may expect martensite to form on the edge before the austenite in the thicker section of the blade transforms. Martensite expands and the blade curves up.

 

If you quench quickly to above martensite start and hold then the majority of the blade will be austenite. There is not a lot of phase change, so not much curving (until you do somthing else).

If you quench more slowly to above martensite start, then the thinnest sections will get past the nose, and remain austenite, but the thickest sections will start (and maybe complete) transform to ferrite/cementite.

A slow quench to below Mstart is a bit more complicated and may go either way But it should be repeatable in the same conditions, so should be predictable when you have done enough.

Don suggests that in a water quench you can see the tip drop and then rise, so the simple model above doesn't explain it all.

 

It seems that for shallow hardening steels like 1095, a quench into engine oil, even below Mstart, is slow enough to bend towards the edge.

 

This is what I understood from an explanation by Howard so credit goes to him if it helps, and blame falls on me if it's wrong.

Edited September 30, 2008 by jhobson

[Howard Clark]

Real close. The one thing I would say differently is that in a slower quench like oil or salt, the austenite in the back, (with simple low hardenability steel), changes into pearlite very quickly, usually in the first seven or eight seconds, and it forms at around 900-1000f as the temp drops in the quench. The edge is still metastable austenite for a bit longer, details depending on the quench media, it's temperature, and the steel, and clay.

 

It is just my opinion though. Your results may vary (and likely will due to different steel, equipment, and practices).

[Ben Potter]

Thanks for all the information, it is very interesting

[kb0fhp]

Let me clear up a misunderstanding. The purpose of martempering or marquenching is designed to minimize distortion. It allows the inside temperatures to catch up to the outside temperatures so it transforms to martensite uniformly and at the same time. In this case, the back of the blade and the edge contracted uniformly, but because of the larger mass on the back, the blade bent forward. Both sides grew, but because of the larger mass of the back of the blade it bent because of the shape of the blade. To get a sori you need to quench it to room temperature so the edge cools first and you get differential cooling from the edge to the back.

[Ben Potter]

Let me clear up a misunderstanding. The purpose of martempering or marquenching is designed to minimize distortion. It allows the inside temperatures to catch up to the outside temperatures so it transforms to martensite uniformly and at the same time. In this case, the back of the blade and the edge contracted uniformly, but because of the larger mass on the back, the blade bent forward. Both sides grew, but because of the larger mass of the back of the blade it bent because of the shape of the blade. To get a sori you need to quench it to room temperature so the edge cools first and you get differential cooling from the edge to the back.

 

Is that why when you full quench in oil you get forward curve, but when you edge quench in oil you still get backward curve?

[kb0fhp]

Is that why when you full quench in oil you get forward curve, but when you edge quench in oil you still get backward curve?

 

Well - standard metallurgist answer - it depends. By edge quenching, you get the edge to transform properly. It also depends on the speed of the oil. If it is slow oil, you would have a more uniform quench - and likely would transform the edge to martensite - maybe - and the thicker parts might get transformed to martensite. The steel used may have adequate hardenability to transform completely to martensite. The idea is to have a sharp edge of martensite, and the backing edge of pearlite. If all martensite, the thicker part would expand last, and force it to expand last - causing a forward curve.

 

That is why a low hardenability steel is used - to force the edge to martensite and the back portion to pearlite.

[Mike Krall]

Sorry, I didn't catch the oil temp on the first read. Sam's correct, your oil should be much cooler. Heatbath is showing a working temp for their #50 oil of 50 to 120 degrees F. You need a fast oil, like the #50 to have any chance of getting all (or mostly all) Martensite out of 1095. You can get it from Darren Ellis if you are interested. I don't know were you buy the Houghton oil, but here's the company's link.

 

Look at the last post (#14) for small quantity of Houghton quench oils... http://www.bladesmithsforum.com/index.php?sh...c=11644&hl=

 

Scott McKenszie works for Houghton.

 

Mike

Edited December 9, 2008 by Mike Krall