PDA

View Full Version : Heat treating question



dewat
12-20-2008, 02:18 AM
The only experience I have with heat treating is with O1 and case hardening CRS with Casenit. My problem is with the CRS, for small parts I use a large coffee can lined with refractory and a home made propane burner, I put the parts in a steel open container usually welded up from 1/8" flat, cover the parts with Casenit and heat to orange, let soak for 10 or 15 min. and plunge into cool/cold water, during the summer I add ice,if I don't they come out soft. So my problem is, twice now, with very little torque I've snapped bolts, one was a standard grade of the shelf 1/4-20 the other was a part with a 10-24 stud that I made from a piece cold rolled 3/8. I didn't think one could make cold or hot rolled that hard, the cross section looks just like when O1 shatters when I didn't temper it. Am I getting it too hot or cooking it too long ?

Thanks in advance

Jim

darryl
12-20-2008, 04:26 AM
It's either quenching too quickly, or you need to temper it after hardening. If it was coming out soft after the quench, then you probably went the right way to cool the quench medium, but you need to temper after that- that's what I see indicated anyway.

Ed Tipton
12-20-2008, 08:05 AM
Dewat...My response only addresses a portion of your question. In your question you said you didn't know that you could get CRS or HRS hard enough to shatter as does O1. This to me implies that perhaps you do not fully understand the terms CRS/HRS. These two terms refer only to the process by which the steel left the foundry. For example, it would be possible to have two very different types of steel such as 1018 or 1095, and have either provided as CRS or HRS. The 1018 is not hardenable without using the casenite whereas the 1095 is readily hardenable without any additive. Generally speaking, at least within the "shop parlance" of my circle, the term CRS or HRS is reserved for or implies low carbon steel, but... it does not have to be low carbon steel. Also, the CRS has no scale and is provided in higher degrees of accuracy such as in shaft work where it must have a tight fit as in a bearing. The fact that the bolts are breaking implies to me that they are too hard and need to be tempered. If you were working with "Mystery Metal", it's possible you did not need the casenite at all. As an experiment, you might try heating/quenching the remaining 3/8 rod without using the Casenite and see if it gets hard.

boslab
12-20-2008, 08:12 AM
theres only acouple of more severe quenches, salt water and liq nitrogen [ ok helium/whatever.
if you are case hardening then an air quench should be ok, oil if through hard required
consider sending them to a heat treat shop for salt bath or ion nitriding, if its only small numbers it fairly cheap
regards
mark

Davidhcnc
12-20-2008, 08:46 AM
Jim

'Am I getting it too hot or cooking it too long ? '

I wouldn't have thought so.


yes, it is funny that your CRS hardened in that way. maybe we should consider it as mystery metal, as Ed suggests.

I suggest a solution is to make your part from Silver steel (drill rod) or o1 which has guaranteed ingredients and known hardening /tempering characteristics.

Heat treat instructions available here

http://www.westyorkssteel.com/silver_steel.html


Then the rest is history:)

Evan
12-20-2008, 08:48 AM
Case hardening using sodium ferrocyanide (Casenit) is a process known as Carbo-nitriding. It adds both carbon and nitrogen to the steel. The parts you mention are very small. The method you are using will produce a very deep case, especially of the nitrogen component of the case. Also, when heated to 435C (815F) sodium ferrocyanide decomposes, releasing hydrogen cyanide gas. By case hardening in a container as you do this evolved gas is trapped with the parts.

What is happening is that you are producing a carbon case that is at least .010" to .020" deep, a nitride case somewhat deeper and introducing hydrogen to the metal causing hydrogen embrittlement.

To prevent this on small parts you should change the method by eliminating the heat soak period. It only takes a few seconds to produce a case depth of a few thousandths at red temperature. The total time based on your description is probably something like 1/2 hour or more of diffusion time and that is sufficient to produce some hardening completely through a thin section part.

It is sufficient to heat the part to red heat and roll it in a small mound of the chemical. Then, with the part well coated heat it with the torch for perhaps 30 seconds and then quench. This may be repeated before quenching if a deeper case is desired.

To further strengthen the case hardened part draw it back slightly in temper by heating to 350F in an oven for long enough for the part to be evenly heated to that temperature, then air cool. This will also drive out trapped hydrogen and largely eliminate hydrogen embrittlement caused by the process.

dewat
12-20-2008, 12:29 PM
Thanks for all the information guys. I'm only looking for enough surface hardening to resist wear on these parts so I will experiment with the advice given.

Thanks
Jim

Spin Doctor
12-20-2008, 12:46 PM
Any time you HT something, whether it is case hardened or an alloy* that is directly hardenable (steels with suffucient carbon content) it needs to be tempered. If for no other reason than to relieve the stresses that build up after quenching. Tempering or drawing back does not always imply that you reducing the RC of the material post HT

* The use of the term Alloy Steel has always confused me, at least the way most people use the term. To them 4140 is an example of alloy steel while 1018 is just CRS, O-1, A-2 etal are "Tool Steels". in truth they are all steel alloys that have been developed for specific purposes

Mcgyver
12-20-2008, 03:19 PM
boslab, I don't think an air quench will do, not fast enough for kasenit.

Evan, I'd surprised if you're able to get 20 thou in 10 or 15 minutes, my heat treat guy tells 50 thou is about the limit and that is soaking for hours - where'd you get the 10-20 thou from? would be great if it is so, main problem with kasenit is the case is so shallow.

It is possible to get cracking if the parts have lots of stress, like a rolled thread for example. top stop this, add an extra step - anneal the part prior to heat treating. To anneal, the metal must cool slowly - heat to red then let cool in a metal container full of ashes...this anneals and somewhat normalizes the metal...annealing some parts prior to heat treating can reduce risk of cracking. also true when HT parts that have been forged.

RobbieKnobbie
12-20-2008, 03:22 PM
On the other hand, why are you hardening fasteners? You might consider buying grade 5 or 8 fasteners, or socket head cap screws. They are made to exacting standards for hardness, tensile strength etc etc.

I've used Unbrako's at work for years and find them to be completely reliable, but there are other manufacturers of industrial grade fasteners that work well too.

Evan
12-20-2008, 06:55 PM
Evan, I'd surprised if you're able to get 20 thou in 10 or 15 minutes, my heat treat guy tells 50 thou is about the limit and that is soaking for hours - where'd you get the 10-20 thou from? would be great if it is so, main problem with kasenit is the case is so shallow.


It isn't ten or 15 minutes. How long does it take to bring the container to ORANGE heat? Orange is much too hot as well and increases the diffusion rate considerably. Also, the material is being soaked in the decomposing salt and is getting a much more concentrated dose than if the recommeded procedure is followed.

I am basing that judgement on how thick a case I get when treating a small part three times by the usual method. The ratio of surface area to volume grows exponentially as the size of the parts goes down so the area available to absorb the carbon and nitrogen per unit volume increases per the cube/square ratio of volume to area.

Mcgyver
12-20-2008, 07:54 PM
size of the part doesn't matter to depth, its a surface absorption and is time dependent, I thought the OP said 10 or 15 minutes...at that you're not getting 10 -20 depth I don't think

dewat
12-21-2008, 12:14 AM
I tried Evans method with good results, I got the wear resistance I was looking for and the bolt (the bolt not the threads) failed only after considerable torque was applied with a couple of nuts, what I would normally expect.

Robbie, I modify some bolts for other than intended purposes, just easier than turning the whole thing from round stock, in this case the hex is turned to round and the head is faced off then counter bored a bit.

Again thank for the info and continuing education.

Jim

Evan
12-21-2008, 12:34 AM
size of the part doesn't matter to depth, its a surface absorption and is time dependent, I thought the OP said 10 or 15 minutes...at that you're not getting 10 -20 depth I don't think


It does when it is a small round object. Any point within the volume of a .250 round rod has diffusion occurring from all directions on the surface. A point at a depth of .010" inch is closer to being within that distance of a large part of the surface because of it not being flat. In fact, it can't be further than .240" distant from any part of the suface in a cross sectional plane. And, as I said, it's matter of surface area vs volume. The greater the surface area to volume ratio the more diffusion that occurs per unit volume per unit time. Jim said 10 to 15 minutes at temperature. How long did it take to heat up to orange heat?

Evan
12-21-2008, 12:54 AM
The use of the term Alloy Steel has always confused me, at least the way most people use the term. To them 4140 is an example of alloy steel while 1018 is just CRS, O-1, A-2 etal are "Tool Steels". in truth they are all steel alloys that have been developed for specific purposes


Alloy steel is steel that has other metals as added elements to modify the properties of the iron. Unalloyed steel only has carbon, a nonmetallic element as an added element. It is normally classed as carbon steel. Minor exceptions exist such as leaded carbon steel but that is not the rule. Carbon steels are not alloys as the only metal present is iron. The proper name for a compound of iron and carbon is an "interstitial compound" which consists of a metallic and a non metallic element.

Metal alloys contain two or more metallic elements in proportions significant enough to alter the properties of the mixture.

Spin Doctor
12-21-2008, 01:03 AM
Well the SAE site lists 1018 as having .6 to .9 manganese so IMO it is an alloy

Evan
12-21-2008, 01:18 AM
That amount of manganese isn't sufficient to make a difference to the properties of the iron. It is used in those small amounts to remove sulphur and oxygen from the iron, not as an alloying element. It is also naturally found in iron ore and it is difficult to remove all of it. A iron/manganese alloy may contain around 15% manganese to obtain the properties desired.