cryogenic tempering appropriate for this application? 1

[JJuhlin]

I have no experience with cryogenic "tempering" but have been offered access to the process and would like to see if it will have any beneficial effects in the following application:

The part in question is a pin 0.55" diameter x 4.5" long. It is part of a roller bearing cage assy in a large (about 40" diameter) thrust bearing with 40 radially arranged tapered rollers. The pins serve as "axles" that pass through axial holes in each roller and hold them in their assigned position within the cage/roller assy.

The pins are made of 4140Q&T material (120ksi yield) that has been induction hardened to yield a surface hardness of about 53-55RC or 8620 case carburized material hardened to about 57-60RC. Both methods have proven acceptable. The rollers spin at about 2000rpm and there is circulating ISO150 gear lube present.

Pin wear and fatigue strength are the primary factors of concern. On paper, the load should be constant, but the bearing is in a rock crusher and everything is subject to high vibratory and impact loading. Fatige failures of pins have been a problem in the past. The current pins have good fatigue life that I attribute at least in part to the ductile core and hardened surface. Through-hardened pins have not performed as well.

Is Cryogenic tempering likely to be an improved substitute for surface hardening treatments in this type of application and will the part be any more brittle than other through-hardened methods?

 

[arunmrao]

There are diverse views ranging from the best to no significant change. Though, there is no harm in attempting them,I know crusher service conditions are quite severe. A matter of curiosity, what crusher machines are you involved with. I am now involved with developing business for cast composites wear parts for HSI and VSI machines.

 

[swall]

JJuhlin--If a subzero treatment is found to be beneficial for the application, you do not need to go cryo (i.e. 300F below zero). Temperatures of -100F are adequate. But, you need to be careful with this, as a certain amount of retained austentite is thought to be beneficial in rolling element bearing components. You need to correlate bearing life with levels of retained austenite in the part before you decide on what the processing will be. Otherwise, you are just guessing.

 

[tomwalz]

We take a long look at cryogenics every couple years. So far we haven't found a reliable process. Sometimes it works very well and other times not at all.

Thomas J. Walz
Carbide Processors, Inc.

http://www.carbideprocessors.com

Good engineering starts with a Grainger Catalog.

 

[tbuelna]

JJuhlin,

I don't have any experience with "cryogenic tempering", but if what you really mean is sub-zero treatment, then this process is beneficial for minimizing retained austenite. But it would not provide any increase in surface hardness.

As swall notes, some retained austenite can be beneficial with case hardened race structures, since it has a strain relief effect. But I don't believe having any retained austenite is desirable in thru hardened race structures, since the austenite will eventually decompose into soft martensite under repeated mechanical loads during bearing operation. If the soft areas are close to the loaded race surface, they can easily fail due to sub-surface shear stress.

Maybe someone with a better background in rolling element bearing metallurgy can elaborate or correct me.

Regards,
Terry

 

[JJuhlin]

The part in question is not part of the load bearing element of the bearing (neither race nor roller). The part in question is part of the cage assy that holds all the rollers in place so that they roll together as a group without bumping into one another or getting out of alignement. Imagine an array of tapered rollers arranged like the blades of a windmill or fan. The races are profiled such that they rollers are constrained radially, but there is space between each roller (not much). The cage is what holds all the rollers in position relative to one another. The pins in question are "spokes" within this wheel-like structure. Each spoke passes thru an axial hole in a roller. Consequently the cage assy keeps all the rolls separated and moves along with the rollers as they travel in a circular path about the center of the "windmill".

The pins are just loose fitting (about .050" diametral clearance) axles running through the axial hole in the spinning roller.

The load on the pin is not particularly great but each roller is spinning about its pin at about 2000rpm. The nature of life inside a rock crusher results in a certain amount heavy vibration and banging around, so the pins must be strong to take this fatigue loading. The 4140 material in its commercially available Q&T state is sufficient for the loads but not hard enough for the wear from the sliding of the roller ID rubbing on the pin.

Information I've found claims that cryogenic tempering is supposed to result in improved "wear" properties. Where this improved wear comes from if the surface is not hard is a mystery to me. Discussions on the topic (see the razor blade on this forum) results in proponents extolling its benefits but then also saying that actual reasons why it works is poorly understood.

Does anyone on this forum know of an application where a cryogenically tempered part is subjected to plain vanilla sliding in the presence of lubricating oil without any actual surface "hardening" (induction, carburizing, etc)?

 

[Frederick]

The US Army Aviation and Missile Command funded a study of Cryo treated 9310 gear materials at the Illinois Institute of Technology Research Institute. They found considerable increases in wear life.

Question: If all cryo does is convert retained austenite to martensite, why does cryo cause G3000 J431 cast iron brake rotors with no austenite and a pearlitic micro-structure to last 4 times as long as untreated ones in both laboratory tests and in practical tests?

For more info on cryogenic processing go to

http://www.cryogenictreatmentdatabase.org/.

 

[JAWright]

It sounds like the problem you have is with the pins, where you need structural steel to the extent that they act as 'axles' but you also need surface wear resistance, like a bearing race or gear flank.

Cryogenic 'tempering' has been studied in a variety of applications, and as far as I can tell, the only identified mechanism that might have benefit is the transformation of austenite to martensite. In particular, there is a distribution of stability amongst the austenite particles in steel, and it will transform the least stable particles of austenite, which is probably rather helpful. That being said, it's also applied to brass musical instruments and cast iron, where these mechanisms don't play a role. Without understanding how something works a lot of testing is needed to demonstrate repeatable performance benefit.

Another solution you might consider is upgrading the steel that you use for pins. Two options that I can recommend are used for aerospace applications like helicopter transmissions and driveshaft/masts and carrier-based aircraft landing gear:

Ferrium C61, a carburizing grade(60-62 Rc) with about 240 ksi tensile strength and 130 ksi-sqrt(in) toughness in the core material, and available from Latrobe:

http://www.latrobesteel.com/assets/documents/datasheets/Ferrium_C61.pdf

Or use M54 which was developed by the Navy for landing gear of carrier based aircraft. Its got minimum 285 ksi strength and 100 ksi-sqrt(in) fracture toughness at 54 Rc, also available through Latrobe:

http://www.latrobesteel.com/assets/documents/datasheets/Ferrium_M54.pdf

good luck!

Jim

 

[Frederick]

JAWright:

Your suggestion of Ferrium steels is a good idea. Your statement, "Without understanding how something works a lot of testing is needed to demonstrate repeatable performance benefit." is not.

Extensive testing has been done on cryogenic processing in laboratories and in actual field tests. We get very consistent results. The idea that we should not use something because we do not understand it is not a valid idea. The Wright brothers did not understand why a wing produced lift, but they created airplanes and eventually the aerospace industry. We did not understand the hardening of steels until the late 1800's but we used hardened steels. We do not fully understand the mechanism of gravity, but most of us use it.

Cryogenic processing is a valid, repeatable process that has huge economic benefits. People are researching why it works, but there is no reason to forgo the economic benefits until someone figures out exactly why it works.

 

[JAWright]

Frederick,

My point was that if you have a developed process for a part and material, then it makes sense to use it whether you understand the hows and whys of its mechanism. But not understanding how it works can limit how fast you can apply it to new applications/materials. For example, experience with cryo-treating brass musical instruments won't tell you what to expect for cast iron brake rotors.

 

[Frederick]

JAWright,

Your clarified point is correct. But the alloys mentioned (8620 and 4140) are commonly cryo treated successfully and the application of true deep cryogenic processing to the part is not something that has to be developed much the same as it is not necessary to develop the carburizing of the 8620 or the quench and temper of the 4140.

Deep Cryogenic Processing of the part will increase the fatigue life and will not result in the part being more brittle. It is a good way to increase the life of the part. We treat a lot of gears made of 8620 and see a large increase in gear life from both the breakage and tooth wear standpoints. You will not see this large increase if the parts are cold treated to -100F.

 

[unclesyd]

Quantifying improvements in wear generated by a sub-cooling c treatment is like jousting with windmills. I've tried it a good number of applications involving tool steels and have only been able to measure any change in only one Oise. This was tested involving very small tear drop shaped broaches 0.0009" dia. We achieved several orders of magnitude improvement in the wear properties of the broaches as measured by the number of holes sized before the broach needed replacement. The tests was across the board, that is all materials used for broaches and broaching all the materials used for spinnerets.
When we tested on several thousand straight cut D2 spur gears used in polymer metering pumps there was no measurable improvement in gear and housing wear. The data was passed through several statistical routines, each with the same result, nothing there.
We had no definitive results when we treated components that were full contact self-mated used in Beamers. Beamers wind the fiber on large beams, large spools.

Have you looked at making the pins from a tool steel like D2. H11, etc?
We have thousands of D2 (Rc 60) shafts running against D2(Rc60) that show very little signs of wear after years of use.

You also might want to look at a work hardening steel like Astralloy V. We use this in sveral applications that require good wear prperties.

http://www.astralloy.com/pdf/astralloy_v_plate.pdf

 

[Frederick]

Unclesyd,

I don't doubt that you have tested a lot of items. But what is a "sub-cooling c treatment"? What does it consist of? If you want to cite your research, give us the details.

 

[unclesyd]

Though I site a lot of references in my posts I'm very reluctant to site anything that I haven't encountered or know a little about.

When I first suggested that we try the use of LN2 there were no references or equipment, only one or two lines in some papers on metallurgy. My first query went to Bethlehem Steel and was told that they were doing some work in this area but it was proprietary.

The basis of my original experimentation was based on work done by the little old Swiss Watch Maker who chilled his steels up in the Alps for two winters to stabilize them. Henry Ford also did the same with CI motor blocks in Detroit.

My initial production work was done when the majority of the people in industry had never heard of LN2 used for anything besides gaseous Nitrogen. This was also prior to people getting fancy boxes with fancy controls so nearly everything we tried was new at the time and others who were experimenting with it keep the information close to the vest. We were the same way.
When LN2 became readily available in bulk we had a storage tank installed next to our lab where it was available at all times. Our first tests were done in an open Dewar in a fume hood on parts generated by the lab. While a year we had a separation plant on site so we had plenty of LN2 to work with.

Our first production test was on large D2 gears to stabilize the growth material without having to do so many tempering cycles. At the time it to 9 tempering cycles to get the growth to a measurable "0". Our first production run was one high temperature(-100F) tempering cycle, cool to LN2 hold for 4 hrs, warm up in cold box, another high temperature. We were able to the go get the growth rate to an acceptable level with this treatment. My next test was trying to stabilize the spindle and drum on some very high speed air bearings post heat treatment. We were able to stabilize about 75% with LN2. We jsut reground the other , there was no investigation as to why. At last we got a little information about some work done in Russia on improving wear of tool steels. At this time we wee repeated the LN2 treatment of the D2 material and as noted the results were inconclusive. We tracked some new gears and center plates that were given the LN2 treatment and again tee results were inconclusive.

Our first effort with a different materials came when we were having trouble with the broaches, holding dimensions and wearing. The problem was resolved for while by having the operator put a monomolecular layer of Cu on each broach prior to the first use. This would not be acceptable in the long term so we put in an order for broach blanks 1mm dia x 35 mm long made from three different materials, D2, S7, and O?. There were 20,000 of each material which after being properly heat treated the heat treaties shipping department commingled the whole lot.
We sparked tested and got about 200 of each material and gave them a LN2 treatment. At this time we cooled to LN2 temperature for 8 hours and quenched up. This treatment has subsequently been changed. The wear were tests were a resounding sucess as the wear rate decreased on all test broaches of all materials. After this we treated the whole lot in an open Dewar and made what we call plugs out of the blanks. As we learned our lesson we ordered the same materials one lot at a time.

There are several process being touted to do different things to and for metals that I take with a grain of salt. I an still interested in reading about results from a well designed experiment that is impartially evaluated. I'm nothing saying some people are not telling the truth, they are stretching it up to it's elastic limit. You can consult with a statistician and tell him the result you need and he and his computer will get them from your data no matter what the right answer is.

 

[Frederick]

unclesyd:

Thank you for this information. It kind of clears up in my mind why you have a negative view of deep cryogenic processing. You are right, the Swiss often would cold treat metals in the snow. Not only Ford, but Pierce-Arrow would cold treat engines block by setting them outside over a winter or two. Race car engine builders did the same. This is a cold treatment and not a cryogenic process. If the early Russian papers you cite are the same ones I read, they were dipping the parts in liquid nitrogen. They had about a 10% success rate. You would not believe the number of people who tell me that they dip parts in LN2, test them and therefore prove that cryogenic processing is useless. A proper cryogenic process involves a slow cool down to cryogenic temperatures, a hold at those temperatures, and a slow warm up and depending on what is being treated, by a tempering cycle. The Cryogenic Society of America has defined cryogenic temperatures as those below -244F. Thus, normal cold treating down to -120F is not a cryogenic process.

Dipping a part in liquid nitrogen creates extreme temperature gradients which cause stress and possible cracking at the surface. Imagine dropping a cannon ball into liquid nitrogen. The outside wants to be at -320F and attempts to shrink to the size it would be at that temperature. The inside is room temperature and wants to be the size it would be at room temperature. The outside is stressed and brittle and will crack. The slow cool down will prevent this by keeping the temperature gradient small. Your success with the broaches was probably because they were small and had little mass. Our machine designs use a heat exchanger technology to prevent high temperature gradients. When tested against other machines, it was found that our process out performed others where the parts were cooled by spraying LN2 into the chamber.

If I am correct and you tried dipping the parts, I can see why you had variable results. It would be the same if I claimed that heat treating did not work because I heated a piece of D2 with a torch until it was cherry red and tossed it into a bucket of water and it did not give the results I expected. I am not criticizing you because you were working with the information available at the time. My point is that there have been changes in methods and knowledge and the process has advanced. My first experience with cryogenic processing was very similar to yours. A person described as an expert in the field dumped a test die into LN2. The results were uninspiring. There is plenty of research out there by credible organizations that have no financial stake in the process that proves that the process works. The Cryogenic Society of America is working to gather even more.

Please remember this process is in its infancy. We have only had about 150 years where extreme cold has been available, whereas humans have been modifying metals with heat for 75 centuries. It is no mystery why we know so little about why cold affects metals and much more about why heat does.

It is very encouraging that research is now turning away from proving that the process works and to why it works. For instance, recent papers have tested different temperature profiles to determine the proper profile for specific metals. Also, Air Liquide, a company that provides liquid nitrogen in 75 countries and has over 40,000 employees is working to find out why the results are so good. They are convinced it works, now they want to find out why.

Take some time and read the research. It is getting very interesting and promises to be more interesting in the future as we find out exactly what is going on.

 

[unclesyd]

To respond to the OP for such an application their is no proof that you will gain anything, in the bane you cannot hurt anything. The only added expense would be the processing. If you have a processors you might want to inquire if you can piggyback your parts. I so this all time with small lots of parts to be processed. The only thing is doing it this way might add a delay to your time line. Again I would look at making the pins from a wear resistant tool steel and at the same time look at the lubrication system if you are getting wear on an unloaded part.


Federick,

We could go at this ad infinitum with no clear resolution as to merits of a Cryogenic treatment. I haven't read any of the latest papers as I don't have access to a Technical library or receive specific industry magazines any more, but I will keep my eye out for published articles. The study by Air Liquida is relatively new according to my information they had nothing going on in 2002. Another thing is I don't have any information on the Cryogenic Society of America or in other words is this Trade Association or a Technical Association. Case in point, the only Cryogenic Society I'm familiar with wanted to store my dead body in LN2 in hopes that technology would catchup and figure how to thaw me out and cure my Ill's. This not a rejection of your society it only that I have no idea who they are.
As for your temperature and the approach the question to me is who picked this temperature and why. It would be easier and much cheaper to drop the temperates to the LN@ boiling point and do away with away with achieving a set temperature and controlling it. I've seen nothing that would indicate that a ramping cooling rate is critical. The cooling rate of the LN2 temperature is easily managed. On large parts that we tested we did slow cool them to prevent gas blanketing by the boiling LN2.

To be able to state something doesn't wear as bad after treatment is hard to do as most test are subjective and the part itself like the brake rotors is used under such varying conditions I would be hard pressed to pin down any real change. With my broaches we used such a quantity and every aspect after treatment was under our control and the use only of a few people the variability was quite low. I'll give you this that in this particular case the improvement was verifiability real.

You are right in stating this is a technology looking for a place in the field of metallurgy. Just remember that the ubiquitous kitchen toaster sat on the shelf for about 40 years until someone came up with pre-sliced bread.

 

[Frederick]

unclesyd:

It really doesn't matter to me that you personally don't think cryogenic processing is real. I am not trying to convince you. What matters to me is that the proper information about deep cryogenic processing is disseminated. This is a forum for discussing things. If I believe that your statements about the process that I have studied for years are misleading or maybe misguided, we can discuss it here. If I allow such statements to stand, others who read them will take them as gospel. It is clearly evident that what you experimented with is not what is done in the current state of the art. That being said, I applaud your willingness to experiment. We need more of this in our world today. By the way, my first exposure to the process was by an "expert" who treated the parts just as you did. The results were not encouraging.

Cryogenic Society of America

http://www.cryogenicsociety.org/

CSA is a non-profit technical society serving all those interested in any phase of cryogenics, the art and science of achieving extremely low temperatures — almost absolute zero. Cryogenic processing is only a small part of their subject matter.

The proper phrase for the field of freezing of dead bodies is cryonics. CSA has nothing to do with cryonics.

cryogenictreatmentdatabase.org is a data base where anyone can look up information and research papers on cryogenic treatments of materials, including unclesyd.

Another trade organization that has a committee on cryogenic processing is ASM International. This is a very prominent organization of materials scientists. ASM International serves materials professionals, nontechnical personnel, and managers worldwide by providing high-quality materials information, education and training, networking opportunities, and professional development resources in cost-effective and user-friendly formats. ASM is where materials users, producers, and manufacturers converge to do business.

http://www.asminternational.org

The ramping of temperature slowly up and down is traceable as least as to the 1960's work of Dr. Randall Barron. Dr. Barron's work is in the database mentioned above. It is almost universally accepted as part of any real cryogenic treatment. The theory is that it prevents high thermal gradients and it allows time for some of the changes in the crystal structure to happen. For instance, going down in temperature too quickly will allow excess vacancies in the crystal structure to "freeze in." Sudden changes is temperature will also inhibit the formation of very fine carbides in tool steels.


We, and others have done extensive testing of parts. You can see examples in the data base above. In regards to brake rotors we have done extensive laboratory tests to SAE standards. The US Postal Service had extensive tests done to SAE standards. We have also tested on actual vehicles to prove the lab tests were relevant. Let's face it. If you test a hundred parts under different conditions and they all wear less, something is happening. The work that Air Liquide is doing now was inspired by their examining the research that has been done in the past and has shown beyond all scientific doubt that the proper application of deep cryogenic processing yields huge increases in wear resistance.

 

[swall]

In the metal casting course I took as part of my metallurgical engineering degree, we were told that leaving the engine block castings outside for a period of time sufficient to develop a rust layer improved machinability because the rusting process caused any sand imbedded in the metal to spall off. Thus, cutting tools lasted longer because they were not cutting an abrasive surface.

 

[DonkeyPhysics]

Hi Frederick,

I've really appreciated your replies on this thread. I understand cryogenics is still a relatively new field. Oddly enough, I have a similar question in another post I just started (right before seeing this thread). I'm less interested in cryogenic work though, and more interested the lesser "subzero" treatment for relief of internal stresses (i.e. of 6061 aluminum).

Rather than have me hijack this thread though (since my question is sufficiently different), I wonder if you wouldn't mind popping over to my other thread and seeing if you may be able to offer any insight there?

http://www.eng-tips.com/viewthread.cfm?qid=299491&page=1

Thanks a lot!

-Alex-

 

[cloa]

Please consider posting in the cryogenic engineering forum although a few poster have been directed away from it such as Boiler/Pressure Vessel forum.

http://www.eng-tips.com/threadminder.cfm?pid=774