Practical determination of machinability 4

[Viktor]

Could someone explain me how to determine machinability of a given alloyed steel without cutting tests providing that all mechanical characteristics of this steel as well as its chemical composition and metallurgical structure are known.
It cutting test is needed, what it should be? I mean what to measure and what to compare?
Thank you

 

[jcrowe]

Compare the composition against a know metal like ASTM A36, if you have a sample, try filing it with a commom metal file, if it files easy it will machine easy, etc.

 

[sbc]

viktor,
There are hardness testing files or portable
hardness testers which will tell you the hardness
of a material & therefore it's machinability.

 

[Viktor]

Dear cbs
Unfortunately, I do not know how to correlate hardness with machinability. For example, cast iron could be quite hard but it has excellent machinability. On the other hand stainless 303 is rather soft but when it comes to machine it….Steels 1045 and 303 may have the same hardness – doe it mean that they have the same machinability?
I would appreciate your thought on the matter.
Regards
Viktor

 

[TVP]

I would caution you against any determination of machinability that does not involve actual machining of the steel using appropriate cutting tools, fluids/lubricants, etc. Even steel within a given specification can vary substantially depending on the microstructure; residual elements like Ni, Mo, & Cr; interstitial atoms like N, O, and P; amount and morphology of sulfides; amount and morphology of other inclusions (aluminates, globular oxides, silicates, etc.).

Using only hardness as a determination is not a very good practice, especially if you are planning to machine components for serial production.

When conducting a machining trial, you should keep as many of the variable constant as possible: use the same type of tool you are planning to use for the actual machining, as well as using the same type of fluid/lubricant, machine set up (tool rake, etc.). One parameter to modify would be feed rate for a given depth of cut. The better machining steel will give you a faster feed rate for a constant depth of cut. Surface finish is another variable to characterize. Better machinability will yield a better surface finish at the same feed rate/depth of cut, or will yield the same surface finish at a higher feed rate/depth of cut.

Analyze the chips-- are they small and well-broken? Or are they long and stringy? Smaller and well-broken is obviously better than long and stringy, which tends to damage the tool surfaces, degrading tool life and cutting efficiency.

Other variable to test and record would be cutting forces, power consumption, and or cutting temperature. All of these will vary based on the machinability of the steel. A good reference to consult is Volume 16 of the ASM Handbook series, entitled Machining. A good university library should have this, or you can obtain it from ASM directly at

http://www.asminternational.com.

 

[joedem]

Keep it simple. Contact one of the major metal cutting tool makers i.e. Kennametal, Iscar, Sanvik etc. These companies have all done extensive testing of metal machinability. Most have handy cross charts they will give to you for free. Why try to reinvent the wheel? Or in this case why dry to develop metal cutting speed charts when they already exist. The charts cover most materials. If the materials you are cutting aren't on the charts most of these companies offer detailed papers / reports on the machinability of exotic materials. Call them.

Finally, if you are dealing with a particular type of material not listed in the charts mentioned above, the actual maker of the material should be contacted. Before these companies release a given material to the market, they always conduct machining tests and publish the data. The data is usaully available for free.

 

[tobavan]

I agree that a right way is to use available databases from both cutting tool makers and metal makers, at least for start. However the first usually provide "safe" cutting chart to protect their tools while the later may give "enhanced" data to promote materials. One need to consider that not only machine tools are different from shop to shop (in terms of power, rigidity, acceleration etc.) but your goal also may be varied (saving cutting tools or pushing up to meed a deadline, low or high surface finish and so on). So my point is whenever possible try testing cuts and build up your own cutting database.

 

[Viktor]

Dear Tobavan

Please read attentively what you’ve wrote in you very good message. Machines are different from one shop to another so do coolants, workholding and toolholding fixtures, etc. Then, tool design, too geometry, tool materials – another three independent variables. On the top of that, the cutting regime: surface speed and feed (rate). If I would be naïve enough to start to build such a database, it would take me until Chinese Ester to complete even its small part not to mention system properties and the cost of tools and work materials.

Regarding the databases of cutting tool companies. They are rather poor – nobody can afford to cut tons of different materials to get one experimental point so they do only very few tests in their laboratories. Besides a very few, they have not much knowledge in metal cutting so their data are rather random. As a result, a great number of different tool company could “peacefully” co-exist in the market place.

In my strong opinion, something is deadly wrong with this business. Metal cutting is the subject of many studies and books. What is a single practical result of the metal cutting theory? Why is cannot provide us with a simple criterion (or few simple criteria) to determine machinability of any given material without cutting tests.

I do believe that machinability is the absolute property of a material so it can and should be determined without cutting tests. In simple terms, we have to fracture a part of the work material (in form of chips) from the rest. Note that “fracture” is the correct term because after machining we have separate pieces of the work material. The process of separation of one piece of a material into two or more is called FRACTURE. All sweet dreams about shearing deformation in metal cutting as the cause of chip formation are wrong and stem from complete misunderstanding of this process. Accepting this point, one should be able to calculate the minimum energy needed to fracture the required volume of the work material from the rest. This energy should be considered as the theoretical machinability if a given work material.

One can then judge how good is a particular tool and cutting regime when he compares this result with that obtained in the cutting test. Should be simple and straightforward.

Viktor

http://viktorastakhov.tripod.com

 

[jjjaos]

This is a good question. I don't know of anyone who has come up with a methodolgy to calculate machinability yet. I am sure there have been attempts. I would certainly also like to see one.

One problem is that machineability is a response variable that has not been clearly defined. What exactly is it that you are trying to measure? Traditionally machineablity takes takes four factors into account:
- tool wear
- magnitude of cutting forces
- chip shape
- surface finish

There are a magnitude of factors that affect these values. It is impossible to practically take them all into account. You are right you should be able to calculate magnitude of the forces required to separate a chip from a workpiece. However are you also interested in calculating the other factors? They are not so easy to quantify. You must clearly identify just what it is you want to estimate and under what circumstances you want to estimate it. Because the cutting process will also affect the machineability.

Furthermore how exact of a estimate do you wish to obtain. Certain rules of thumb will tell you if a material is hard or easy to machine.

Generally the harder the material or the higher the tensile strength the more difficult it is to machine. However copper is very soft, but difficult to machine because it is very ductile and chips do not break away. This creates a tendency for it to grap the cutting tool, causing breakages. Conversely, harder materials tend to grind better than softer materials.

Higher carbon and alloy content usually make steel more difficult to machine. Especially those alloys added for hardening characteristics (i.e. chromium, molybdenum, tungsten, etc.) because they increase the material strength and will cause the material to work harden. Nickel and aluminum tend to stick to the cutting tool, causing build up which causes chipping and poor edge retention.

The addition of certain alloys improve machinability. They include sulphur, phosphorous, lead and graphite.

Coolant usuallyimproves machinability, but sometimes actually inhibits it.

Plastics and rubbers machine well if they are not too soft or too brittle. Fiberglass is very abrasive and quickly wear high speed steel and carbide cutting tools. Diamonds work well on aluminum, fiberglass and plastics, but are not tough enough to work well with steel.

I don't personally believe that this is an exact science, but I do feel like you should be able to estimate machineabilty based on a few material characteristics (i.e. material strength, hardness, ductility, thermal conductivity, alloy content, etc.) However I think this would require a major study to determine an equation. If anyone has one, I would like to see it. But any estimate is subject to the real world application, because things are subject to change under different operating conditions. Experience is the main thing.

If you need to select material for a high volume part based on machinability. The best thing to do is to get a sample pieces and try them in the actual machining process you are going to use. Invite cutting tool salesmen to bring test tools and work with you to find the best combination of tools and materials. There is no substitute for trial and error. If you want something to compare it to, use 1212 steel. It has a machineability index of 100% and is used as the basis of comparison for other machineability indexes of other steels.

 

[gbent]

The discussion of the relative machinability of different materials seems to me to be mostly acedemic. In my experience, I have had more trouble with shapes than materials. In spite of the difference in machinability, I would rather make short titanium rounds than long, slender 12L14 shafts.
Another important properity is the ratio of cutting force to Youngs modulus. Steel has rather low cutting forces, but is fairly rigid. Titanium, has only about 1/2 the rigidity of steel, but requires much higher cutting forces. This makes support and clamping much more difficult. It also makes chatter very likely.
I think TVP is completely correct. What will the actual production conditions be? What are the chip-forming tendancys of the material? Some materials are easy to get the part to shape, but getting rid of birdnests of chips is difficult.
Another important item of practical machinability is workpiece condition. It is a forging, casting, bar stock, or ingot?
I try to compare the material to other materials I know, and use cutting tools I know. If the material is new to me, I try to find someone who has experience cutting it. If I can't find anyone anywhere who knows about machining the material, I either try to decline the job, or take it on a time and materials basis.

 

[Viktor]

Dear Gbent
As most practical specialists, you think that machinability is an academic issue rather than a vital problem which any practical engineer faces every and single day. Well… what could I answer? I think that the answer to this question is given in a quote from CIRP working paper “A recent survey by a leading tool manufacturer indicates that in the U.S.A. the correct cutting tool is selected less than 50% of the time. The tool is used at the rated cutting speed only 58% of the time, and only 38% of the tools are used up to their full tool-life capability…” If we recall that the U.S spends more than $120 billion annually to perform its metal removal tasks using conventional machining technology, the price of neglecting the discussed issue as well as the theory of metal cutting is a bit too high. Don’t you think so?

Now about Young’s modulus. Unfortunately, it has a little to do with machinability because it relates to elastic properties of a material. On the contrary, metal cutting involves significant plastic deformation of the work material so strain hardening and strain at fracture should be considered if one tries to correlate the mechanical properties of a work material with its machinability. The product of stress at fracture by strain at fracture gives you a good estimation of energy you have to spend for cutting of this material. If you use this parameter, it would become clear why machinability of Ti is much worse than a medium carbide steel although their “standard mechanical characteristics” (I could not stand this expression) could be the same.

As for materials of various tool and carbide (and other tool materials-PCD, HSS…) companies. Although some of them have extensive testing program, they do not know what they are doing. The logic is very simple – lets try and see. What to try and how to see? Cutting tests are very expensive and time consuming. Moreover, they require very expensive measuring equipment (Kistler dyno’s, data acquisition systems, FFT, laser measuring equipment (vibration and alignment), digital high speed infrared cameras, etc.) and high qualification of test specialists. On the top of that, one should have a suitable experimental methodology and a way to extend the results of a given test on different conditions (similarity theory). Now, you have three attempts to name a tool producers that has all these. As a result, many of them do only very general, superficial test to make sure that “everything is OK in principle” and then use us, users (as Microsoft does) as a laboratory animals to test their product.

If you do not believe, just try to call a tool company (please, select a respectful one to avoid misunderstanding) and ask a contact person to quote you a cutting insert for turning of a very particular work material (hardness, train at fracture, stress at fracture, grain size, chemical composition, inclusions, etc.) on your particular machine (you specify its stiffness, static and dynamic rigidity, range of cutting speeds and feeds, available control unit, etc.), the needed productivity and quality (diametric accuracy, surface integrity and residual stresses including their signs and distribution. You should see his/her reaction (smile, you are on …camera) – they simply do not know more than a half of these technical terms.

Regards
Viktor

http://viktorastakhov.tripod.com

Viktor

http://viktorastakhov.tripod.com

 

[gbent]

Dear Viktor
I won't disagree with your statistics about the underutilization of most cutting tools. I think much of the underutilization comes from making a SWAG on first proof, and then using the theory "If it ain't broke, don't fix it". There are always areas on the first proof which must be fixed, so those operations or tools which performed acceptably well are ignored. After the job makes production, only the bottleneck operations are examined for productivity increases. In this respect, machinability is only a small part of making good parts.
I differ on your thoughts about Young's modulus. I will agree it is not directly related to the chip forming process. I think it is important in a practical sense. The stiffness of the part directly influences the fixturing necessary to hold the part with no motion or deflection. A part with little rigidity which requires a large force to remove a chip will be difficult to achieve a satisfactory level of precision or repeatability.
You have mentioned many of the variables which affect the machinability of a material. The large number of variables which can't be eliminated under production shop floor conditions is what injects a large amount of "black art" into what should be a science. We who practice the art look to people such as yourself to shine a bright light on our art, and help convert it into science.

 

[tobavan]

Viktor,

Like most members here I have no belief that our academics can give an applicable solution (science) for a workshop practice, at least in the near future.

Until this exact science is delived it is probaly more interesting and useful to know how our machinists cut a certain material on his particular machine. With a wider implementation of HSM and HPM much of handbook's data on machining are out of date. So such kind of "community's" database surely is very helpful.

I have visited your site. Very impresive! Can you host that kind of information? And how do other members think of it?

Regards,

 

[Viktor]

Dear Tobavan

I would love to host such a database.
I am trying to survey some modern machining regimes used in the automotive industry. I found, however, that it is not easy task due to huge, unbelievable discrepancies from one manufacturer to another. For example, the penetration rate of gundrill at GM is almost twice slower that that at Ford for the very same work material and drill producer. You would not believe how much time and money the automotive companies spend for re-tooling of their new production lines due to the lack of suitable database on the performance of common cutting tools.
Viktor

http://viktorastakhov.tripod.com

 

[CNCMADNESS]

A good reference to start with is Metcut Research Associates Inc. Machining Data Hand Book. We use this for a good start when machining tough steels like titanium and 17-4 . Usually these speeds are on the low side but again they are just a start.

 

[archers]

I believe one of the data bases would be a good starting point. It provides a history of what has been done and a recommendation for various machining parameters. I also agree that each machining situation is different.

I would recommend starting with the insert, speeds and feeds recommended in a common data base as mentioned above and then create a matrix for experimentation, including such things as rake and relieve angles, speed, feed,type of insert, etc.

Ultimately this is a question of cost per insert or part whichever is more important. Try a few experiments changing parameters until the cost is minimized and product quality or surface finish is acceptable.

This would be a "practical" approach.

 

[metman]

Viktor
I agree heartily with most of what you say as I have worked both sides of the fence metallurgist/strength-of-materials-fracture mechanics and machinist/toolmaking. I have seen the practical guys way underutilize. gbent's comment about Youngs Modulus vs cutting force has great validity. Rigidity in metal machining if not the primary concern is right up there and if not properly addressed nothing else will help.

Have any of you read the book "Chaos?" It is primarily concerned with making realistic approximate mathematical models for nonlinear relationships. With this type of approach possibly a machinability model could be appropriatley derived. Jesus is the WAY