Consistent Units for Thermo-Mech Analysis 2

[Eser]

Hi all!

I have a question regarding the selection of suitable units for the calculation of a thermal-mechanical analysis. Inorder to have the stress in Mpa what should be the unit of:

-Specific Heat
-Conductivity
-Mass Density

If the units of force is N and the distances are in mm.

Thanks in advance,

Eser

 

[Denial]

If you want a rigorous set of consistent units, and you have already specified that you want force in N and length in mm, then (unless you are dealing with a problem that involves electricity) there are only two more units to be specified. These are the unit for time, for which I assume you will accept seconds (s), and for temperature, for which I assume you will accept centigrade degrees (degC).

Everything else then follows automatically.
Mass. Must be in kg (to be "consistent" with N for force).
Stress. Will be in N/mm^2 (ie MPa).
Acceleration. Will be in mm/s^2 (and this will apply to "g" as well).
Density. Will be in kg/mm^3.
Energy. Newtons times millimetres, ie millijoules (mJ).
Thermal conductivity. mJ/mm/s/degC.
Specific heat. mJ/kg/degC.
Etc.

 

[jhardy1]

Sorry, Denial, but when working in N mm units, the consistent unit of mass is the tonne, NOT the kg. The following is a fairly comprehensive list of consistent units in the N mm system:

Length: mm
Mass: tonne (T)
Density: T/mm3
Time: second
Energy / Work: Joule
Power: Watt (=J/s)
Force: Newton
Modulus / Stress: N/mm2 (=MPa)
Temperature: Kelvin (for temperature differentials, 1 K = 1 degree C)
Conductivity: W/mm/K (= J/s/mm/K)
Specific Heat: J/T/K
Flux: W/mm2 (=J/s/mm2)
Convection: W/mm2/K (= J/s/mm2/K)

 

[Denial]

JulianHardy is correct, and I was wrong.
(Now I know why my bridges fall UP.)

 

[ThomasH]

I'm not sure that I agree on this one.

Energy/Work: Joule = N*m

That is NOT consistent with N and mm. I have not studied the details but the proposed is not consistent as far as I can see.

Please correct me if I'm wrong.

Regards

Thomas

 

[jcac]

I agree with Thomas,

Length unit: mm
Mass unit: tonne
Time unit: s

1 force unit = tonne x mm / s^2 = 1N
1 work unit = N x mm = 1E-3 J
1 stress unit = N / mm^2 = 1E6 Pa
1 density unit = tonne/mm^3 = 1E-12 kg/m^3
etc.

Now I have to hope I'm corrrect.

Eser, I assume that this is a set of units that you are not familiar with. Do you have to use this set of units? My experience of using a set of units that I am not familiar with is that it is easier to make a mistake and harder to spot it (see above).

Regards

James

 

[EricZhao]

I suggestion is unless you are comfortable and confident with a unit system, don't use it. I would just take the time to transfer everything in to kg-m-s unit system. As we can see from the previous post, it is tricky to make a consistent unit system by yourself.

 

[ThomasH]

This is by no means a new discussion. Some prefer SI units others the so called costumary or imperial units.

There is, in my opinion, no "best" unit system. Use whatever system you are comfortable with. But so far I have never seen anybody who got it correct on the first try when creating a "new" system. There are a lot of misstakes to be made, unfortunately.

Good Luck

Thomas

 

[jhardy1]

ThomasH,

Oops! You are quite correct - in the N/mm system, the consistent unit of energy / work is the N.mm or the mJ (millijoule). the correct list is as follows - I hope!

Length: mm
Mass: tonne (T)
Density: T/mm3
Time: second
Energy / Work: millijoule (mJ)
Power: milliwatt (= mW = mJ/s)
Force: Newton
Modulus / Stress: N/mm2 (=MPa)
Temperature: Kelvin (for temperature differentials, 1 K = 1 degree C)
Conductivity: mW/mm/K (= mJ/s/mm/K)
Specific Heat: mJ/T/K
Flux: mW/mm2 (= mJ/s/mm2)
Convection: mW/mm2/K (= mJ/s/mm2/K)

It is for precisely this reason that i customarily use SI units whenever possible. Whenever you need a material parameter or similar, you can be sure that the data should be expressed in basic SI units. Getting such material properties in derived units can be problematical, to say the least!

 

[rb1957]

guys,

are we being careless with mass and force ...

the last time i checked a tonne was a force,
and THE unit for mass was the gram (i'll accept kg)
1N = 1g*1m/sec^2

we, over here on this side of the pond, still have the fun of using slugs as the imperial unit of mass

 

[TGS4]

rb1957,

Wrong - tonne is another way to write the Mg (1e6g or 1e3kg). Also, the Newton is defined as 1N=1[(kg*m)/(s^2)].

Also, it doesn't matter what you use as the BASE units:
length
time
mass
energy
temperature

so long as the derived units are CONSISTENT. Isn't that the definition of a consistent set of units. For all I care, one could use km as length, Mg as mass, year as time and GJ for energy, and K for temperature so long as the other derived units are consistent with this set of BASE units.

Perhaps a refresher on units is needed for some... For example, force is [M]*[L]/([t]^2), and pressure is [M]/([L]*[t]^2).

 

[johnhors]

Am I alone in finding this thread rather worrying ?

That so few people seem to fully understand the meaning and use of units, especially the correct definition of the SI units. TGS4 is right in saying that "it doesn't matter what you use as the BASE units, so long as the derived units are CONSISTENT" and that your work is clearly marked with your choice of units, but straying from SI units to a derived set of metric units appears to create many problems for many people, whilst users of the Imperial system of measurement IMHO appear not to have the same problems. Is this because an "inch" is more natural and easier to work with than a "metre" ? In the mechanical world, how many 3D models and 2D drawings use the metre as the unit of measurement ? Every drawing or model I see uses the "millimetre" in the metric world or the "inch" in the Imperial world. Is it then any wonder that things go wrong in the metric world if the product definitions do not use SI units ?

 

[ThomasH]

Like I said before, I don't think there is any "best" unit system. Use whatever you are comfortable with. Problems usually occur otherwise. If you don't have any system you feel confident with you have another problem.

In my experience problems can occur when someone is comfortable with "his own" unit system and then work outside his/her normal scope. Stress engineer working with dynamics for example.

Newton and millimeters gives MPa but is usually the best choice for dynamics.

As for THE unit for mass it is kg, it's a flaw in the SI-system that means that the base unit has a prefix. As for tonne it's mass but tonne force is force.

For more on SI-system I recommend this:

http://physics.nist.gov/cuu/Units/units.html

and if you want to play with units:

http://www.digitaldutch.com/unitconverter/

Regards

Thomas

 

[TGS4]

johnhors - I too found the direction of this thread very worrying, which is why I wrote in. In my experience, I have chosen, as ThomasH suggested, to work with a set of units thta I am familiar with - and for me that is the base SI units. Sure, I get stresses such as 0.113e9Pa, but that's OK, because I learned a little trick in school called "I can use exponents". After about 1.5ms, I can tell that 0.113Pa is equal to 113MPa. And, since I know that 24" NPS pipe is 0.6096m in diameter. If anybody wanted to look at my analyses to try to figure out what that diameter is in mm, then I would tell them to use ~1/1000000 of their engineering ability and do the conversion themselves.

Rant over.