[mem_shz]

Hi Everybodi..My project is a midwall cooling in gun. the innerface of the tube reach the maximum temp 2000 K in 10 milisecond and between 10 milisecond to 60 milisecond reach 300 K (as continously shooting) I have defined this function in Cel(Like step function),this function is a tarnsient temprature (Time versus temp) to give it as input i have 2 question..
1) how can i define this cycle repeatedly in total time 7 second?
(i mean after 60 milisecond reach 2000 K again in 10 milisecond and reach 300 K again in 50 milisecond)
2)where should i give it to innerface b.c ..in fixed temprature or somewhere else?
Thanks.

[mem_shz]

No Idea ?? Please Help Me If Anybody Can...

[ghorrocks]

If the firing cycle repeats continuously you can use the mod function to give the time in the current firing cycle.

Is you second question what boundary condition to apply to the inner wall? The heat transfer from the gasses to the barrel is going to be complex as it covers a massive range of conditions (ambient air, then a high pressure/velocity/temperature pulse which quickly fades). I suspect you might have to do a CHT simulation to model this, but I doubt the physical models in CFX are designed to cover such a massive range of conditions.

You could just give it a go a see how close you get.

[mem_shz]

Originally Posted by ghorrocks

If the firing cycle repeats continuously you can use the mod function to give the time in the current firing cycle.

Is you second question what boundary condition to apply to the inner wall? The heat transfer from the gasses to the barrel is going to be complex as it covers a massive range of conditions (ambient air, then a high pressure/velocity/temperature pulse which quickly fades). I suspect you might have to do a CHT simulation to model this, but I doubt the physical models in CFX are designed to cover such a massive range of conditions.

You could just give it a go a see how close you get.

Yeah i use the mod function for 60 milisecond of my simulation (see the picture)
http://picpaste.com/Capture8-tQSzZb0s.PNG
and even run in this timesteps but my main question is how to repaet this cycle(maybe in Cel) for 7 second for my total simulation (total time 7 second).

this transient temprature is an input for innerface of barrel in my simulation and i want to contour a transient thermal response of outerface in result.
Thank You Very Much

[ghorrocks]

Here is some CEL which might do it:

CycleTime = mod(t,0.006[s])
TInput = if(CycleTime<0.001[s],1000[C],30[C])

[Lance]

Originally Posted by ghorrocks

Here is some CEL which might do it:

CycleTime = mod(t,0.006[s])
TInput = if(CycleTime<0.001[s],1000[C],30[C])

Dont use mod with decimal numbers: http://www.cfd-online.com/Forums/cfx...ction-mod.html
In v16 and earlier, mod may give a totally wrong result and in v17 it can result in numerical noise.
Example:
mod(0.3,0.05) should = 0, but
v16: mod(0.3,0.05) = 0.05
v17: mod(0.3,0.05) = 7.45058e-09 <--- still not zero,

[ghorrocks]

Thanks Lance, I could recall an issue like that but I could not find it. Thanks for pointing it out.

You might be able to use the time step count instead of the physical time in the mod function.

[mem_shz]

Originally Posted by ghorrocks

Here is some CEL which might do it:

CycleTime = mod(t,0.006[s])
TInput = if(CycleTime<0.001[s],1000[C],30[C])

Oh thank you very much my friend a little mistake in mod function i have made

[mem_shz]

Originally Posted by Lance

Dont use mod with decimal numbers: http://www.cfd-online.com/Forums/cfx...ction-mod.html
In v16 and earlier, mod may give a totally wrong result and in v17 it can result in numerical noise.
Example:
mod(0.3,0.05) should = 0, but
v16: mod(0.3,0.05) = 0.05
v17: mod(0.3,0.05) = 7.45058e-09 <--- still not zero,

I have V17 but this expression was exactly what i want when i plot this and works for me !!
see this
http://picpaste.com/Capture9-zQugtLTm.PNG
where is the problem ??! i doubt it s correct or not...can you explain more
thanks

[ghorrocks]

If it works for you then happy days. But proceed with caution.

[kabuforever]

Originally Posted by ghorrocks

If the firing cycle repeats continuously you can use the mod function to give the time in the current firing cycle.

Is you second question what boundary condition to apply to the inner wall? The heat transfer from the gasses to the barrel is going to be complex as it covers a massive range of conditions (ambient air, then a high pressure/velocity/temperature pulse which quickly fades). I suspect you might have to do a CHT simulation to model this, but I doubt the physical models in CFX are designed to cover such a massive range of conditions.

You could just give it a go a see how close you get.

In such a small time scale, i think we must consider about the non-Fourier heat conduction effect. and i don't know if CFX covers this kind of problem.

[ghorrocks]

I thought non-Fourier effects became significant on the nanosecond time scale, temperatures just above 0K or extreme heating rates. This case does not seem to qualify as its time scale is milliseconds. So the Fourier heat model is applicable.

[kabuforever]

I think the standard for non-Fourier effect should be the changing rate of temperature (in this case more than 1e4K/s) or dq/dt, not only the time scale.

[ghorrocks]

I don't know what the thresholds are for heating rate so cannot comment. You might be able to develop a model for non-Fourier heating in CFX, I have not looked into it. But I would certainly start with a normal Fourier heating model and get that working well before I considered it.

[kabuforever]

I haven't dealt with this kind of problem, but got this information from some books, non-Fourier effect might exist when the changing rate of temperature reaches 10e3 to 10e7 K/s.