How do you design a flame holder for a jet engine?

[Serj]

I don't quite understand how it is supposed to work. How do you determine what the flame holder is supposed to look like ,where to put the holes, and how big the holes should be? What are the physics and equations for it?

 

[Clausius2]

I don't quite understand how it is supposed to work. How do you determine what the flame holder is supposed to look like ,where to put the holes, and how big the holes should be? What are the physics and equations for it?

Do you mean the fuel injectors?

If so, the physics of the stuff is a bit complex, believe my words .
Problems on flame stability in unsteady running, or in supersonic flows (Scramjet) are severe difficulties one has to solve at the time of design. Also power is very influenciated of how is the burning process and how the combustion takes advantage of the chemical energy.

Be a bit more accurate in your question, I think it is a bit general. There are million of books and pages which try to answer your simple question.

 

[Serj]

Sorry,I meant the combustion chamber. Or more specifically the flameholder (also called flame can) . The diffuser slows the air down so why does it need a flameholder?

 

[FredGarvin]

Flameholders are used in afterburners. Combustors (that I know of) do not have an equivilent.

As for your questions regarding the physics, there are a few things happening in a combustor that makes design extremely complex: There are multiple flows in a combustor. You also have to balance the need for complete combustion and the reduction of temps so you don't fry your HP turbine. There is also the need for proper mixing. There's no quick and easy physics lesson to describe what is happening.

 

[Serj]

Do you have any links containing equations? Thank you.

 

[brewnog]

As Fred and Clausius have (reliably) told you, it's not just a case of putting your numbers into a formula to find the right answer. But, if you insist... They're not equations, but some information about tools used to numerically solve the type of problem you're badgering about.

http://www.navo.hpc.mil/Navigator/sp04_Feature3.html
http://www.osc.edu/research/pcrm/publications/NumSimTwoPhs/index.shtml
http://www-acerc.byu.edu/papers/97-cosmo-gt.html

 

[Serj]

sorry, i didnt know i was badgering. I was just curious. Thank you for the links.

 

[brewnog]

Sorry, you weren't, I'm just grumpy today! But as you can see there's a lot more to these kinds of analyses than just plugging numbers into an equation.

 

[Averagesupernova]

http://www.nickhaddock.co.uk/turbinepage.htm
http://www.reality.demon.co.uk/gasturb.htm
http://www.wclintdavis.com/
http://www.gas-turbines.com/
http://blastwavejet.com/pulsejet.htm

Here are some people that have made their own gas turbines.

 

[Clausius2]

But, if you insist... They're not equations...

No equations?

There are several people who spend and have spent the rest of their lives trying to solve THE equations. Today, they remain as a mystery of science. Combustion equations are the generalization of Navier Stokes equations with reactive flow. Besides such equations are strongly nonlinear, the reactive terms usually attaches an Arrhenius factor which make N-S equations to be more non-linear.

A simple example of how a laminar diffusion flame works mathematically, can be found here:

$$\frac{\partial \rho u}{\partial x} +\frac{\partial \rho v}{\partial y}=0$$

$$\frac{\partial}{\partial x} (\rho u^2+E_{u}P}) +\frac{\partial}{\partial y}\Big( \rho uv-\rho \frac{\partial u}{\partial y}\Big)=0$$

$$\frac{\partial \rho uY_{O}}{\partial x} +\frac{\partial}{\partial y}\Big( (\rho vY_{O}-\frac{\rho \partial Y_{O}}{PrLe\partial y}\Big)=\frac{DaS}{1+S} Y_{O} Y_{F} exp(-\frac{\beta}{T})$$

$$\frac{\partial \rho uY_{F}}{\partial x} +\frac{\partial}{\partial y}\Big( \rho vY_{F}-\frac{\rho \partial Y_{F}}{PrLe \partial y}\Big)=\frac{Da}{1+S} Y_{O} Y_{F} exp(-\frac{\beta}{T})$$

$$\frac{\partial \rho uT}{\partial x} +\frac{\partial}{\partial y} \Big(\rho vT-\frac{\rho \partial T}{Pr \partial y}}\Big)=Da T'Y_{O} Y_{F} exp(-\frac{\beta}{T})$$

That's only a model (which has been simplified using boundary layer approximation and steady flow) of a diffusion flame of Oxygen and Fuel. As you can see, the equations are coupled and the fluid field is very complex. The equations are written into non dimensional form, as a function of non-dimensional parameters which are famous Numbers: Euler, Damkhöler, Prandtl and Lewis. The reactive term is on the right side, and as you can observe it depends on the reaction chemical kinetics.

Although it can be extracted analytical conclusion of this combustion system using a Burke-Schummann analysis, a numerical computation is needed almost every time you formulate something like this. Theoretician physics and engineers are searching for analytical solutions of this kind of complex fluid flows, and a lot of papers are published every year about that. It is a field of research which is only in its early stages.

 

[FredGarvin]

I always considered the guys working on our combustors as practicing black magic. There's a definite art to designing them.