How to set power of bulb?

[petr_AL]

Hi,

we would like to perform Steady state Heat transfer analysis of automotive headlamp to examine temperature on plastic parts surface.

How should be set the Heat generation of bulbs? When the bulb has 70W power, it should be divided by volume of bulb or by volume of all headlamp assembly?

Internal heat generation should by applied to glass part of bulb, I suppose.

The ambient temperature should be 50°C. I set this temperature using Surface convectional load applied on all parts. Is it OK?

Thanks

[Joey.X]

You can model this by steady state heat transfer analysis, the tricks are

(a) Parts involved

    The bulb, reflection parts and plastic parts should be the least parts in this case, you add more with increasing model complexity. 

(b) Thermal boundary conditions 

   - For model simplification, whole bulb (including glass) could be modeled as the heat generation part which has 70w power, and the whole part has uniformed thermal property. 

  -  It may need convection loads on some boundaries, these is needed for steady state analysis the thermal balance in whole model, the ambient temperature could be used as reference temperature. 

  -  For the external surfaces without any thermal boundary condition, it implies perfect thermal isolation, be sure that's what you want or not.

(c) Radiation modeling 

  - The heat transfer in this model includes radiation and conduction; convection heat transfer could be ignored for model simplification.  And be careful on radiation modeling which appears necessary in this case.

  - Define radiation surfaces and enclosure, where the surface of reflection parts is supposed to have low emissivity (close to zero).

  -  Remember turning on nonlinear iteration since the radiation heat transfer makes the model high nonlinear

[petr_AL]

Hi Joey,

thank you for your answer.

One question yet: how to set power of bulb (for example 70W)? The heat generation parameter is defined as power/volume (W/mm3) - see image. What volume should be taken with bulb power?

Thanks

Peter

[John_Holtz]

Hi Peter,

Based on your image, you are applying the 70 W to one part (see attached image). Therefore, divide 70 W by the volume of that part to get J/(s*mm^3). If you were applying the power to 2 parts, you would divide the power by the total volume of the two parts, and so on.

If you do not know the volume of the part(s), you can run the "Analysis > Analysis > Weight and Center-of-Gravity" calculator.

[petr_AL]

Hi,

I examine simplified test headlamp assembly with two bulbs

I set heat generation to glass part of bulb. It's volume is 1123 mm3. I want 70W bulb, I set 70/1123 = 0.06 value of heat generation.

Second bulb is smaller, 20W. From glass volume and power I have value 0.03 for heat generation.

Other assembly parts have set Surface conventional loads with 50°C ambient temperature.

For Steady state heat transfer I get temperature result - see image. Maybe I made some bad settings, because the maximal temperature is about 30 000°C. I thing, it is too much.

Can you help me, where I made a mistake?

Thanks

[Joey.X]

hi, Petr_AL

The heat generation setups look good (but beware it’s under an assumption I will talk in the bottom). And I agree that the result is too crazy for maximum 30 000°C as you mentioned, 

Do you use steady state thermal analysis?  I guess so according to your previous post, just confirm here.

In steady thermal analysis, the converged result depends on the heat balance of heat source (here the two bulbs) and heat sinks (all the participated part surfaces to the ambient). 

The issue appears from the other model setups, please check following items

- Check all (or major) surfaces which transfer heat energy out, do you leave some surfaces without any thermal boundary conditions? If so, it was assumed absolute thermal isolation, check if you have some surfaces violates the physical fact. 

- Check all part material properties are correct

- Check if the simulation is converged (usually from log files) 

- Do you setup radiation heat transfer in this model? From your picture, the result pattern conduction dominated. 

Back to the bulb modeling, it assumes the heat generation comes from the whole bulb uniformly instead of from the real filament wire inside bulb(albeit you can do this with much more complicated model), the result could be some off from reality, but what you majorly care is the thermal impacts on other part in whole bulb assembly, right?  If so, you can allow some tolerance for the maximal temperature which is located in center of the bulb, but you don’t care it too much, but care the temperature on other parts.

Regards,

[petr_AL]

Hi Joey,

thank you for your answer.

I'm using  Steady state Heat transfer analysis. All my settings you can see at attached .pdf file.

Heat generation is set to glass part of bulb - here it is simple cylinder.

We are not interested in temperature of bulbs, but in temperature of other plastic parts. My colleauges make real temperature testing on real headlamps using temperature sensors or infrared thermometer. They said, on bulb is about 400°C. This temperature should influence to reflector and other headlamp parts.

I set no radiation. I understand the radiation principle, but I do'nt know, how to use radiation here.

When I made weight analysis, some errors occure - see log file ds.lg9.txt.

Thanks

Peter

[Joey.X]

Peter,

Thanks for sending more details.

First, the log file shows error messages on the bottom, that indicates the result can not be trusted, so some model setting and probally some upstream log files need to be investigated.

Secondly, this model really need using radiation heat transfer to get resonable result, refer to my first post for the tips (c). For the fundamental ASIM/MP radiation modeling , you can refer to Autodesk simulation online wiki help from here

http://wikihelp.autodesk.com/Autodesk_Simulation/enu/2012

Regards, 

Spoiler