So you are saying natural convection cooling and still air cooling are the same thing?

I am saying that the heat sink specs for natural convection assume no airflow. Period. Obviously the air does not sit dead still, there is some movement but there is no forced airflow. Convection effect on a heatsink sitting on top of a PCB is basically useless since the board blocks vertical airflow. Even if you get the board oriented vertically and have an air path, convection on small SMD heatsinks is pretty useless because they are close to the PCB surface and there are other components about the same height blocking the air flow so it basically "ski jumps" across the top of the components greatly reducing effect. You also have other power devices "pre heating" the air in the vicinity which raises the ambient temperature the heatsink is surrounded by. SMD heatsinks generally perform a lot worse than expected due to these effects.

 

You make many good points. Still, I would say that if someone ballparks some numbers and gets one answer, and someone else builds and tests something directly, and gets another answer, I lean towards the latter answer. Then I ask some questions about how the tests are done.

So, why not ask how the testing was done and how the design is validated? We don't even know what specifications are established. What is max operating temp? What is the required MTBF?

Personally, I've run D2PAK transistors at 3 W in a closed package with no heat sink on the board. I had the thick compressible thermal pads on the board which helped conduct heat away to the aluminum housing that had dimensions of 1x6x5 inches. The package was intended to be mounted to a heat sink but tests we did in a thermal chamber did not use one because the chamber caused airflow around that package. However, since the package was completely sealed without vents or fans or anything, the internal volume really was still air. Also, in this package was a DPAK running at about 1.5 W. This entire package went through thorough normal industry validation steps in design and testing. The max temp specification was 65 deg. C which was validated by manufacturing dept. testing which monitored board and component package temperature. MTBF calculations were done on the design with consideration of component stresses with temperature and the validaed MTBF exceeded the required 100,000 hours spec. Aside from the calculations, various accelerated aging tests were done, and the unit passed. Also, as far as I know, this design was never found to have any reliability issues by customers that used it.

So, in my personal example, we're talking about 4.5 Watts on the circuit board in a closed space. Although there was a thermal pad to help reduce hot spots, these pads are not all that great in terms of thermal conductivity. So, the possibility of 9W with free air convection with PCB mounted heat sinks does not sound outlandish to me. Maybe max temp is less, and maybe MTBF is less, but if it meets the established specs, no problem.

 

the problem i have is i have no idea how other people are going to use it.
With the different kinds of lasers and different power levels.
Some only need 2.2V others need 4.6V some people may run all of them off 7.5V and let it drop the voltage.
And in case they do i want to make sure it can handle it.
I have looked for other packages that have the same high hfe gain but it seems hard.
Also i am only able to use STmicro parts.
the best i was able to find was a unit no longer in stock http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATASHEET/CD00172270.pdf
So i went looking for others but still cant find any in a package meant to be mounted with that hfe.

If you are going to be selling these to people giving them a range of voltages that can be used as input, you REALLY should switch to a switched power supply solution. ALL users assume a "safe-zone padding" of your min and max voltage specs, so if you publish the max at 7.5V, somebody will try it at 9V, it will smoke, and they can state it was running at 7.5V

With a switch mode supply, you can make the range double that of a linear supply, and get more consistent response from what you are powering. Dissipation doesn't usually even reach 1W, let alone 9Watts of heat to worry about.

Lots of sent back products due to failures can give your company a very bad name, so test and don't assume your supply will always be used with what it was designed to power.

End users also tend to abuse electronics, stacking them atop each other, putting papers or books on/beside them, blocking cooling, etc. Keep that in mind when designing, a warning sticker works rather well, such as "Do not place objects within 3" of any side of this box"