You will not be able to determine the spectra response from this, but you can determine power density in an approximate manner. I have not verified this method, but have compared it to the acnelamp result. At a later date, I should be able to compare this with test data using an optical power meter.

There are 3 steps.

1. Determine the power density at the source.

2. Determine the power density at the target.

3. Compare your result to published data.

Step 1.

http://heelspurs.com/led.html has a method to determine the power density. This is only approximate and not verified. I believe it is based on the conservation of energy. You can read how it was done, but here are few derivatives that I took.

- Obtain two foam coffee cups. For the first one use a caliper to measure exactly 2cm. You can scribe a line on the foam cup very easily and it will be fairly accurate.

- Then scribe another line about 2.5-2.75mm and cut the cup along this line.

- I first used a black marker to cover and relatively large sheet of alum foil black. Then carefully fold the entire piece inside of the cup, bottom and the walls, making sure not to leave air pockets between the cup and foil. You can cut the foil a bit to now see the scribe at 2mm.

- Make sure the top is very flat by turning it upside down and working it some.

- I placed this cup inside another cup that was cut down to make sure the original cup's top edge was still exposed. This provides additional insulation. Remember, you are trying to use conservation of energy so any leakage will affect the result.

- I placed the LED array on top of the cup and then covered the area with saran wrap to ensure that you would not get airflow between the unit and cup.

- Other than that, its basically like heelspurs.

Now, one important thing.

My results were coming out ridiculous at first. But a simple calculation will show you that the equation he uses is not correct. He says that he got 100mW/cm^2, but if you insert this into his equation, you would find it would need a delta of 7142 deg C, obviously erroneous. So I manually derived the equation and found that the error was that it should read 1W/cm^2 = 2 cm x 4.2 x C / 600.