Setup to Measure Nav/Strobe Intensity (candela)

[az_gila]

Same ole...

They have not been able to figure out how to manufacturer consistent balanced combinations. What they do to overcome this is called "bining". They check each LED White component block against a standard, whether that is warm, or cool, or in between. It is then "binned" according to where it falls. Manufacturers can then order from that "bin". ....

....actions (binning) as the silicon manufacturers have been doing for years when they can't control a process accurately....

I first saw this in action at Signetics in the late 70's watching them make op-amps. The good ones with a low offset cost a lot more, but came off the same line.

These were under $2 items and the testing machine was spitting them out many a second. The output "bins" were actually 1 gallon plastic kitchen pitchers!

The final price of each part (same part number, just different "dash" suffixes) was determined by the fullness of the appropriate bin at the end of a production run...

 

[Dan Langhout]

The denial is over . . . .

. . . . . about this new light. I went up to my friend David Edgemon's shop tonight for a reality check. David has an actual Whelen A555 rear position light, a real Whelen strobe tube driven off of a strobe power supply, and a rear position light of his own making with a Luxeon Rebel LED mounted in a Whelen A555 housing. I brought my LED position / strobe light and my light meter and we proceeded to hook all this stuff up, put it to the eyeball test, and take a few measurements with the light meter.

First comparison was between my light and David's home brew light.

Just in case you can't tell , David's Luxeon Rebel light is on the right. The eyeball test was no contest. Hand held light meter measurements at about 1 foot initially gave around 100 foot-candles (candela) directly on axis which tapered off to around 26 or so around 70 degrees off axis. This is more than adequate for a rear position light . . . . and the LED doesn't even have a focusing lens on it! This light is drawing around 350 ma or so. We did notice that after the light had been on for 45 minutes or so and had heated up a good bit that the light output dropped somewhat - to about 70 candela on axis! Even the 70 degree off axis level was still close to 20. LEDs light output are sensitive to junction temperature so this is an expected result but the light looks like it would still meet the spec.

Next up is David's Rebel and an actual Whelen A555 incandescent light.

The Whelen is on the right. These looked pretty close to the same brightness but the meter said the Whelen was around 57 foot-candles at 1 foot (about 57 candela). The incandescent light seemed to taper off less as you went off axis. I don't recall the exact value but it was still well above 20 at the 70 degree mark.

We also tested the Whelen flash tube. I believe David is using a NOVA EPSX power supply (maybe he will chime in here and correct me if I'm wrong). The eyeball test here was REALLY no contest! No photos of that but I did grab some peak foot candle readings using the peak function of the light meter - the first peak was up over 4000 foot-candles at 1 foot. That's over 4000 candela peak folks! Of course the effective candela is much less than that since the pulse width is so short but the point is that the meter certainly has no problem capturing an intense short pulse.

The bottom line here is that I'm rather inclined at this point to believe that my fairly crude initial measuring methods are perfectly valid for ballpark numbers and the hard truth here is that my LED light is nowhere close to being intense enough to meet the requirements for a rear position and strobe light. Somebody please convince me I'm wrong.

 

[AeroLEDs_Support]

Measuring Candela

Dan,

To answer your questions about how to measure your position lights/strobes:

1. The meter you have should work fine for measuring the position lights. The main question is how well it is calibrated, but assuming it is reasonably well in calibration, you can use it to do what you are doing.

2. Yes, multiplying by the distance in feet squared will convert foot-candles to candela.

3. You need to make your measurements far-field, not near-field. SAE test standards call for 60 foot measurement distances.

4. You need a non-reflective room (painted black) or an outdoor test range that has a reasonably dark ground (black asphalt or green grass) to avoid having significant reflections affect your readings. You need to first measure the ambient light level so you can subtract that from your readings.

If you want to measure the effective candela, you need an integrating photometer that can measure foot-candle-seconds. The effective intensity per flash is then calculated by multiplying the foot-candle-seconds for a given flash by the distance squared, then dividing by 0.2 plus the flash duration in seconds (typically a fraction of a second). The meter you have can't measure foot-candle seconds, but if you can measure the foot-candles when the light is on, and can measure the flash duration, you can calculate the foot-candle-seconds yourself...

Hope this helps,

Dean Wilkinson
CTO, AeroLEDs

P.S. I have been wondering about Aveo's claims for that product myself, as I have a hard time believing they are meeting the 400 effective candela requirement. They may be only meeting the 100 effective candela requirement that applied to airplanes certified prior to 1971. If so, they can probably be used on experimental airplanes, but they can't be used on airplanes that were certified after 1971. If I had one of their units, I have the ability to measure it. I may have to get around to doing that before too long...

 

[Dan Langhout]

Thanks . . .

. . . . . for the information Dean. As I had mentioned in an earlier post, I was just looking for a sort of "order of magnitude" check about the intensity levels. I figured that I could probably get a reasonable ballpark measurement of the position light and took the approach that if I could capture a peak intensity value for the strobe, I could do a back of the envelope calculation about the strobe being close (the peak would need to be a fair bit above the required effective value to have a chance of working out through the Blondel-Rey equation). I really had no intention of trying to accurately characterize the performance.

Myself and several others had wondered how compliance testing was actually done - so your information is interesting. I've ended up learning way more about all of this than I intended!

 

[JonJay]

Current limiter

. . . . . This light is drawing around 350 ma or so. We did notice that after the light had been on for 45 minutes or so and had heated up a good bit that the light output dropped somewhat - to about 70 candela on axis! Even the 70 degree off axis level was still close to 20. LEDs light output are sensitive to junction temperature so this is an expected result but the light looks like it would still meet the spec.

My understanding was that the Power Supply monitors the LED somehow and when the junction gets too hot, it reduces the current to the LED to keep it from destroying itself. I do know some power supplies function this way, and I am pretty sure Luxeon does.

 

[Dan Langhout]

My understanding was that the Power Supply monitors the LED somehow and when the junction gets too hot, it reduces the current to the LED to keep it from destroying itself. I do know some power supplies function this way, and I am pretty sure Luxeon does.

In this particular setup, David is using an LM317 linear regulator set up to do constant current regulation. It's simple and electrically quiet. The Luxeon Rebel stars don't have any sort of internal power supply. The LEDs just get as hot as they will based on power input and heat transfer through their mounting.

 

[az_gila]

I was fairly....

........

The bottom line here is that I'm rather inclined at this point to believe that my fairly crude initial measuring methods are perfectly valid for ballpark numbers and the hard truth here is that my LED light is nowhere close to being intense enough to meet the requirements for a rear position and strobe light. Somebody please convince me I'm wrong.

...convinced you were in the ballpark quite a few posts back....