If you can make a poor man's runtime meter...

[AilSnail]

If you can make a poor man\'s runtime meter...

I can make an even poorer integrating light meter!

The apparatus is a 2"wide white paper arc over a photoresistor from Doug Owen. the stars were placed beside the PR facing the same direction. a collar is around the PR to stop direct light and to mount Doug's 50% filter.
three different badboys were used for a variety of amperages.

the bb500 has become a bit unreliable after i cut and reattached the fb pin, but it seems to be consistent when it works.

It looks like the filter does not remove 50% of the light in this application. Hopefully I can take some pictures of the setup.

Edit: I believe the decrease in light from 1250mA to 1450mA might be due to the low Vin of the BB700 and BB750.

The battery pack were 2x2 D alks.

The measurements were taken after about 3 seconds.

The units of R are Ohms of the photo resistor.

Double pole switches were used on the Vin and Vout of the badboys to quickly set the current.

Conclusion: Though the test method is probably riddled with error sources, it leads me to believe that parallelling wayne's boards to get more than 750 mA tohrough the single dies (1w and 3w) is an excercise in futility. I don't know if the apparatus favors low domes, but it looks like it based on the flux bins.

 

[AilSnail]

Re: If you can make a .... [pics]

Added pics and some more comments.

 

[Doug Owen]

Re: If you can make a .... [pics]

Awesome!

Thanks very much for sharing.....

Doug Owen

 

[AilSnail]

Re: If you can make a .... [pics]

Thanks for the encouragement, /ubbthreads/images/graemlins/blush.gif

I intend to make a bit more reliable setup, covering the led instead of the resistor, getting a stronger batt pack (buy new batteries) and add more comparisons when more leds arrive. The two Q4H's on one BB750 each could be used to qualify the filter a bit better. There is certainly something wrong somewhere, and I suspect it is (at least) the modded bb500 putting out too little juice. If the filter (in my setup, with leaks, fingerprints and all) really takes away 50% of the light, then the 700 puts out twice as much light as the 500, as seen in the graph.

Hmm, I want to plot some of the other variables against each other as well...

 

[INRETECH]

Re: If you can make a .... [pics]

Lots of tape, Red/Green would be proud

 

[AilSnail]

Re: If you can make a .... [pics]

Forgive me, for I have sinned. I did not have any dyke tape.

 

[Chop]

Re: If you can make a .... [pics]

Geez, all of that to replace a stop watch?

Nice setup. /ubbthreads/images/graemlins/thumbsup.gif

 

[Doug Owen]

Re: If you can make a .... [pics]

I'm not sure I fully understand the data, but agree that it seem that the light out at '500' is half that at '700'. I assume the current is estimated, based on the model? That is you didn't adjust anything to exactly 700.0 mA for the tests?

There's some interesting data WRT Vf (which is a good direct 'measure' of LED current). Note that there is very little gain in LED current (as seen in either Vf or reistance) in the 'high power' tests? I agree again, no advantage.

Interesting data, please keep it up. BTW, the best testing I can do (or anyone has told me about) leads me to believe that the filter is within a percent or two of exactly half value.

Doug Owen

 

[AilSnail]

Re: If you can make a .... [pics]

Chop: Are you referring to the "PM runtime meter"? This is more like a "PM relative light versus current and vf meter"

Doug: Yes I have not measured the current, so the current estimates are only based on the names of the badboys. The Iadj on the BB500 may be floating, as I am not sure I successully soldered it back to ground after it broke.

"There's some interesting data WRT Vf (which is a good direct 'measure' of LED current). Note that there is very little gain in LED current (as seen in either Vf or reistance) in the 'high power' tests? I agree again, no advantage."

If you are correct that the led current follows Vf in some way, then where is the extra power dissipated?
Look at Vf (or R) against the Vin as well. Take 750mA and 1450mA (BB700+BB750) for instance. The Vin to the boys goes down, indicating a much higher load, but both the Vf of the emitter and light output goes down as well, or stays about the same.

One potential big source of inaccuracy is the temperature. I did not let the stars cool off very much between each power setting. If I do more testing that will have to be done. Also I need to do a proper amp reading when the smt resistors from dat2zip arrives.

"I'm not sure I fully understand the data, but agree that it seem that the light out at '500' is half that at '700'."
snip
"BTW, the best testing I can do (or anyone has told me about) leads me to believe that the filter is within a percent or two of exactly half value."

Two things that could account for that: too little output from BB500, and leakage around the filter and collar.

Ok, I'll read through this post after some sleep, see if it makes sense.