AA vs. AAA & Arc vs. CMG - The Numbers Make Sense

I've pulled together a spreadsheet to compare various data from Roy's runtime charts, his initial AA/AAA/Ultra brightness comparison, and mAH ratings from the Energizer site. These increased data points indicate the following:

-- The Arc circuit, or circuit in combination with their LEDs, is 60% - 70% more efficient than the CMG circuit.
-- The Arc AA is 50% - 60% brighter than the Arc AAA

I have 2 data sources for each conclusion - sources that agree with each other. Roy's Sonic/AAA comparison shows the Arc with 67% more footcandle hours. Roy's Ultra/AA comparison shows the Arc with 63% more footcandle hours. Probably not coincidence.

AA batteries have about 230% of the power in AAAs. The Arc AA doesn't go 230% of the AAA runtime, but is brighter. It's footcandle x hours figures add up to 240% of the AAA's. Not coincidence.

Now I'm really looking forward to getting my Arc AA. It looks like Peter took the well-established CMG on and beat it soundly. Smaller, lighter, brighter, longer lasting, better made, tougher finish, and it stands up on its tail. Well worth the extra money. Peter seems to have badly undersold the AA concept in years past.

Anyone have knowledge of how the CMG circuit works in comparison with the Arc? What could account for such a large efficiency difference? Electronics, bin rating, or both?

CMG uses a Zetec circuit, I think. I didn't realize the efficiency difference was THAT large. Also, if the Arc AA is that much brighter than the AAA, it sounds like the LED is severely overdriven, enough that I wonder whether I should worry about it popping. The manufacturer's maximum current rating in the spec sheet is there for a reason after all. It sounds like it has to be up in the 100 mA range. At that current level I'd feel more comfortable with an underdriven Luxeon than an overdriven Nichia. Does someone want to do a runtime test with a new L91 lithium cell, measuring the brightness carefully at the beginning of the run, and then at the end of the run put in another L91 and see if it's as bright as it was the first time?

Dave, the CMG circuit have been floating around the forum for quite sometime already. Do a search on it. For ARCAAA circuit you might want to contact dpr, he is selling ARCAAA circuit boards from the recent auction that Peter put up.

Great analysis there /ubbthreads/images/graemlins/thumbsup.gif

Vince,

Nice sleuthing, Dave! I think you hit the nail on the head.

I'm guessing Mr. Gransee bumped the output/current draw up a bit on these latest offerings.

The CMG uses an off-the-shelf Zetex circuit, the ARC is... well... nobody has said for sure (and shouldn't! /ubbthreads/images/graemlins/sssh.gif), but it's not the Zetex. Peter's attention to circuit/component detail are likely the reasons for the efficiency advantage.

If your AA ends up comparing to the one I recieved, I think you'll be very, very pleased!! /ubbthreads/images/graemlins/thumbsup.gif

Bob

Oops... paulr beat me to it!

I do wonder how overdriven the LED is, and if that may significantly shorten its lifespan... /ubbthreads/images/graemlins/thinking.gif

Bob

My Arc AA smokes my Arc AAA Standard, but is barely brighter than my Arc AAA CPF edition. /ubbthreads/images/graemlins/icon3.gif

I haven't gotten an ARC AA yet but from the sound of things, the LED doesn't really seem to be overdriven anymore than usual. I'm sure that the apparent increase in brightness is due to the larger reflector, which from Peter's video, seems a little more polished than a AAA.

could it be that since some of the recent buyers of the aa and cpf aaa don't find a visual brightness distinction the same circuit used in the aa is now used in the later versions of the cpf aaa?

Thanks for the circuit information!

The Arc AA seems to start out a good bit above 100 mA to the LED. It may average below that though. It will be interesting to see how it does in the long run.

There are certainly more qualified people than me to take on this technical question. But...I just spent a little time studying output graphs on the Duracell website. Seems to me that the difference in output between the Arc AA and Arc AAA may be nothing more than the fact that one is driven by a AA battery and the other is driven by a AAA battery. Current draw similar to that imposed by the Arc circuit draws down the voltage on a AAA much more than it does a AA. Peter probably provided the exact same circuit for the AA. He just didn't tip us off ahead of time that it would make for a brighter light when fed by the larger battery.

Hopefully Peter will chime in here somewhere. For now he should be enjoying the weekend. Maybe this does raise reliability issues for the emitter. If the Arc AAA was designed to be as powerful as possible without frying the LED, then the AA might be over the line. Then again, there might be enough safety margin in the figures, the LED might be well thermally connected to the body, and the emitter is only running very hard when the battery is fresh.

It does raise the lithium battery question though.