I can't wait to get mine, and see what it looks like with a MadMax+ Q3 in it.
I did a test run with the sandwich, 1 AA & the optic last night, and it looks adequately bright. But, I was just holding everything together by hand, it was kind of awkward.
I wonder if any of you plan on doing this, or if you already have. If so, are there any beamshots comparing the MadMax running on one cell & two cells?
I've put a MadMax+Q in it. Preliminary testing shows the mini-mini to put out about 75% of the amount of light that the MadMax+Q generated in a 2xAA minimag! Used the same Li batteries in both lights.
Added: The 75% value turned out to be on the high side. Using batteries from the same package, this value came out to be 55%-60% of the 2xAA value.
Future plans are to see if the madmax can be adjusted to higher output on one Li battery and then run a runtime plot on it. Then reduce the output of the madmax by 25% and do another runtime plot to see how much runtime is gained by reducing the output. [img]/ubbthreads/images/graemlins/grinser2.gif[/img]
Roy, I'm very anxious to see your results! I also bought one of the little torches, but don't have the MadMax yet. I would like to find that adjustment that will give me the most LUX for the most runtime! Don't we all? [img]/ubbthreads/images/graemlins/grin.gif[/img]
Mine should be arriving today, can't wait to see this!
Are you going to post the graphs in the Reviews forum, or elsewhere, Roy?
Now, some possible improvements to the Mini-Mini-Mag:
1. A lanyard ring
A small flat ring, such that it would fit between a Kroll and the body of the light, with an extension for attaching a small split ring for the keyring attachment. Picture this part as a very disproportionate figure eight. This is a non-issue if the original tailcap is used.
I don't find the size of the light to be too big for keyring carry, but then, I carry my keys on a full size Petzl carabiner, not in my pocket.
2. A smaller head.
The head on the smaller Mini-Mini-Mag looks as though it may keep it from being pocketable, pants pockets, at least.
A small head with a lens, or half ball, or a turned down NX05 would be nice, but probably very expensive and difficult to make. I imagine matching the thread pattern on the Mag body would be the hardest part.
I'm just brainstorming here, but I would really like a lanyard attachment for mine, any other ideas?
I must say that I'm pleasently supprised with the results of runtime testing so far! I'm not sure that I'll keep to my earlier statement and change the setting of the MadMax+Q! The plots will explane why.
I started by doing a runtime plot of the MadMax+Q with two AA Energizer e2 Lithium batteries.
Notice the almost perfect plateau...flat with a sharp drop to a much lower level of light. The runtime is about 2 hours and 50minutes. The time to 50% brightness is also 2 hours and 50 minutes.
Here's the runtime plot for the MadMax+Q in the Mini-minimag using one Energizer e2 LIthium battery from the same batch that was used for the AA Minimag and Madmax+Q. The runtime for the MadMax+Q in the Mini-MiniMag is 2 hours and 28 minutes to 50% of max brightness and 1 hour and 40 minutes to the break of the output plateau.
Using fresh batteries from the same package I calcuated that the Mini-MiniMag and MadMax+Q generated about 55%-60% of the light of the MadMax+Q running on two AA Li batteries. This figure is a gross approximation as I do not have a Light Meter. I'm using the miliamps being generated by a bare photocell. I'm assuming a linear relationship between light brightness and the current generated by the photocell. Few things in nature are linear.
With a runtime of 2 hours and 28 minutes to 50% of max brightness, and a brightness level of 55%-60% of 2xAA, I think I'll skip trying to adjust the ManMax+Q. That is a tedius job...for each adjustment its is necessary to take the madMax out of the flashlight, change the TINY pot, re-assemble the light, take a reading, and start all over again maybe 20+ times.
Very good! Thanks Roy.
I am surprised at the runtime on one cell, to say the least.
I calculated from your graphs, at least 55.5% of the two cell MadMax+ Q3's brilliance, and 98.6% of the two cell runtime. I would rather have a higher percentage of the brilliance, with less runtime, but oh well.
Half the brilliance of a MadMax+ Q3 is not inconsiderable.
(I guess the fact that it is more than half the brilliance, on half the supplied voltage is better than I have a right to expect.)
Perhaps two ½ sized AA NiMH, would supply half the runtime, at full brilliance.
I absolutely concur with the 50% point being the best choice for the definiton of "runtime", as well. (I think you and I both have argued in favor of this being a CPF standard.)
two 1/2 aa would do almost full brightness bumping up against the discharge limit of the batteries in my case the 210mah 1/3aaa nimh batteries run about 20 min in my single aaa nexstar at so close that its difficult to tell the difference in brightness but I can see a difference.
WOW! I just got my Mini-Mini-Mag today. Very impressive, even on one alkaline cell. A voltage check reveals 3.15V, at the luxeon.
This thing would really be kickin' on a lithium, and even more so on 3V.
It just fills my fist, with the Kroll protruding. Quite comfortable in the front or back jeans pocket.
One of the reviewers needs to get their hands on one of these (But I don't think I'll be letting go of mine). Rothrandir, has got a real winner here.
This body, combined with a MadMax or BadBoy, will not disappoint. I imagine, though, that a MadMax+ would be best in it, since you are taking a hit in brightness, and the MadMax+ is brighter than a BadBoy. It may also be so that a Q3L upgrade would guaranty more brilliance, as the luxeon may have a lower forward voltage than what comes on a standard MadMax+.
Initially, my 1 AA MadMax+ Q3L did seem about ½ as bright as my other MadMax+ Q3L in a standard Mini-Mag. But then I switched sandwiches between the lights, and I got a surprise.
The 1 AA light is almost as bright as the 2 AA! I would say it is closer to Roy's first estimate of brightness, maybe more. It seems to be over 70% as bright as the 2 AA.
The only idea I have, is that perhaps the forward voltage of one of the luxeons is lower than the other, resulting more output from the same input voltage.
I measured the voltage at the luxeon on both sandwiches, w/fresh alkalines: the 1 AA - 3.27 V, and the 2 AA - 3.35 V.
[ QUOTE ] Roy said:
I think I'll skip trying to adjust the ManMax+Q. That is a tedius job...for each adjustment its is necessary to take the madMax out of the flashlight, change the TINY pot, re-assemble the light, take a reading, and start all over again maybe 20+ times.
[/ QUOTE ]
Roy, I was thinking about this, and I think there's an easier way.
If you have some tiny clip leads, you could clip onto the luxeon leads, apply voltage to the MadMax sandwich, and adjust it until you find the max voltage across the luxeon. This would be a one shot adjustment, with no guessing.
It has been stated ( by unknown authority) here on the CPF, that the output of a light has to change by 50% or better for the human eye to detect the change in the output of the light.
If the location of the adjustment pot were rotated 90 degrees so that the wiper arm faced to the outside, adjusting the MadMax+ would be much easier. I'm sure that there is very good (unknown to to me) reason for it's current position! [img]/ubbthreads/images/graemlins/smile.gif[/img]
Sorry, I wasn't clear as to which thread I was refering to in the 09:52 thread. I was refering to a visual estimation of light change.
As I see it, the problem with ONLY checking the voltage Is a plateau effect, where the light output stops changing with increasing voltage. I don't know if the MadMax+ has a plateau effect or not. Someone with the patience to measure voltage vs light output and plot the results, could answer that question. [img]/ubbthreads/images/graemlins/grin.gif[/img]
I was basing my assumption on the reports I've read here, that an increase of 0.1 V on a luxeon, could result in an increase in current of 100 mA or more. (after the forward voltage is met)
I haven't the resources to test luxeons like this, myself. I figure it sounds reasonable. Once the forward voltage is exceeded, any increase in voltage would increase the current through the luxeon, to some degree. But the question is, is the increase in brilliance worth the extra spent power? The answer would be different for each user, I guess.
What do you think, Dat2zip? Is the above method viable for adjusting MadMax brilliance? How do you do it before shipping them?
The Madmax Modules are tested before they are shipped, but, only for a general gauge of Pass/Fail criteria. Depending on the module with 3V applied the module should draw some amount of current. Module to module and LED variances create a small range and we generally look for gross errors. Like way too little current, way too much current and no light conditions.
The trim pot creates a voltage divider from Vout to the feedback pin. Adjusting it to wide open is basically setting the trim pot wiper to GND.
In the new board, there is a SMT resistor from the feedback pin to GND which accomplishes the same thing.
Any diode or LED follows some IV curve. An increase in Vf is definitely an increase in current (I).
Since Madmax is a voltage regulator the Wide Open mode is really setting Madmax to try to regulate to 5V. Since it can only deliver X amount of power, The LED clamps this to some lower level and the Madmax dumps the maximum power to the LED in an vain attempt to try to raise Vout to 5V. Since it never acheives that stability it is running wide open... It's like you have the petal to the metal on your old car, but, it never goes faster than 45MPH.
There are only a few tricks that will boost Madmax output. 1) Use a higher Flux LED like an R or S ranked Luxeon. 2) Get a Luxeon with a lower Vf bin code. The lower Vf means more current to the LED. P = V * I
Where P = Power (Constant in this case)
V - Vf of the LED.
I - Current to the LED.
This simple equation shows that if the Vf of the diode goes down with fixed power, then the current (I) would go up to keep the equation balanced.