Warning: pic heavy, as usual.
Reviewer's note: This is the second prototype sent to me by Niwalker for review and feedback (see my initial prototype assessment here). To see my review of the final shipping version, please follow this thread.
As before, the MiniMax is a flood light that comes with two high-output MT-G2 emitters, and is powered by 4x18650. Specs are largely unknown as present, so I will jump right into the comparisons.
I will also give a summary of all the changes from the first prototype at the end of the review.
From left to right: Eagletac Protected 18650 3400mAh; Niwalker MiniMax Nova MM15 prototype #2, prototype #1; Foursevens MMU-X3; Jetbeam SRA-40; Sunwayman T45C; Eagletac SX25L3.
All dimensions directly measured, and given with no batteries installed (unless indicated):
MiniMax Nova MM15 Prototype #2: Weight (with handle): 332.7g (516g with 4x18650), Length: 115.3mm, Weight (bezel): 60.7mm
MiniMax Nova MM15 Prototype #1: Weight: 268.3g (452g with 4x18650), Length: 114.0mm, Weight (bezel): 58.0mm
Eagletac SX25L3 3x18650: Weight: 315.9g, Length: 150.2mm, Weight (bezel): 47.0mm
[Crelant 7G10: Weight 643.4g (827g with 4x18650), Length: 198mm, Width (bezel): 79.0mm
Fenix TK75: Weight: 516.0g (700g with 4x18650), Length: 184mm, Width (bezel): 87.5mm
Nitecore TM11: Weight: 342.6g (476g with 8xCR123A), Length 135.3mm, Width (bezel): 59.5mm
Niwalker BK-FA01: Weight: 687.6g (870g with 4x18650), Length: 209mm, Width (bezel): 80.0mm, Width (tailcap): 50.3mm
As before, the MiniMax is the smallest 4x18650 light I've tested to date – compare it to Nitecore TM11 to get an idea.
Let's start with a reminder of the what the first prototype looked like:
And now, the new prototype #2:
I said it on the first prototype review, and I'll say it again – this is a remarkably tiny light!
Exterior styling has been updated, but overall size and dimensions are not very different. The head is slightly wider at the top, and the light is slightly longer – but only by a millimeter or two in each direction. Anodizing is a black matte finish (presumed to be HA), and is in good shape on my samples (except for the screw threads, see below). There was still no labeling on my second prototype sample.
There is also still no real knurling to speak of, and ridge detail has actually been reduced from the earlier sample. But there is a new removable aluminum handle and lanyard attachment point now, so overall grip and anti-roll is good. One comment here – the handle is located very close to the body, so there is not a lot of room for your fingers.
The integrated battery carrier has been updated slightly, and there is now a protecting piece of plastic at the end of the exposed carrier end plate. Height of the built-in carrier appears to have been increased slightly, as it is now a bit easier to insert longer 18650 cells (the original carrier was very tight for length). That said, you may still find longer or wider high-capacity 18650s a bit of a challenge in this light.
As before, the four 18650 cells are in series, not parallel (i.e., 4s1p). There have been some circuit updates however, so please see my analysis below for user interface changes and standby drain/lockout.
As before, the body's aluminum handle is really just a protective cover - you can run the light without it. Threading is unchanged from first prototype – square-cut threads, but with worn anodizing. Of course, that doesn't matter, since there is no current passing through the body cover.
The head region has been re-designed, and is better organized. The switch is located directly in front of the handle, with a revised button cover (it now appears to be made of stainless steel). The center "N" logo is raised with a "grippier" texture. Switch traverse/feel is pretty much unchanged, and about typical for an electronic switch. The tripod attachment point has been move to directly opposite the switch on the head.
Again, here is the MiniMax Prototype #1:
And the MiniMax Prototype #2:
As before, the MiniMax runs off two MT-G2 emitters in relatively shallow reflector wells. The main difference now is that the reflector is more of a "light orange peel (LOP)" finish, compared to the heavily textured earlier prototype. This has not appreciably changed the beam – it is still very much a flood light (scroll down for beam shots).
A stainless steel bezel ring has replaced the previous black (presumably aluminum) one.
The user interface has been updated from the earlier prototype.
As before, turn the light on/off by the electronic switch. There is no momentary mode – once you release the switch, the light comes on in constant output.
Previously, the light always comes on in Hi mode first – the new prototype has mode memory (for the regular non-Turbo modes). It also has one additional regular mode, intermediate in output from the previous Hi and Turbo levels (I call this level "Higher" in my assessments below). This nicely fills in a gap that I felt was missing in the original prototype.
As before, press and hold the switch to cycle through all the modes in sequence: Moonlight > Lo > Med > Hi > Higher (i.e., there are now 5 regular levels, up from 4 previously). Release the switch to select the desired level. Note that that you can now restart the level ramp at any time (previously, you could only activate it once while the light was on). But as before, Turbo is not available on the regular mode sequence.
As before, double-click from On to access Turbo. Double-click again for strobe. Another double-click brings you to SOS (new on the second prototype). Note there is no mode memory for Turbo or the strobe modes.
There is now a lock-out mode, accessed by a rapid triple-click of the switch from off. The two MT-G2 emitters will flash twice to indicate the lock-out is engaged. Another triple-click re-activates.
Also new on this sample is a standby indicator that flashes when the batteries are connect but the light is not on. A brief green flash of the switch occurs every four seconds.
As before, the switch "N" logo lights up green when in use. The N changes to red as the cells near exhaustion – this is a warning to switch down to a lower level, as the light will shut-off automatically soon (due to an internal shut-down feature in the circuit). I provide details on this in my testing below. You can re-activate after the light shuts down, by pulling one of the cells out and re-installing.
For more information on the overall build and user interface, please see my video overview:
Video was recorded in 720p, but YouTube typically defaults to 360p. Once the video is running, you can click on the configuration settings icon and select the higher 480p to 720p options. You can also run full-screen.
Please make sure you watch on a device that supports annotations (and have then turned on). This is one video where I had to update the UI description after the fact – I erroneously refer to the new "Higher" mode as "Turbo" in a few places in the video. Please see the annotations in the video, or refer back to the description above – thanks!
As before, there is no sign of PWM that I can see, at any output level – I believe the MiniMax is current-controlled.
However, I did detect high frequency noise with my oscilloscope on some modes (Lo through Higher – but not Moonlight or Turbo) – consistent with the earlier prototype.
The frequency was extremely high – about 21kHz (up from 14 kHz on the first prototype sample). Either way, neither of these signals are visible to the eye in actual use. Rest assured, the light is fully flicker-free in all modes.
The strobe mode was a fairly typical 10Hz fast strobe.
Again, a fairly typical SOS mode has been added.
A standby current drain is inevitable on this light, due to the electronic switch. Despite how the carrier looks, the batteries are actually all in series, as before (i.e. 4s1p arrangement).
I have measured the current on the second prototype as 630uA initially, but it rapidly drops down over 30 secs or so to settle at 520uA. This is slightly higher than the stable 405uA of the first prototype, but still fairly low (i.e., would still take over 8 months to drain 3100mAh 18650s).
However, the new standby indicator on the second prototype now flashes once every four seconds, causing a jump in current to 2.63mA when it is lit. If we use ~1mA as a rough average current (i.e., averaging the current over 4 secs), that would give you a little over 4 months before 3100mAh cells would be exhausted.
As before, there is no physical lock-out available (i.e., you would need to pull one of the cells). But Niwalker has added an electronic lock-out mode (triple-click the switch from off).
I have measured the current in lock-out mode, and it is no lower than the stable standby drain (i.e., 520uA). However, the standby indicator no longer flashes, and the light cannot be activated accidentally. So, in this lock-out mode, except a little over 8 months before 3100mAh cells would be fully drained.
And now, what you have all been waiting for. All lights are on their standard battery, or AW protected 18650 2200mAh for the multi-18650 lights. Lights are about ~0.75 meter from a white wall (with the camera ~1.25 meters back from the wall).
Automatic white balance is used on most of my wall beamshots (to minimize tint differences), but in this case I went with a Daylight WB on my Canon for the MiniMax prototypes.
Note the original prototype is simply labelled "Prototype" below. The new second generation prototype is labelled "#2 Proto".
Note: No matter what white balance I use, these comparisons will never be entirely accurate for tint. In real life, I find my MT-G2 lights all to be relatively neutral white, without a huge difference between them.
As before, the MiniMax has an unbelievable amount of output on Turbo. Hard to directly compare, but my ceiling bounce results tell me that the second prototype is slightly brighter than the first. Scroll down for my detailed tables and assessments.
But first, here are some indoor shots in a basement, to allow you to better compare. For your reference, the back of the couch is about 7 feet away (~2.3m) from the opening of the light, and the far wall is about 18 feet away (~5.9m). Below I am showing a couple of exposures, to allow you to better compare hotspot and spill. And again, the camera is set to a Daylight white balance for all lights below.
Again, these are all done with a Daylight white balance, in better show tint differences. But a single white balance will never capture the true difference between lights – in practice, the X6 is not that cool, and the Minimaxes are not that warm.
As with the white wall beamshots, the second MiniMax prototype seems to be slightly more "throwy" than the first – but it's not really noticeable in real life, even with the lights side by side. Rest assured, this is still a pure flood light. Scroll down for actual output and throw measures.
Outdoor beamshots aren't possible right now, given the several feet of snow on the ground around here. But for your reference, here are the outdoor beamshots taken from my original prototype review. As always, these are done in the style of my earlier 100-yard round-up review. Please see that thread for a discussion of the topography (i.e. the road dips in the distance, to better show you the corona in the mid-ground).
Given that this location was picked to illustrate relative throw (which the MiniMax is not designed for), it doesn't really capture the overall brightness of this light very well – recommend you stick with the indoor basement shots for now.
That said, you can get a few hints of its relative brightness if you examine the far right end of the zoomed-out shots above (i.e., the tree line in the distance on the right). Depending on your monitor calibration, you may be able to faintly make the trees there (which are more than 100 yards away). You'll note how much hard it is to see these on the comparator SX25L3 or X6.
All my output numbers are relative for my home-made light box setup, as described on my flashlightreviews.ca website. You can directly compare all my relative output values from different reviews - i.e. an output value of "10" in one graph is the same as "10" in another. All runtimes are done under a cooling fan, except for any extended run Lo/Min modes (i.e. >12 hours) which are done without cooling.
I have devised a method for converting my lightbox relative output values (ROV) to estimated Lumens. See my How to convert Selfbuilt's Lightbox values to Lumens thread for more info.
Throw/Output Summary Chart:
My summary tables are reported in a manner consistent with the ANSI FL-1 standard for flashlight testing. Please see http://www.flashlightreviews.ca/FL1.htm for a discussion, and a description of all the terms used in these tables. Effective July 2012, I have updated all my Peak Intensity/Beam Distance measures with a NIST-certified Extech EA31 lightmeter (orange highlights).
Since my high-output lights don't fit in my lightbox, I am really relying on my ceiling bounce measures here. The second MiniMax prototype manages to squeeze out a few more lumens than my original sample – a very impressive ~5500 lumens in my testing
Throw remains definitely pretty minimal, given the overall output of the MiniMax. As always, the MT-G2 produces a very smooth beam profile.
Let's see how the rest of the output levels compare, between the two prototypes:
Again, take these with a bit grain salt, given my limited ability to measure output on these lights. But it gives you a general idea of spacing. As you can see, Niwalker added a fifth regular mode (which I am calling "Higher" above). This is the intermediate output level I recommended on my initial prototype review.
Another change according to Niwalker is the use of thermal step-down now on Turbo, instead of a timed one. Let's see what the difference is like, with and without cooling (on my standard relative output scale):
First off, I'm happy to report that the light steps down to the new "Higher" output level of ~2000 lumens.
But in terms of step-down timing, not much has practically changed. With fan cooling, the second prototype stepped-down after 4 mins runtime (just as the first prototype did on its timed step down). Without cooling however, the second prototype stepped-down slightly faster (3.5 mins in my testing). End of the day, this doesn't really translate into much of a runtime difference – although I agree that a thermal step-down is a safer design than a timed one, for super-high output lights like these.
So let's see how it compares to the first prototype, and others in this 3x/4x18650 battery class (again in my standard relative output scale):
UPDATE May 18, 2014: I believe there was an output reporting issue with the "Higher" mode test on my second prototype sample shown above. Please see my final shipping review of this light for updated runtimes.
I realize this graph may be a bit hard to decipher – the brown line that ends around 130 mins is the max Turbo mode of the second prototype, and the new "higher" output level is the brown line that terminates around 340 mins (off-scale).
As you can see, you get a lot more runtime from "Higher" mode than the Turbo mode (i.e., those ~4 mins at ~5500 lumens really drains off a lot of battery capacity). But I'm quite surprised by the relative efficiency of the new prototype compared to the initial one (i.e., compare to the Hi mode of first prototype).
Let's see if that continues at the lower levels:
And indeed it does – the revised prototype runs twice as long as the original prototype on "Hi" level, and several times longer on the "Med" level. Niwalker has clearly done something to boost the efficiency of the circuit, or is using higher output bins emitters (or both) on this second sample.
I am frankly surprised to see how efficient the second prototype is, compared to other lights in this battery class. This is making me think my lumen estimates may be a bit high for this light … it is hard to accurately compare lights with such widely different beam patterns.
I realize people may find it hard to relate to my relative output scale (which is based on the numbers coming out of my lightbox. Here is a re-plot of the max levels of the MiniMax in my estimated lumen scale:
UPDATE May 18, 2014: Again, I believe there was an output reporting issue with the "Higher" mode test on my second prototype sample shown above. Please see my final shipping review of this light for updated runtimes.
For those of you wondering what repeated re-starting of the light in Turbo mode would like, I have done the experiement below. Note that I actually let the light cool for ~10 mins between each re-start. I am only ploting the time the light was actually on below, so that you can match up the time scale.
As you can see, the output keeps dropping over time, on successive re-starts. For example, after 30 mins,you are down to ~4000 estimated lumens, compared to ~5500 estimated lumens at the start. You also hit ~2000 lumens just before the light shuts-down completely, after ~50 mins. Sorry I can't be more accurate for time, as I accidentally let the light step-down for several minutes before one of the re-starts.
I've done some detailed testing of the light once the switch indicator changes from solid green to solid red.
When the indicator first goes red at the "Higher" level, I measured an immediate resting voltage of the cells as ~3.46V on average. Re-installing the cells and switching down to the Hi level, I got another 8 mins of runtime before the indicator went red again. At this point, the initial resting voltage was ~3.40V on average.
I then ran the light on the Med level on these same cells. After 10 mins more at this Med level, the indicator went red again. I measured the initial resting voltage as ~3.36V on average at this point. I next ran these cells in the light on the Lo level. After another 50 mins at this level, the indicator went red again, with an initial initial resting voltage as ~3.30V on average.
In my view, this is MORE than sufficient warning to switch down to a lower level. I am sure the light would have run for a quite long time further on Moonlight, but was not prepared to wait and watch.
Note as well that even if the light does shut-down on you (and won't reactivate), just give the cells a couple of minutes to recover. All cells bounce back somewhat from their discharged voltage on their own. Once the voltage recovers enough for you to re-active, there is no restriction on what output level you can select (i.e., all modes are available). But keep in mind that you likely don't have much runtime left at the higher levels, if the light has already shut-off once.
Potential Issues and Updates from the First Prototype
I made a number of specific recommendations in my review of the first Prototype. Let's see how Niwalker responded to these, and then look at what else has changed.
- Selfbuilt recommendation #1: Need a higher "Hi" level, between the original ~870 lumen Hi and ~5200 lumen Turbo (preferably in the 1500-2000 lumen range), and use this for the step-down from Turbo.
- Check. Niwalker has added a fifth level to the regular sequence, of ~2000 estimated lumens in my testing. Turbo mode now steps-down to this higher Hi level.
- Selfbuilt recommendation #2: Allow the mode selection ramp to be re-started at any time, with mode memory upon re-activation.
- Check. You can now access the ramp repeatedly, and it recalls the last setting when turning off/on. Note that it still starts at the Moonlight level and ramps up.
- Selfbuilt Recommendation #3: Allow access to max output from Off.
- Sorry, still not available. Turbo level is still not available as part of the main memorized sequence, so you have to turn the light on first and then double-click for Turbo. You can do this pretty quickly though (i.e., single click, followed by a double-click). You just need a slight pause after the first click, or you risk activating the triple-click lock-out mode.
- Selfbuilt Recommendation #4: Add an electronic lock-out mode to prevent accidental activation.
- Check. Note the standby current is not altered when in lock-out mode (but I previously indicated that I thought the standby current was reasonable)
- Selfbuilt Recommendation #5: Add a standby/locator flash to the switch
- Check. The electronic button now flashes once every 4 secs when in standby. This can be turned off by entering the lock-out mode.
- Selfbuilt Recommendation #6: External build issues: the switch is hard to find by touch alone, the tripod mount should be moved opposite to the switch, and some sort of lanyard/carry device should be added.
- Check, check and check. The switch now has a metal finish (making it easier to see), with a raised rubberized logo (making it easier to find by touch). The tripod mount has been moved directly opposite. A removable handle (held by screws) has been added, with a split-ring and wrist lanyard attachment on the opposite side. The handle also helps you locate the switch easier (i.e., located right at the head).
- Selfbuilt Recommendation #7: This wasn't an official recommendation, but I did note in my internal build overview that the integrated carrier was very tight for 18650 cell length, and had exposed contacts on the end plate.
- Check. Niwalker has extended the length of the carrier slightly (although longer cells are still a relatively tight fit). There is also an additional piece of plastic protecting the end plate of the carrier.
What else has changed? Well, Niwalker has made a few other enhancements, based on member feedback here:
Step-down: The second prototype now uses a thermal step-down feature, instead of the original timed step-down. This was highlighted by several members here. However, as my testing shows, it doesn't make a practical difference – in typical usage, the new thermal step down will occur around the same time as the previous timed version.
Strobe modes: A SOS mode has now been added to the blinking modes, after Strobe.
Styling: It's a minor point, but the styling of the battery cover/handle has changed. A pretty neutral change in my view. I suggest you scroll back up to my build overview section, for additional commentary of changes over time.
Efficiency: Something unexpected was the greatly increased efficiency on the regular, non-Turbo mode. Check out my runtimes above, but the new prototype appears to be quite a bit more efficient than the previous sample (still with a largely direct-drive-like pattern). I have confirmed with Niwalker that their engineer has continued to revise and tweak the circuit (the second prototype I tested is apparently the 6th iteration he developed).
Other Potential Issues
New issues with the second prototype are fairly limited. One issue is that the new handle is located pretty close to the body – so unless you have very small fingers, you won't be able to grip it by the handle proper.
After twenty or so battery changes, I started to notice intermittent battery contact issues. This was caused by the loosening of the screws holding down the tailplate. Note that the carrier struts are required to carry current (given the 4s1p arrangement), so tension on the tailplate is important. As such, you may need to check and carefully re-tighten the exposed screws from time to time.
Niwalker advises this isn't the final shipping version of the MiniMax – but it is close. You should take this analysis as a near-final overview.
If you have jumped right to the end of this rather long review, I suggest you start with my summary of the build changes above, and then go back into the data/testing results sections if you want more info.
As always, I am impressed with the responsiveness of Niwalker to incorporate design feedback. Pretty much all of my suggestions were taken into account with this updated build – as well as some of the key points raised by members here.
That said, it's important to realize that this is not a blank check – Niwalker has worked to revise the existing build, not replace it. This is still a massive flood light, just with improved build, user interface and mode spacing (not to mention efficiency).
I am looking forward to seeing the final shipping version, which should be pretty close to this revised prototype.
UPDATE FEBRUARY 13, 2014: Niwalker has informed me of what plan to update for the final shipping version - see the list below. I must say, all sounds good from my perspective.
Final version changes, according to Niwalker:
- The lowest output will be lower, at least half as bright as it is right now.
- Momentary Turbo (5233 lumens) when in stand by.
- Double click for instant strobe when in stand by.
- A beacon mode after SOS (2000 lumens every 2 seconds)
- 2000 lumens SOS, instead of 5233 lumens
- Change the way to mount the handle and raise the handle (there won't be two empty space when handle is not installed)
UPDATE May 18, 2014: The final shipping version of this light has now been reviewed. Please continue all discussion of this model in that thread, thanks!
MiniMax prototypes were provided by Niwalker for review.