I really dont think your understanding HKJ’s discharge testing data. The P42A is good, but not the best. Certainly not the hardest hitting.
Here is the 20A test.
You can see that when fully charged the 30T has less initial voltage sag meaning it will “hit harder” and produce higher lumens. It will also pull higher amperage due to it’s lower internal resistance.
Just keep in mind that this is only important to drag racers or braggers that have integrating spheres and can measure the slightly higher output. If your a normal user, you are better off sacrificing that max output (which you can’t see with your eyes) to gain some additional capacity.
I suppose its close enough I guess, to be fair; but I still feel the Molicel is better overall fpr most people where overall battery life and performance becomes more important over some extra brightness for 30 seconds or so, but yeah I guess it all boils down to preference on those three (Molicel P42A vs Samsung 30T vs 40T). How long does 0.5 amp hour last at 20A? Because that’s only how long it will be brighter for…
That's ridiculous, calling the 40T average - plainly wrong. Look at the 40T at 20A. The 40T is my preferred 21700 for high performance. The P42A drops quick early on, other than that it's a very good cell but still prefer the 40T. I can't use the P42A for output test measurments for lumens and throw because of that initial drop because all that matters is the first 30 secs.
Yeah I was thinking 30t which suffers more overall voltage sag, so I’ll remove the comment RE: 40t/30t. I guess if one wants to drag race for a few seconds or so, the 30T and 40t are actually a bit better choices because of the higher initial voltage; but still… the P42A holds the higher voltage throughout for most of the battery life (over a lot more time) making it the hardest hitting battery overall; which is why I think it is still the superior choice for the average guy when using the regulated modes (which is most of the time after FET is no longer in use). How long does FET last starting @ max before going to regulated modes (7135 chips)?
Okay, your not understanding. When people measure lumens they are only refering to fully charged cells. Mainly just the first 30 seconds or so. The term “hard hitting” refers to the initial hit of power, not the later 95% of it’s life. Once you drain the battery a little the output drops, on these FET powered lights at least. No one really cares, performance wise, about voltage sag in the other 95% of the battery output.
What your refering to has more to do with Low Voltage Protection (LVP). This is a different aspect of the battery not really related to lumens or brightness. Keep in mind that a flashlight will not still be pulling 20A after the first few percent of it’s use, amperage slowly drops with voltage. The advantage of the P42A is that it allows longer and slightly brighter output before LVP kicks in. Note, this is a legit performance aspect for Boost drivers, but not FET.
With the 30T example above, you can see the P42A is the more practical choice because is sacrifices a bit of max output initially for more capacity later on.
The 30T, on the other hand, sacrifices capacity to squeeze out the maximum output possible in the first 5 percent or so. The 30T is so extreme it can do 50A no problem.
Anyway, it’s a bit like a dragster against a camaro. The dragster (30T) is extreme power for a short amount of time and the camaro (P42A) is not as extreme, but can go a longer amount of time. Battery manufacturers can tweak their formulas to produce different output curves. Just look at the Samsung 30T, 40T and 50E. They are all newishly designed cells with 3 different capacities and 3 different output curves to meet different product needs.
Yeah that helps explain FET a bit better; once voltage drops enough, FET is not used anymore and it goes to regulated, so hard hitting doesn’t matter so much at that point. Thanks for that info. So I think I now see what your saying; and so the advantage of Molicel’s higher overall voltage is actually not so relevant of a thing for FET flashlights (using FET mode), for a couple of reasons (lower initial voltage and then going into regulated after FET), but still, the Molicel will help overall with boost drivers and also any regulated voltage levels including the ones after FET becomes unavailable (which is like 85-90% of the overall run-time and where the Molicel will outshine the others, right)? Thanks for clearing that up, I don’t claim to know everything and will accept good information when it presents itself!
The four copper heatsink plates I ordered should arrive soon. Thank you, man of light. USPS tracking shows they arrived in New York City on February 7. Oddly enough, the tracking number never worked in 17track.net, but it does in Postal Ninja.
I have a question while I’m waiting for the plates to arrive. I own four Astrolux MF01 Minis and six other Anduril lights. I’ve watched YouTube videos showing how to configure the current and maximum temperatures. I had no trouble setting the current temperature but, even though I duplicated the exact steps shown in the YouTube videos, I simply cannot set the maximum temperature. All ten of these lights give four flashes followed by a brief blink as the maximum. I presume that means 41 degrees C. Again, doing exactly what I saw on YouTube, I’ve entered anywhere between 22 and 30 clicks to set the maximum temperature, yet the lights always remain at four flashes and one blink.
I must be doing something wrong, but I can’t figure out what it is.
That’s a good question. I’ll do some research on it.
Keep in mind that the batteries are tested on a special rig that maintains a constant 20A load from 4.2v down to 2.8v or so. This is very different from a flashlight with a FET driver whose amperage starts off high and gets hot, then throttles down to prevent burnt hands. Even with temperature protection turned off, a FET light will steadily drop amperage as the voltage drops.
I’m still curious how long 0.5 amp hour is time wise. It can tell us how many 30 second bursts of Turbo a certain battery can do before the output matches another cell.
BTW, once a battery hits 3.7v-3.75v (resting voltage, not under a load) it’s mostly discharged. That’s a good time to recharge it. Using Turbo on a FET light with a resting cell voltage of 3.7v will probably start tripping LVP. What you can do is use HKJ chart and compare 2 batteries under a light load like 3A and see how much extra run time you get at 3.7v. Below that voltage I mostly ignore.
Certain lights you might use for long durations at even lower amperage, so I look at the 1A output down to 3v. Those types of lights (headlamps) really benefit from the higher capacity cells.
Hhmm. Anduril does not blink out the max set temperature, it only blinks out current voltage and current measured temp. The brief blink means 0, not 1. A regular single blink means 1. It almost sounds like you are seeing the voltage blink out, not the temp maybe?
--> From OFF: 3 quick clicks to voltage reading, then 2 quick clicks done 3 times to get to temp reading.
(ugh -- this is why I eliminated those 2 modes between voltage and temp (modding Anduril). It's a PIA skipping thru them).
we are talking here about a low power density of like 0.35W on 10mm², a 0.1mm gap with paste is enough,
you don’t need like 10-15um to get ultra low thermal resistance
in any case the AMCs won’t be totally even and the housing thickness can vary like 0.1mm from batch to batch
I’m not sure where you are located, but Banggood sells lots of thermal paste. I hear the MX4 is non conductive and non drying. You’d need to do some research to find a good one. You only need a tiny amount.
Conductive thermal grease is big $$$ in general, I know - I have some (great on conductive thread joints), but supposed non-conductive could have low conductive properties - to me it's a little scary using a lot of thermal grease on components in a driver. I would suggest not to over do the grease, and a thicker type would hopefully be better - some are soupy, or get soupy under high heat. Wonder if the cheaper thermal 'glue' would be better - the kind that hardens to a soft malleable consistency.
Wait, are you sure about this? Wouldn’t a zero be a blank? Like 4.0 volts would show 4 blinks and longer than normal space (no blink for the zero), then 4 blinks, etc…?
Your saying Anduril does a shorter than normal blink for zero?
