TK's Emisar D4 review

Probably not. If she is an exceptionally cautious child, then maybe… but that would be pretty rare. Many adults can’t even use a D4 safely.

Start with something safer, and let her work her way up to more powerful items as she demonstrates she is responsible enough to use them.

Hmm, maybe not. I think I’m gonna try it though, mostly because I only have four Nichias to my name, 3 4000ks and a 3000k. There’s a fourth 4000k in my buddies H03 that I’m gonna swap for a LH351D next time I see him, but that won’t be for several weeks. I really want a nichia D4, the D4 was my first light that I got before I knew what tint I wanted, so it’s a 6500k xpl-hi, no gracias lol.

Will do that :slight_smile: Would a convoy s2 be weak enough? It needs to use rechargeable batteries and I have lots of 18650 around.

Is there more info about eye damage from powerful flashlights around here?

Some degree of trust is required, of course. People can’t even go outside without being exposed to something dangerous. I mean, some kids stare at the sun just because they can. So this is really something you have to determine on your own, based on how trustworthy your child is.

Almost any LED can cause eye damage if it’s running at more than a few milliamps. Staring at a modern LED is a lot like staring at the sun. And almost any battery can start a fire if the user shorts it to itself or something. So I can’t really say any flashlight is 100% safe.

A S2+ with a 3x7135 driver (or, even better, 1x7135 driver) is a lot harder to cause damage with. However, I’d suggest going even smaller if possible. Maybe something with a single AA or AAA Eneloop battery at first.

OTOH, a parent’s job isn’t to protect their child from the world, but rather, to help their child explore the world. If they make a mistake and start a fire or hurt themselves or something, it’s not necessarily a bad thing, as long as the damage isn’t permanent. It may help them learn caution. Making mistakes is one of the fastest ways to learn, as long as the mistakes aren’t too big to recover from.

TL;DR: It really depends on what kind of person your child is.

Hi TK & all. Looks like I’m getting sucked in to BLF again, at least for a little bit. :slight_smile: Thanks for posting!

A:
I apologize in advance for anything I’ve missed in the thread - I’ve only hopped around in it some, there is a lot of good content.

@TK - Something from the OP confused me:

I saw in the OP that you’d improved this ~4.5mA figure in Anduril but I couldn’t find a post discussing the improvements or how they worked. I also saw your comment (“”#1570”:TK's Emisar D4 review - #1570 by ToyKeeper “) about how the light didn’t go into standby immediately and the next few messages between you and rizky_p explaining that the actual standby drain (as measured by rizky_p) is 24.5μA and that a project of your eliminates the delay.

I guess I’m asking for more information on the quiescent current / parasitic drain situation.

  • Do you know whether currently shipping lights have a firmware with low parasitic drain?
  • Was the move from 4.5mA down to 0.02mA purely a matter of firmware and/or changes in measurement strategy?

B:
I’m also interested in what’s needed (from a hardware standpoint) for a good implementation of something like Anduril, Narsil, or RampingIOS on an FET+1 driver. After reading about it I really want to order an Emisar D4 right away… but I’ve got a small handful of good momentary switch lights which need drivers and I want to rehab at least a couple of those before buying even MORE flashlights. I have an embarrassing number of non-working flashlights.

As many of you are probably aware I am VERY out of the loop. :frowning:

  • I’ve read that a low-value resistor has been added to FET drivers in order to reduce ringing and keep the MCUs from rebooting at high currents.
  • I’ve read that resistor dividers for momentary button flashlights should use high value resistors to minimize drain.
  • What do I need in order to get started with thermal regulation?
  • … what else did I miss? :wink:

I’m no expert on the matter but I’d hazard a guess: spot brightness is probably the greater enemy vs total light output. With that said, even 300lumens from a single LED can leave after images that stick around in your vision for hours… so I’d take TK’s suggestions very seriously. In addition to those suggestions maybe you could consider using a diffusing film. eg site:budgetlightforum.com film - Google Search

Driving very large emitters (or multiple emitters) at low currents is also a way to reduce spot brightness. I suspect that you could (safely?) get away with a much higher output if you used a large emitter coupled with a diffusion film vs a single XP-G / Nichia 219 / etc.

A custom low-powered triple or quad might be the cheapest way to get spot brightness down while still retaining a large amount of out-the-front illumination. Or something like a C8 host with DC Fix film with an XP-L or larger-die emitter (such as an SST-series emitter). The C8 gives a large window to apply the DC Fix over.

FWIW I’m interested in this topic as well, although I haven’t really explored it. High powered flashlights are dangerous and I’m uncomfortable with the idea of children using them. Suggestions or explanations of why I’m wrong about spot brightness or how to deal with it are welcome!

Will need to read more up on it. The D4 is def bright. I read that my Maxa Beam does do damage to the eyes but from what I understood that was not because of the intensity but the fact that short arc creates a lot of UV light and not all are filtered out by the glass. The maxa is going to be locked in my weapons cabinet tho as it’s def not something for a kid…

I’m with toykeeper on the part about parenting. This is why I do let here play with it while I’m watching and explain that it’s dangerous to point it directly at someone and that it’s very powerful so she can not adjust it with the button. This is teaching here about the dangers but still allowing here to explore. There’s lots of things that are dangerous when not handled properly so they might as well learn instead of avoiding… But if short therm exposure of the beam can be harmful it’s not worth the risk of a kid not listening because there is no second chance with the eyes.

She has weak AA and AAA flashlights but it’s more fun to borrow mine…

Maybe I should make a separate thread instead off taking this off topic :zipper_mouth_face: ?

I don’t know if it’s allowed with links to another forum but this thread below was a good thread to rise awareness about lasers and might be somewhat relevant to powerful leds also so I’ll post it. Mods delete if it’s not allowed. This was what’s got me thinking about eye safety. It didn’t have a good outcome but it shows how fast things can go bad.

Small die laser diodes will always be more dangerous than LEDs because of their insane candela/lumen ratio compared to LEDs, especially at long distances. They can do much much much more damage at similar power levels.

Even I’m not keen on the Emisar D4. Its potential for damage, and huge heat generation means it’s not a light I’ll be carrying, ever. I prefer a regulated 1000 lumens from a quad LED setup, for very long runtimes of regulated brightness.

Eye damage? Not that I’ve heard of. I have 210 lights and the majority are heavily modified, folks around here know by now that I can’t help but modify a light for all it’s worth… never heard of anyone actually having damage from a bright flashlight. Sure, if you look right at it from up close it hurts and you look away, but it’s not lingering and causes no “damage” to my knowledge. Sure aren’t as bright as the sun day in and day out… :wink:

Hi Wight, long time no see! :slight_smile:

Adults have to use common sense, Li-ion cells and bright flashlights are not toys for children to play with. Like guns and knives, screwdrivers and wall outlets, common sense must prevail…

No, not eye damage of course.

I was writing about physical damage, like on your skin, your backpack.

And no! Actually, on very close distances, most high power flashlights actually have higher intensity than the sun. This is how I can see my modded BLF Q8 in the brightest part of the day, and when I pull it straight to my face, it gets very hot quickly.

Welcome back! (for a little bit)

The stock D4 has a standby drain of about 0.02 mA. If I recall correctly, it goes into standby about 6 seconds after turning “off”.

When the light is awake but has the main emitters off, like between blinks in beacon mode, it uses about 4.5 mA.

The stock light’s moon mode is higher than that, like maybe 7 mA or so. I forget the exact value, but the moon level is higher than I wanted.

In newer firmware, I reduced the power used while awake. To do this, I made it under-clock the MCU at low levels, and also made it go into an idle state between interrupt events. So by reflashing a stock D4 to the latest Anduril, moon mode drops to about 1.7 mA and the between-beacon-flashes power use drops to like 1.1 mA. Also, it goes into standby a lot sooner after going to “off” mode. In most cases, standby is immediate.

To reduce standby power, you can change the fuses a bit while flashing. Basically, turn off BOD and it’ll drop standby from ~0.024 mA to ~0.002 mA. This change doesn’t really affect battery life much though, and mostly just determines whether the light will reboot after changing the batteries, or if it can go without power for long enough to swap the cells.

I’m not sure what exactly was done to eliminate ringing, but if I understand correctly, it’s a resistor on the FET control pin to soften the edge of the curve a bit. It may be possible to determine how this works by looking at a recent driver from MtnElectronics or Emisar or Texas_Ace, but I don’t really know enough about circuit design to explain it.

The voltage divider can be eliminated entirely on most 4V linear-driven e-switch lights. There is no need to use pin 7 for that any more. Eliminating this reduces standby drain quite a bit compared to older designs, makes PCB layout easier, and frees up a pin for other purposes. The MCU uses VCC on pin 8 to measure voltage.

To use thermal regulation, at the moment, it needs a tiny25/45/85 MCU with a sensor built in. For running Anduril (or other FSM-based interfaces), I’d suggest tiny85. I plan to add support for tiny84/841/1634 sometime too, but I’m not sure when.

One other thing to note is, for e-switch lights, the UI development is now a lot easier than it was in the past. Take a look at the FSM UI toolkit if you have any interest in making or modifying interfaces. This allows people to define interfaces in a much more intuitive way than before, and makes it fairly easy to support a variety of different hardware.

Also, aux LEDs are a popular thing now. They’re blingy like tritium vials, except they don’t wear out and they can be turned off or turned down.

You’ve probably missed other things while you were gone too, but I’m not sure what.

There is no rule against outside links on BLF. :slight_smile:

As for lasers… I have a list of “cool toys which are too dangerous for me to own”, and high-powered lasers are on that list. I could buy or make one, and it would be glorious, but someone would probably get hurt. Many of the high-powered ones are bright enough that even just looking at the hotspot on a wall, without protective goggles, can cause eye damage.

Yes, indeed, Lasers are another story altogether! I have a 3.3Watt Blue laser that can burn a yellow jacket nest from 25-30’ away! It WILL burn a blister on your hand in less time than you can jerk away from it, so the eyes are definitely in danger from something of this power. These should be treated just like a gun, locked away from children and monitored to the same extent. Very dangerous stuff, the high powered laser.

Can’t the thermal level just be set way down so it doesn’t get hot? I really liked to have this kind of power available but it’s indeed to hot. I always unscrew the cap before I pocket it due to the heat. The D4 is also very good looking :stuck_out_tongue:

No, better not. These are not toys. A 500 lm XM-L with a 20mm head diameter is safe though, with protected batts….won’t get too scalding hot also. If you get an AA NiMH XM-L, best. About 200-300 lm.

I nearly got blinded with a HID and 55W bulb, with 75W ballast so that one was overclocked. Actually for that one wasn’t even the hotspot issue, but even the spill is sufficient to screw up your vision for a good couple of hours. The light accidentally fell on its side and my eyes directly caught the spill just some 15-30 cm away. I had a bright spot vision right in the middle of my field of vision for a few hours. Forget about the blink reflex or look away reflex, they don’t help enough.

The 55W bulb just has an arc gap of 3mm and you are doing like 5000 - 6000 lumens easy. Go calculate the surface luminosity.
eg relative to a XHP 35 is 3.5mm x 3.5mm. How many lumens is that, should you even catch the side-spill up close?
Not sure how many lux is the side spill that close though, havent measured that. Don’t get mixed up lux with surface luminosity though, i sometimes mix them up.

I also have a couple of lasers small to big weak to powerful, those are even worse as the laser lights are literally infinitely point source wrt LEDs. 5mW in certain scenarios can cause eye damage. Their low divergence means their beams stay dangerous, so forget about being 10m away from the laser, coz even the reflections might be dangerous.

That point source light would be focused onto the retina, and these point sources burn your retina the same way as a magnifying glass + sun combo.

With a real intense LED high cd hotspot, you’d get an even bigger hole in the retina wrt your eyes catching the spill, if both are pretty up-close.

One thing that also needs to be mentioned, is that laser light is coherent light. It loses very little energy or get “out of focus” (lack of a better word) after passing through the lens, a lot of energy is left to be focused on the retina. LED “non-coherent” white light behaves a bit more differently. But still get enough white light brute force candelas onto the pupils, bad things would happen still no matter how it’s badly focused onto the retina and energy transmission to the tissue cells. :slight_smile:

Oh btw, if anyone wants to know a number, once i had some free time and tried to measure my cheapest and lowest powered 50mW greenie from several tens of metres away, typical aperture size (no beam expander) and no real effort done to focus it accurately at that distance also. It’s works out to be 8 Million cp on the lux meter, seriously not very high. I’m not sure how the green light is being accurately measured by the lux meter, but well it’s just a number. If it so happens that you focused it to be around 1m (for burning purposes) and it feel from the tripod and with sheer luck the beam gets through your pupils, it’s well well above that 8 M cp. Up close, that “8 M cp” is still from an infinite point source of light that the eye lens can focus more or less nicely onto the retina, unlike a big reflector HID or BLF GT hotspot, even if we keep the lux readings on the pupil diameter to be the same for both LED and laser light sources. It’s this point source that significantly makes laser light so much more dangerous even with the same lux, coz of the way it’s focused on the retina to be a point source image.

PS. It’s actually the thermal and photochemical effects that the laser light has on retina tissues. Another issue is that green causes more damage to the retina (red) than red laser light, something like green lasers can pop red balloons and it’d take much higher powered red lasers to pop that same balloon.

Oh yeah, my youngest (5 yr old) has his own Xeno E03. A XM-L 20cm head light. When he’s alone he uses the NiMH. If i am with him outdoors, i’d pop in a protected 14500. Safe enough.

A D4 would be too bright too dangerous up close, shd the power settings be accidentally set. Young kids have short hands, really too close for comfort.

I can’t remember on the other thread, did you get the latest gen Maxabeam or the earlier ones? 12 Mcp or?

I am willing to bet that 12 M cp hotspot + pretty much point source at longer distances with that relatively small 5 inch aperture, if your eyes do get caught within that ten plus metres (not sure when the hotspot would really start to become focused though), is dangerous and has potentially enough lux/cd at that 10+ metres with that 5-inch aperture literally becoming as a point source of light, to cause permanent damage be it UV or visible in times shorter than your blink reflex. Even if not permanent, it’d take a fairly long time to heal (not hours definitely).

Generally, UV and IR radiation causes damage to the cornea/lens, while visible to NIR causes retina tissue damage.

Give these things to a normal person who always jokes around, i am also willing to bet that he or she would point it at someone’s eyes. I already had that happen with my “only” 3.8 M cp HID during a flashlight meet. Never ever hand over a laser pointer to a layman without a safety brief even outdoors where reflection risk is low.

edit - who wants to volunteer to get hit by a Mateminco MT35 Plus’ hotspot at just a few metres where it’s beginning to focus and has the highest lux. :smiley:

Similarly, the new Maxtoch L2K LEP light, with 2000 metres (not sure what is the cd for this, been several years since i did such calculations) and front aperture of just 60mm, is going to be pretty dangerous as well, even though pure cd numbers are not high relative to say a BLF GT. It’s both a function of the cd and also the aperture, that determines how it’s going to be finally focused onto the retina. It’s also possible due to LEP technology that it is already focussed into a tight beam within 1 metre! :cowboy_hat_face:

PS. Sorry i may not be very good in conveying all my thoughts and tehnicalities as English is not my first language, I hope it does not confuse more. :stuck_out_tongue:

I look at atomic blasts all the time and I don’t have a problem. I just watched 10 of them on YouTube and my eyes are fine!

Sorry, I’m trolling since I can’t go fishing today

Welcome back ! too :slight_smile: . Where have you been ?

Sorry, not enough time to tidy up this quoting but:

When we moved on from the Attiny 13 to e.g. the 85 (via the 25) it was found that they became more and more susceptible to the FET/LED current switching spikes, that messed them up.

DEL sorted this out by properly de-coupling the MCU from the main power rail. With a 5R in series, and 100 nan across the MCU.

Also added a pull-down to the FET gate drive to properly define it.

Here is a link to the schematic of the current FET+1, which might work:

https://ucde87931876e80c19701f070b4d.dl.dropboxusercontent.com/cd/0/inline/ASt6Pynxq0sCId3N_TA_erpu9zpJvHAx1BEBGzvaTgDHcQsY6GoeeOgCNJWWFkLq9gf8_ewj2XsOaHmDVzfHbsRriGtfERtQnmpc2AewHOPMABAIR8Kw_Wf35uwnNgmI7TPuHkbOzhJBosF1abW28svMf0K4ZxzF-dZoYSzcvwCqNZC3Gr5pUdnw_8f-DdqGxGp-L0gFjpyWYd04cddqB0BU/file

No voltage divider needed now, we first bumped up the values x10, which helped a lot, but then eliminated it altogether by letting firmware do a sort-of double measurement (TK can explain). Combined with getting rid of the voltage divider, and the OTC (ugh), this means the MCU can now do all the switch timing in firmware. And learning how to use the Attiny’s power saving capabilities.

Many designs now use an e-switch rather than a clickie. But both are valid.

Quiescent current is now so low that you can even change cells without re-booting.

Moving to e-switch drove quiescent current drain development (always on).

There has been a retrograde trend to lighting up switches, tailcaps, aux-led boards etc which has wasted some of these improvements, but some seem to like this sort of flashy nonsense.

It has all become very commercial now, hardware designs are not usually shared (even though they are still mostly derived from your and Comfychair’s pioneering). Commercial contracts are at stake, and any constructive criticism is not usually well received.