Skilhunt E3A (AAA twisty) - First Look

163 posts / 0 new
Last post
iamlucky13
Offline
Last seen: 2 hours 27 min ago
Joined: 06/22/2018 - 09:18
Posts: 1041
Location: USA

If you look at the graph from zeroair, he stopped the test at 2.5 lumens.

That’s clearly far dimmer than the regulated output. The battery is listed at 0.82v at that time.

That’s not optimal for an NiMH battery, but I’m pretty sure it won’t seriously harm it. You’ll get fewer cycles out if it, but Eneloops are so robust, even a 90% reduction in cycle life would mean several years of weekly charges.

Lightbringer
Lightbringer's picture
Offline
Last seen: 5 hours 47 min ago
Joined: 08/30/2016 - 14:12
Posts: 15882
Location: nyc

Yeah, but it’s rock-solid flat flat flat flat flat output ‘til you actually watch it dim down to nothing in probably only seconds. There’s no “100lm” at startup and then a gradual constant fade as the cell wears. I like that.

Using a LFP cell in my (Xeno) E03, I get about the same thing: constant flat output for almost the entire SOC of the cell, and then a noticeable dimming, letting me know I’ve got about seconds of light left. Not a problem because I only use it around the house, so it’s not like being stranded on a dark trail with no other light.

Hmm, so I wonder if that’s because of the cell’s flat voltage, or if it’s the same with even an alkaleak…

09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0

gchart
gchart's picture
Offline
Last seen: 8 hours 35 min ago
Joined: 03/19/2016 - 11:57
Posts: 3173
Location: Central IL

I think I’m fresh out of alkalines to test with, but I imagine it’d look vaguely similar to the NiMH runtime curve – nice and flat. That should be a function of the boost driver providing a constant current.

hodor
hodor's picture
Offline
Last seen: 10 hours 48 min ago
Joined: 07/11/2018 - 03:58
Posts: 1064
Location: UK

Nice teardown and photos! Did you ever get round to playing with the current set resistor?

gchart
gchart's picture
Offline
Last seen: 8 hours 35 min ago
Joined: 03/19/2016 - 11:57
Posts: 3173
Location: Central IL

No, I forgot about it in the midst of other mods I’ve been working on. I might have to give it a shot. Are you thinking higher or lower output? Seems crazy, but I had thought about dropping the output to around 3 lumens or so just got fun (and maybe to use as a reading light).

Maybe I’ll try a runtime as is and then one higher and one lower. We’ll see!

hodor
hodor's picture
Offline
Last seen: 10 hours 48 min ago
Joined: 07/11/2018 - 03:58
Posts: 1064
Location: UK

I’m thinking the same as you! I don’t have one yet though, bought too much recently Blushing

gchart
gchart's picture
Offline
Last seen: 8 hours 35 min ago
Joined: 03/19/2016 - 11:57
Posts: 3173
Location: Central IL

Ok, I just found an alkaline and I’m doing a runtime with it now while I top up some NiMH. Thumbs Up

Unheard
Unheard's picture
Offline
Last seen: 1 hour 52 min ago
Joined: 01/16/2019 - 11:38
Posts: 2279
Location: Germany

SammysHP wrote:
Yeah, stabilization and runtime are great, but it will destroy your eneloops because you won’t notice they are empty until they are discharged too far. Or what is the voltage when the output starts dropping? Still above 0.9 V?

Is there a proof that Eneloops can be destroyed by discharging in a flashlight? I assumed that was the case until I got into a discussion with someone doubting it. I have no idea. Never got one below 0.8 V. They suffered, but I’m not sure if that’s not just because of their mileage.

Smile, you cannot kill them all.

iamlucky13
Offline
Last seen: 2 hours 27 min ago
Joined: 06/22/2018 - 09:18
Posts: 1041
Location: USA
gchart wrote:
I think I’m fresh out of alkalines to test with, but I imagine it’d look vaguely similar to the NiMH runtime curve – nice and flat. That should be a function of the boost driver providing a constant current.

Since alkalines can’t maintain voltage under load, I’m sure it will fall out of regulation much sooner. An NiMH AAA can handle the ~700mA load from this light without a problem, but an alkaline gets pushed to its limits and wastes a lot of its energy content in heat at that load. It also will be at high risk of leaking after such a hard drain.

Here’s a test of the similar Olight i3e EOS – excellent regulation on NiMH, but only 7 or 8 minutes before it falls out of regulation on an alkaline:
https://budgetlightforum.com/node/44513

On the other hand, alkalines have pretty good capacity at low loads, so multi-mode lights or low output lights like the Fenix E01 can stay regulated for a very long time when producing only a few lumens.

gchart
gchart's picture
Offline
Last seen: 8 hours 35 min ago
Joined: 03/19/2016 - 11:57
Posts: 3173
Location: Central IL

Yup, you’re right. I’ll post up the runtimes later but the alkaline is anything but flat.

iamlucky13
Offline
Last seen: 2 hours 27 min ago
Joined: 06/22/2018 - 09:18
Posts: 1041
Location: USA
Unheard wrote:
SammysHP wrote:
Yeah, stabilization and runtime are great, but it will destroy your eneloops because you won’t notice they are empty until they are discharged too far. Or what is the voltage when the output starts dropping? Still above 0.9 V?
Is there a proof that Eneloops can be destroyed by discharging in a flashlight? I assumed that was the case until I got into a discussion with someone doubting it. I have no idea. Never got one below 0.8 V. They suffered, but I’m not sure if that’s not just because of their mileage.

AA Cycler did a test on this. Leaving a flashlight on overnight repeatedly resulted in twice the rate of capacity loss and internal resistance increase compared to discharging to 0.9V, which is normally defined as empty for an NiMH cell.
http://aacycler.com/post/deep-discharging-nimh-lsd-cells/

I wish he’d checked final voltage of the test battery periodically to get an idea exactly how low it was getting, but I’m guessing around the ~0.6V that many boost circuits seem to stop working at.

Note that NiMH cycle life is rated by manufacturers per the IEC61951 test cycle. This is honestly a rather gentle test that only discharges by about 60% each cycle at a 4 hour rate, which will probably still be above 1.1V. If you look at AA cycler’s tests, he tests to 0.9V and at 500mA for a AA, which is roughly a 1.5 hour rate. Under that test cycle, he sees a few hundred cycles typically, compared to Eneloop’s 2000+ cycle rating.

If you go low enough, you will start to cause rapid damage to the battery. Energizer suggests (page 9 here) that at somewhere around 0V, the cell starts to suffer from irreversible internal reactions. Also, if NiMH cells are being used in series, the cell to reach empty first can still have current driven through it by the other cells, causing a voltage reversal that is even more damaging. Because of this, it is not uncommon to find batteries that have been used in sets in camera flashes, solar lights, etc, where the weakest cell suffered reversal and is almost completely dead, while the other batteries work fine.

In my opinion, the bottom line is still to charge the battery when it starts to noticeably dim, and not worry about it. You will still be getting far better performance, face less risk of leakage, and spend less money than feeding your light alkalines.

jon_slider
jon_slider's picture
Offline
Last seen: 7 hours 14 min ago
Joined: 09/08/2015 - 12:20
Posts: 5226
Location: Central North America
iamlucky13 wrote:
If you go low enough, you will start to cause rapid damage to the battery. Energizer suggests… that at somewhere around 0V, the cell starts to suffer from irreversible internal reactions. … In my opinion, the bottom line is still to charge the battery when it starts to noticeably dim, and not worry about it.

I agree, when the light gets dim, change the battery

there is No reason to attempt to draw an Eneloop down to 0 volts, and I dont think it is possible in a flashlight, since they stop working around 0.8v anyway.

There was a report recently, that a Sofirn C01 can draw the cell down to 0.4v, IF the light is left ON.. but if the light gets turned off, it wont turn back on unless voltage is at least 0.8v..

I see no reason to abuse an eneloop this way. If you really want to run a light at super dim, relatively useless brightness, just cause you like to count (uselessly dim sublumen) runtime.. then use an Alkaleak..

ChibiM
ChibiM's picture
Offline
Last seen: 1 day 17 min ago
Joined: 05/09/2011 - 10:25
Posts: 6775
Location: Holland

I once had a meeting with a few Eneloop engineers (at the Panasonic office in Belgium) , and they said that deep discharge is worse than overcharge. 

Here are the questions and answers: https://eneloop101.com/interview-with-eneloop-technicians/

Especially Question #16 onward

Unheard
Unheard's picture
Offline
Last seen: 1 hour 52 min ago
Joined: 01/16/2019 - 11:38
Posts: 2279
Location: Germany

Hey, thanks, Lucky!

iamlucky13 wrote:

In my opinion, the bottom line is still to charge the battery when it starts to noticeably dim, and not worry about it. You will still be getting far better performance, face less risk of leakage, and spend less money than feeding your light alkalines.

Yeah, that’s what I do. I’ve never had enough light output when the cells approached 0.8V and so was forced to recharge them. So, I’ll continue to not worry about it.

Thanks to you, too, ChibiM!

Beer

Smile, you cannot kill them all.

jon_slider
jon_slider's picture
Offline
Last seen: 7 hours 14 min ago
Joined: 09/08/2015 - 12:20
Posts: 5226
Location: Central North America

ChibiM wrote:
Especially Question #16 onward
thanks for the interesting story
fwiw, #16 says:
There is no real testing data to prove this point of how much % cycle life at certain voltage is lost etc. but a deep discharge will reduce the cycle life, this can be by 1%, but can be more as well.

so, just FUD and nothing specific

.

Unheard wrote:
I’ve never had enough light output when the cells approached 0.8V

now thats Real World Smile

ChibiM
ChibiM's picture
Offline
Last seen: 1 day 17 min ago
Joined: 05/09/2011 - 10:25
Posts: 6775
Location: Holland

Wink They didn't back up their claims, but I trust they know what they were talking about Big Smile 

iamlucky13
Offline
Last seen: 2 hours 27 min ago
Joined: 06/22/2018 - 09:18
Posts: 1041
Location: USA

Thanks ChibiM! I looked on your website before my long post above for any additional information, but I missed finding the interview. I don’t think it is linked in any of your menus.

I think what they noted about overdischarge is still consistent with what is in the Energizer datasheet. The Eneloop engineers state there should be minimal difference between 0.8V and 0.9V, but clearly at some point, overdischarge does become an acute problem.

Interesting, the IEC61951 cycle test actually over-charges the cell every single cycle (0.25C discharge for 140 minutes followed by 0.25C charge for 190 minutes), so there is evidence right there that the batteries are tolerant of slight overcharge. I guess this is why they’ve continued for so long to sell kits that include cheap dumb chargers, even though they know those are not optimal.

jon_slider
jon_slider's picture
Offline
Last seen: 7 hours 14 min ago
Joined: 09/08/2015 - 12:20
Posts: 5226
Location: Central North America
ChibiM wrote:
I trust they know what they were talking about

then no reason to worry
1% is a very small number

iamlucky13
Offline
Last seen: 2 hours 27 min ago
Joined: 06/22/2018 - 09:18
Posts: 1041
Location: USA
jon_slider wrote:
There was a report recently, that a Sofirn C01 can draw the cell down to 0.4v, IF the light is left ON.. but if the light gets turned off, it wont turn back on unless voltage is at least 0.8v..

I forgot about that. There’s probably not many lights that can drain a battery as low as the C01, and even that light dim significantly when it gets that low:
https://zeroair.org/wp-content/uploads/2019/01/zeroair_reviews_sofirn_c0...

jon_slider
jon_slider's picture
Offline
Last seen: 7 hours 14 min ago
Joined: 09/08/2015 - 12:20
Posts: 5226
Location: Central North America

> even that light dim significantly when it gets that low

exactly..
Fenix called that 10 hours of sunlight and 11 hours of “moonlight”
LOL!

SammysHP
SammysHP's picture
Online
Last seen: 5 min 38 sec ago
Joined: 06/25/2019 - 14:35
Posts: 1035
Location: Germany
Unheard wrote:
Is there a proof that Eneloops can be destroyed by discharging in a flashlight? I assumed that was the case until I got into a discussion with someone doubting it. I have no idea. Never got one below 0.8 V. They suffered, but I’m not sure if that’s not just because of their mileage.

I think I got this info mostly from steidlmick at TLF. Maybe parametrek as well. From my experience I got one down to 0.8 V and it had lower capacity afterwards (but not sure if there was another factor).

roostre
Offline
Last seen: 7 hours 21 min ago
Joined: 03/09/2020 - 23:37
Posts: 279
Location: Pac. NW

My wife has lots of decorative string light arrangements and fake candles that automatically light up for eight hours a day which use either two or three AA sized batteries configured in series with no low voltage protection.

We are using NiMH rechargeable batteries and when the lights stop working (besides having very low voltages) usually one of the batteries in series from the "three-battery" lights has also temporarily reversed its voltage.

I have recharged these NiMH batteries about twenty times already without any visually noticeable adverse issues so it has saved us money over using non-rechargeable Alkaline batteries (which unfortunately occasionally leaked) and I am curious to see how long these NiMH batteries will ultimately last under these conditions.  Being an old guy, I am no longer worried about my NiMH batteries not lasting longer than me and hope to outlast them all.

Please let me know if there are any safety issues that I may not be aware of from recharging NiMH batteries that have reversed their voltage.


Also, I purchased two Skilhunt E3A lights (one regular and one high CRI to compare) and while I prefer the compact form factor and exceptional build quality of the E3A, I use my Sofirn C01S light more because of its additional low brightness mode which is especially useful as a passenger in a car or while inside a dwelling after other's bedtime.

How many flashlights does a "real man" need?

None, real men are not afraid of the dark.

gchart
gchart's picture
Offline
Last seen: 8 hours 35 min ago
Joined: 03/19/2016 - 11:57
Posts: 3173
Location: Central IL

Well, test #1: I reduced the current sense resistance from 0.43 Ohms to 0.30 Ohms. Output increased 42% but overall efficiency took a crap. Once I do a few more tests I’ll post up some graphs.

Next up: do a couple steps of output reduction

JaredM
JaredM's picture
Offline
Last seen: 1 hour 57 min ago
Joined: 10/31/2011 - 13:33
Posts: 2388
Location: Pittsburgh, Pennsylvania

gchart wrote:
Well, test #1: I reduced the current sense resistance from 0.43 Ohms to 0.30 Ohms. Output increased 42% but overall efficiency took a crap. Once I do a few more tests I’ll post up some graphs.

Next up: do a couple steps of output reduction

Very interested in reduction data. 50% would be nice I think for this light.

gchart
gchart's picture
Offline
Last seen: 8 hours 35 min ago
Joined: 03/19/2016 - 11:57
Posts: 3173
Location: Central IL

Ok. I just started a roughly 50% run. Thumbs Up

SammysHP
SammysHP's picture
Online
Last seen: 5 min 38 sec ago
Joined: 06/25/2019 - 14:35
Posts: 1035
Location: Germany

Awesome. What size is that resistor? Looks tiny and close to the boost IC. Not sure if I want to swap that myself. :/

gchart
gchart's picture
Offline
Last seen: 8 hours 35 min ago
Joined: 03/19/2016 - 11:57
Posts: 3173
Location: Central IL

It’s a 0603 but I managed to fit a 0805 in there. I have a bigger variety of those. I don’t have any 0603 on hand lower than 620 Ohms.

This ~50% run is using a 1.0 Ohm resistor and is chugging along at 43 lumens. I’ll do at least another run at around 5 lumens. Any other special requests?

Lightbringer
Lightbringer's picture
Offline
Last seen: 5 hours 47 min ago
Joined: 08/30/2016 - 14:12
Posts: 15882
Location: nyc
gchart wrote:
Well, test #1: I reduced the current sense resistance from 0.43 Ohms to 0.30 Ohms. Output increased 42% but overall efficiency took a crap.

Yeh, that’s par for the course with those boost converters. It’s unreal how they need to be “fine-tuned” for a particular output. Coil inductance and resistance and size are huge factors. Beyond a certain total current in a specific component, they start to saturate the core. Lotsa heat and current, but no extra output. Like overcooking an LED ‘til it turns blue with too much current.

09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0

JaredM
JaredM's picture
Offline
Last seen: 1 hour 57 min ago
Joined: 10/31/2011 - 13:33
Posts: 2388
Location: Pittsburgh, Pennsylvania

I’d be curious if and where the bottom falls out on the low current end. Either in stability, ripple, or efficiency. 1-5 lumens might be something to shoot for.. For data gathering purposes. Innocent

gchart
gchart's picture
Offline
Last seen: 8 hours 35 min ago
Joined: 03/19/2016 - 11:57
Posts: 3173
Location: Central IL

Before putting together the graphs, here are a few numbers using NiMH running to 10%:

  • Stock with 0.43 Ohm: 93 “maukka lumens” for 77 minutes
  • Replaced resistor with 1 Ohm: 42 “maukka lumens” for 210 minutes!

Pages