Review of Spark SG3 and Spark SG5 Headlamp and clip EDC lights. (14500/AA/16340/CR123)

Ok, let me first thank Spark for sending me these lights for review. To be honest I had no idea I’d be getting these lights. They came 2 weeks after I sent them an email inquiry about reviews, and they said they’d send me a video and then a bike light for review. I sent them my address, and that was it- until these showed up on the door step.

And second here is a full video review of the light. In this video the lights are Frozen, dropped, and the review is fleshed out a bit more. That said, the text review has more technical data. All the abuse tests are later in the video in case you want to skip.

Needless to say I was excited, because, up until this point all I had were 2 cheap headlamps from eBay. Let me also mention, I was a little unprepared for these battery wise. I do have three Trustfire flame 14500s I purchased earlier this year and 4 unused Ultrafire “1000 mah” 16340s that I got for $2.50 at a Wallbuys sale 3 months ago.

Finally a light to use them on! I was going to apologize for not having any CR123As to test out the SG3 on, but as you’ll see later in the review that it really doesn’t matter at this point. I don’t have a lot of spare cash to waste on CR123as right now, and the ones I could afford would take several weeks to get here.

Anyway… so enough right? Onto the reviews.

Since these lights are sort of kindred spirits, they will be combined into one review. Otherwise it’d be two reviews with a lot of copy and pasting.

Basically it goes like this. According to the specs the SG3 (71mm long, 24mm diameter) is a slightly smaller light with a slightly higher output but shorter runtimes. It uses either CR123As or 16340s. The SG5 is a slightly taller (81mm long) and skinnier light that can take: alkalines, lithium primaries, 14500s, and Nimh. They both come with nearly identical sets of accessories. The light, a removable stainless steel clip, a headband, a velcro securing strap, a super nice velvety bag, and some basic instructions, and a nice box with magnet. I think they retail for about $60ish.

What else do these lights have in common? Well other than the dimensions they’re identical cosmetically. Specs say they’re made from a class III hard anodized aluminum alloy. I have no reason to doubt that, as they seem resistant to scratches- after having subjected them to filth, dirt, and rusty nails during the photo shoot in an abandoned house. They were also thrown inside a backpack with a bunch of other lights I brought along on a bike ride to the house.

Rounding out the exterior is a Schott ultra-clear glass lens, and a carbon fiber tube that surrounds the main battery tube. I know it’s subjective, but these lights look pretty nice. They have nice rounded curves, and look almost like some sort of classic military light. The knurling around the tailcap and the lens is aggressive and tactile. I like it! And it should also be noted that Spark makes a lens with reflectors for both of these lights, so if the beam is too floody for you they are swappable. Expect to use a pair of pliers to remove the lens though, because they’re tightened too much for hand removal. And yes, this light is about as floody as you can get.

Ok, so what about the tail cap and the switch? Well the tail cap allows the light to tail stand, but only on a relatively flat surface. If I could suggest an improvement to the Spark design team, maybe make the flat area a bit larger on the next revision to aid in better tail-standing? While the diameters of the tail caps on both lights are slightly different they both tail stand about equally well.

You can keep the clip on or remove it. Unfortunately you can only put it on one way. So if you’re using it as a clip light on an exterior pocket, you’ll have to use it upside down. I chalk this up to a minor annoyance, and there’s not much you can do about it without a redesign of the light. For some, this sort of thing is a deal breaker- not for me though. This clip is serious too, and it takes a bit of forcing to go on the tube. It isn’t cheap, thin, or easily malleable. Assume that taking it on and off repeatedly will leave some wear on the clip depression on the tube. This isn’t a knock on the finish, because it seems the anodizing holds up well to something that would rub off the finish on pretty much any other light.

The switch? It appears identical on both lights. It’s a thick rubber boot with the spark logo. It’s electronic, so there’s no click.

What about inside the light? Spark says in the instructions that it recommends you do not take apart the light. I will agree with this on Spark, as you’ll see later in the review. Complete disassebly of the light is nearly impossible without permanently damaging the electronics. It’s like taking apart an iphone. Anyway, on the negative end of the tube the tail spring looks like it’s gold plated. It’s thick and makes good contact with the batteries, on both lights. On the positive end of the tube, it looks like a flat PCB, so there seems to be no mechanical mechanism employed for reverse polarity.

Well this is what the instructions say verbatim…

“Insert Battery. Insert positive end of battery first, towards head of flashlight. Then tighten the tail cap. Mis-placingof battery will not cause any damage because reverse polarity function is embedded.”

Cool right? More on that later. Keep reading.

Now lets throw a battery in both lights and see how the UI works. Good idea! The user interface is pretty simple and identical on both lights.

Click once to turn it on. Now that it’s on, select a mode by pressing and holding the button. The light will then ramp to each of the 4 main modes, in order from lowest to highest. The are as follows: Min, Med1, Med2, and Max. When it hits the mode you like let go and it’s saved. So when you turn it off and turn it back on it uses that last mode it was on. Nice and simple. If you’d like to change modes… Hold the button on the current mode, and it will revert back to min and ramp through all the modes again. I dig this more than if it were to go to the next mode. In addition to the 4 regular modes there’s one called “Super” You can access this mode if the flashlight is off or on, by doing a steady double click.

The absolute best part? Since the manual doesn’t state it, and because I haven’t inadvertently accessed it, there are NO BLINKIES. I can hardly believe it and I love it.

Ok five modes total. What do they output and how long?

Here are the manufacturer specs for both lights. Unfortunately it does not specify which battery chemistry these outputs and run-times are good for. I am going to assume they’re for Li-ion. The manual says simply this light uses an XM-L emitter. I’m pretty sure it’s an XM-L2 because it looks different than all my other XM-L emitters. And having done the “running water trick” PWM is not detected on any mode.

SG5-CW Manufacturer Specs. (reviewed)
Super- 280 lumen at .9 hours.
Max- 110 lumens at 2.8 hours
Med2- 40 lumens at 9 hours
Med1- 8 lumens at 32 hours
Min- 1 lumen at 10 days

SG5-NW Manufacturer Specs (not reviewed)
Super- 260 lumens at .9 hours
Max- 100 lumens at 2.8 hours
Med2- 30 lumens at 9 hours
Med1- 6 lumens at 32 hours
Min- 1 lumen at 10 days

SG5-CW output on Eneloops.

SG5-CW output on 14500s.

SG3-CW Manufacturer Specs (reviewed)
Super- 380 lumen at .8 hours.
Max- 200 lumens at 1.8 hours
Med2- 70 lumens at 6 hours
Med1- 10 lumens at 30 hours
Min- 1 lumen at 12 days

SG3-NW Manufacturer Specs (not reviewed)
Super- 350 lumen at .8 hours.
Max- 180 lumens at 1.8 hours
Med2- 60 lumens at 6 hours
Med1- 8 lumens at 30 hours
Min- 1 lumen at 12 days

SG3-CW output levels on Ultrafire “1000 mah”

Ok, so I do not have a light meter so I won’t be able to give exact light outputs. But I do have some comparison beam shots of other lights. These first groups of pics have the lights at about 1 foot from a white wall.

Here are the higher outputs on a faster shutter speed. The image is darker to better illustrate output and to highlight the beam patterns.

Settings are the same on all shots. First is the control. The 1st two were taken with the lights about even with the camera lens. Then I moved them closer to the wall to illustrate the beam patterns. The Spark seems easily much brighter. I would estimate Sparks claim of 260 lumens not too far off, if we use the 117 lumen Fenix as a benchmark. I apologize, but beam shots are all I have until I purchase a light meter.

Ok how about the runtimes? Well those I can test, and provide you with visual proof on how long they run and if the output in light is continuous or it tapers off. I present to you my output progression charts. These shots are taken at regular intervals with identical camera settings. Some of the longer runtimes I had to take irregularly as my schedule permits, as there is no way I can take pictures at regular intervals over the course of 12 days or even 30 hours. To accomplish this, I had set up a camera and tripod on a ledge in my stairwell. For this test I just turn on the light and leave it run for at least the manufacturer rated specs. The light is not turned off for any of these tests unless specifically noted.

Within each chart all camera settings stay the same. If I start at at 1.8f, 1/100 sec, 400 iso on the first pic in say High Mode, the last and all inbetween pics on high mode are 1.8f, 1/100, and 400 iso. Sometimes on lower or higher modes I will adjust the camera settings for the entire chart so you can see the difference better. For the extremely technical, sometimes I did not press the button exactly on time- the times posted can vary by up to five minutes but rarely more than or or two. Please note the camera specs on each chart.

Ok one more thing before we get to these charts. You’ll notice that I do not have run times posted for the SG3 currently. Well there is a good reason for that. Remember earlier when I said I’d explain the reverse polarity protection in greater detail? Well I decided to take Spark up on this claim, and threw the battery in backward on the SG3. Prior to this I had made sure the light operated correctly, which you can see in the video, and even took some output beamshots, so what you see is what I had accomplished before I did the test. So I threw in the battery and after screwing down the tail cap it lit up like in turbo mode. I unscrewed it immediately, and just for the heck of it I did it again for a brief second and unscrewed it. Since the manual stated it had reverse polarity protection and I didn’t see mechanical evidence of it, I had assumed it was built in. I guess that’s not the case, because now the light now only operates in two modes… and maybe some sort of Super mode. The thing is, I have no idea which modes it left and which are no longer there. The UI operates identically but now with fewer modes. After doing this I decided it was not a good idea to test this out on the SG5, if I wanted to play it safe. I had no idea if my light was defective and didn’t have this mode or if the information in the manual is incorrect for all the SG3s. Anyway please be very careful when putting a battery in the light. Do not put it in backwards. Double and triple check.

Well after this set-back I decided that I’d conduct additional stress tests for the partially working light, since it’s already sort of bunk. But that’s later on.

Anyway. So what about the runtimes on the SG5? How do they stack up against the manufacturer specs? What chemistries are the runtimes good for? For the sake of time, it’s nearly impossible to check the chemistries of every battery in moonlight because it would take a a month, and considering the test area is our main stairwell, it’s impossible to leave the camera and light set up for weeks at a time. I will make an earnest effort to test the moonlight mode in the near future on the SG5, minus a camera, to see if it lives up to the claim of 10 days- on a 14500. By the way, spolier alert, this light works best on 14500s. Also it’s a very minor thing but I believe all these charts are using 500 iso instead of 400. Check it.

First off lets go with 14500s.

Max

Med2 – this one sort of surprised me, as it came up about 3 hours short. I may try a retest of this mode if I get some time.

Med1. With this test I took 3 breaks. I didn’t want to leave the light going while I was sleeping or working, so I could get accurate runtimes. It actually outperformed the manufacturers specs by about 7 hours.

Min.
No data available for moonlight. Spark claims 10 days.

How about eneloops?

Super.

Max

Med2. Yep, you’re seeing it right. I tested this three times and got about the same result every time. It has less run time than the higher mode, and well less than 9 hours. Stick with 14500s. Even then it seems med2 is the one mode that falls short of manufacturer specs. All other modes have been spot on or better.

As of right now I have no data for med1 and min on AA batteries. I may provide this info in the future. My work schedule currently does not provide time for me to conduct these tests, considering I have to be near the light at all times.

So what about the missing info about the SG3? Well since I don’t expect a replacement sample, I’ll try my darnedest to beat the crap out of this light to see how well it holds up. So that’s good right? Make lemonade out of “No reverse polarity protection.”

So what about the missing info about the SG3? Well since I don’t expect a replacement sample, I’ll try my darnedest to beat the crap out of this light to see how well it holds up. So that’s good right? Make lemonade out of “No reverse polarity protection.” So there won’t be any photos for that segment, and you’ll need to watch the video. And it should be noted- all abrasions on the SG3 that you see in later photos (and maybe a few early ones) all came from those tests. The light DID NOT come that way. You’ll see me freeze, throw, and drop the light a full two stories on the pavement. How will it fare

How nice are the head bands? Reading around there seems to be some complaints. Mainly that the SG5 seems to not fit tight enough in the headband. I agree with this, but the SG3 seems to work alright. So upon closer inspection I’ve found that the head bands that come with both units appear to be identical. Since some of the proportions on the SG3, particularly the ones that the headband touches are thicker, the rubber of the head band grips better. So I’ve found one way that keeps the SG5 from sliding around. Leave the clip on, but swivel it, so it doesn’t come in contact with your head. It is guaranteed to stay in place. Anyway, here are the best ways I’ve found to mount both lamps. The way I have each one mounted works the best for me.

Oh and how about the star? It’s right under/at 16mm, if you’re looking to swap it. Here are a few pics of the interior of the SG3. Expect similar for the SG5.

While I don’t think there’s many people who would attempt to mod a $60 headlamp short of a simple emitter swap (which should be fairly easy), I would not recommend it. For the most part this is machined out of a single piece of aluminum. Save for the tail cap, switch and driver. This is good news for thermal management. Because the entire light is basically a heat sink. Behind the star is solid aluminum. There’s a tiny hole behind it that allows the wires to run from the switch to the driver. I found it impossible to remove the now dead driver. Please do not attempt to remove the driver, or you’ll risk bricking the light, much like mine. I thought I could try to remove it and attempt remedy the reverse polarity damage (I’m not sure how), and it killed my light.

So that’s it. I might add additional data for med1 runtimes on AA batteries as my schedule permits.

A quick summary of what I like.

  • Awesome regulation. I am assuming the SG3 is pretty much identical in regulation to the SG5. The sustained brightness times on 14500s is awesome. I could detect no step-downs. AA is no slouch either, but this light shines on 14500s.
  • No PWM. A $60 shouldn’t have detectable PWM anyway right?
  • Flexibility of battery chemistries. I like that you have the abilty to use lithium ions.
  • Excellent runtimes.
  • Super easy lockout. Try as I might I wasn’t able to get a reading on parasitic drain. It has to be small, but my multimeter wasn’t able to detect it.
  • Great thermal management. The whole body of the light makes direct contact with the star. It’s a solid piece of aluminum.
  • Awesome anodizing and build quality. Save for the killer clip area, the SG5 (which did not drop 18 feet onto concrete like the SG3) has no signs of wear after a week of carrying and tests.

What could be improved.

  • How about the light being clipable rightside up in pockets? It’s kind of a minor thing for me (just clip it upside down) But I’d like the light to have the button on the top if I’m clipping it.
  • The fit of the SG5 in it’s headlamp apparatus.
  • Real REVERSE POLARITY PROTECTION. Seriously do not attempt this. The manual says it has it, my real world experience says it doesn’t. Be cautious when inserting the battery.

Anyway that’s it for now! Thanks for looking, I spent many hours on this. And if you’d like to see many many more photos that didn’t make it into this set check out my flickr for the SG3 and SG5.

http://www.flickr.com/photos/mhanlen1/sets/72157635199475527/

Thanks for the review!

I see Spark is still pretty bad about posting their output specs.

HEY! Watch it buddy! I take offense to the use of that word! Just kidding. Thanks for the review. Like the photos!

-Garry

Yep. The manual isn’t super helpful either. It’s very basic. But that’s what reviews are for!

A-ha! so you found one of the hidden BLF member insults I have peppered throughout the review!

Nice review and pictures. Thanks :slight_smile:

Ok folks, here is the full video review much earlier than I anticipated having it done. I am going on vacation later this week, so I needed to crack down on my day off and finish it. All the stress test stuff is later in the video. The light is frozen, thrown at objects, and dropped from a two story window.

Nice video review!

Tank you, love your video reviews

Thanks for reviewing the SG5 mhanlen, i was wondering whether or not the runtimes were anywhere near what they claimed.

love that video review :slight_smile: very nice and funny at the same time

Love the video! Love the use of the abandoned house for backdrop too. My wife did ask me "What the $&%# are you watching?" I think because I was laughing at a video about flashlights!

Are you a comedian for your "day job"?

-Garry

No, but I write a separate script for the video. I revise it over a few days. The first version has a few or no jokes, then I add some more… then I take out a few. I feel if I didn’t add a few jokes these would be unwatchable by anyone but 3 of the most diehard flashlight fanatics- especially clocking in at an excruciating 15 minutes. I literally spent all day on this yesterday. The abandoned house is located in a local state park, right in the middle. You either have to hike to it or ride a bike. I figured it would make a nice backdrop for a review. I don’t like all my pictures and videos from different reviews to look the same. So I read my script into my computer final cut pro… screw up a lot- and edit out all my flubs.

I got these lights last Monday. Not too bad! The side affect is the girlfriend complains I’m not paying attention to her.

“What are you doing on your computer, you’ve been there all evening!”

“Umm, working on this thing.”

Fantastic review. Thanks. Enjoy the holiday.

If it were any other reviewer I would probably have passed up watching the video, but since it was an mhanlen production I knew it was going to be good! This is the honest truth. You do good work. I know you must have hours and hours into this! Now excuse me, I have to go check my freezer and see if I have any misplaced items there.

-Garry

It happens! Thanks again for the kind words- it is my hope that they’re entertaining… and also I try to put info in there for non light people to be able to understand the videos.

Anyway onto some more info. So I was able to finally run the light on Med2 mode with eneloops. I got some very bizarre behavior. Seems the light gets a little warmer than usual on med2 AND it runs significantly less than on Max mode. However the output is sustained… and it just shuts off… sort of like when using a 14500. Anyway, I thought that this couldn’t be right, but after running three separate tests (with 3 different eneloop 3rd gens) yesterday and the light always just shutting of roughly around the 2 and a half hour mark- I guess it’s not a fluke.

I am currently going back and testing the 14500 on med2 and then the eneloop on Max, and see if they hit the same targets as I have posted above. I said it before… but this light works best on 14500s. Anyway here is the chart… it’s also been integrated into the main review. Now I’m really curious what med1 mode is like. I don’t know if I’ll get to it this week, but when I do I’ll post my findings.

Is it just me or is the M2 mode with an eneloop trying to act like the max mode on an eneloop.

I agree. I am going to be re-trying a few of the other tests… med2 on 14500 and eneloop with max, to make sure my light is behaving correctly. I’ll report back.

Edit: 8/28/13 - I tried an Energizer industrial this morning, and it seemed to have behaved the same on Med2 as the Eneloop but with less runtime. So it’s definitely not the batteries.

Edit: 8/29/13 I’ve been going back and trying 14500s on med2 and I’m now getting about 3.5 hours. I barely got an hour out of a newer duraloop today on Max. Seems like the runtimes aren’t getting better.

Seems like it’s a programming error then.
I haven’t seen any others reporting this yet, so i guess it’s a wait and see problem.
Thanks for your efforts in finding out the problems.

We’ll see. I’m on vacation right now, but I’ll be doing some more runtime tests when I get time in a week or two. Otherwise it works fine- it does get pretty warm on alkalines though.

Well an owner of the SG5 over on CPF reports that he squeezed 9 hours out of an eneloop on med 2 mode. Which means my 2 hour runtime, is probably due to a defective unit. So as long as you buy the light through a retailer, if you run into this problem you can exchange it. Or maybe mine is just the lone quirky one.

Read it here.

Or here is the quote.