Review: XTAR S1 (3 x XM-L U2, magnetic ring selector, 30 - 2800 lumens ANSI)

70 posts / 0 new
Last post
xed888
Offline
Last seen: 2 months 3 weeks ago
Joined: 07/16/2011 - 12:22
Posts: 771
Location: BN

Thanks 2100. Can always count on you.

I have a TK70 now but Im quite tempted by this S1. Do you reckon its worth selling the TK70 for this? Although I do hate the PWM of S1 (its such a good torch, it shouldnt have PWM! )

2100
Offline
Last seen: 2 years 6 months ago
Joined: 05/28/2011 - 10:28
Posts: 4512
Location: SINGAPORE

The TK70 is a good flashlight, i am using it in 3D config with 2 x unprotected 32600.  Hence there there is always a small risk even though i keep that at bay with regligious voltage checking.  4 x NiMH is too unwieldy for normal daily use.  It's not the weight but the very high polar inertia, you cannot swing it around easily.

Both put out the same amount of light OTF (you can't notice the diff, trust me).  As for throw, it is 60k vs 90k cd or 33.3% diff, so it is just a little visibly cut in throw. I have always maintained that you cannot tell any difference if it's < 20%.  As you can see in the pictures, there really is just a wee bit of difference and only noticeable when you directly A-B them side by side. 

The Xtar S1 is definitely much easier to use and bring around, even in a small haversack stuffed with many stuff like food/water. It is not a EDC even with a holster though.

 

I am current talking to Xtar HK to see if they can bump up the PWM.  As for efficiency of this vs current controlled, well it is a triple XM-L and 3 x 18650...how much runtime do people want? Triple XM-Ls are already very efficient even at 1000 lumens.  LOL!

 

xed888
Offline
Last seen: 2 months 3 weeks ago
Joined: 07/16/2011 - 12:22
Posts: 771
Location: BN

Thanks! PWM for me is about it annoying me. I can detect it quite easily and its rather distracting. Unless they remove it from their final model, i think this isnt the light for me Sad and i did want to like it very much!

2100
Offline
Last seen: 2 years 6 months ago
Joined: 05/28/2011 - 10:28
Posts: 4512
Location: SINGAPORE

I told them to see if 1-2kHz region is ok to do (that's the iTP flashlights frequency).  They have some reason not to do it at 20kHz, not sure why but i guess for engineering they'd have their reasons. 

But seriously 475Hz is not that bad, it is very usable.  This is 3X faster than most budget lights.

My personal preference and priority is usually to use the light more and as often as possible. We are living in cities, we already don't have much chance to use it.   Hence I am ok to accept some negatives.  Seriously I do use the TK70 quite a lot (several thousand hours), but most of the time it is tailstanded as a night light in the lowest mode.  Even in the house i don't manage to bring it around often.   The Xtar S1 is definitely not small but it's still ok to tailstand in the toilet and move around etc.... I conbsider that as the biggest size to use.

Eg, The XTAR is as easy to use as your Olight M3X, SWM T40CS, just 2X heavier.  If you hold your TK70 in your left hand and a Olight M3X in your right and you move around, the difference is telling.

xed888
Offline
Last seen: 2 months 3 weeks ago
Joined: 07/16/2011 - 12:22
Posts: 771
Location: BN

I think I will hold off on the S1 at the moment. USD200 for a light with PWM is no-go for me, I'm afraid. If they remove, I might think twice. And like you said, the TK70 is just an awesome light!

joe1512
Offline
Last seen: 2 years 4 weeks ago
Joined: 11/03/2011 - 15:31
Posts: 331
Location: Dallas,TX

Where do they get the 2800 lumen figure from?

If you are reading 6.2 amps, thats 2.1 amps per emitter which is what?  600ish lumens each?

The TK70 does 3 amps per emitter which is more like 850 lumens per emitter.

2100
Offline
Last seen: 2 years 6 months ago
Joined: 05/28/2011 - 10:28
Posts: 4512
Location: SINGAPORE

Joe, 

(1) The TK70 does not do 3 amps per emitter.  As far as my knowledge goes only forummers claimed that.  It's probably pretty close, and anyway it's not worth to get into a debate with the other folks regarding this.  I dwell more on measuring with the lux meter.

(2) This is U2, if you use the xls LED calculator and key in 320lm @ 700mA, you'd get 816 emitter lumens for 2.1A.  You get 899 lumens for 2.8A drive for a T6.

http://budgetlightforum.com/node/2808  <<-  calculator

(3) Just measured again by cleaning the MKNE IMR cell's surface and pressing harder, actually i was able to get 6.5A-6.6A peak. Unfortunately it's not really easy to measure.

(4)  Important thing is, if you look above, TK70 reflected bounce is ~ 220 lux.  Xtar S1 is ~ 210 with 3 x 18650.  I was able to get this ~ 210 with the MKNE IMR as well with the tailcap screwed on.  That is what is important and is accurate.   The current reading is for reference only.

RedForest UK
RedForest UK's picture
Offline
Last seen: 1 year 5 months ago
Joined: 06/06/2011 - 11:27
Posts: 1055

Yeah, the final measurements for output are what really matters, everything else is really just working back from that!

 

Remember that a good high capacity IMR 26650 even under load would be well over the forward voltage of the XM-L LEDs (maybe 3.9v under load fully charged with the XM-L vf of around 3.3v @2.3-2.5 amps) with the full calculation you get:

6.6 x 3.9 = 25.74

/100 x 90 (assume 90% efficient circuit) = 23.166

/3 (3 XM-Ls) = 7.722

/3.3 (each 3.3v vf) = 2.34 amps to each emitter.

 

According to the spreadsheet that means nearly 900 lumens for each U2, which would make almost the claimed amount at the emitter(s ).

2100
Offline
Last seen: 2 years 6 months ago
Joined: 05/28/2011 - 10:28
Posts: 4512
Location: SINGAPORE

hehe.... Redforest UK's calculations are deep....i understand part of it only at first glance.

Anyway Xtar S1 is 475Hz PWM for my copy (see if production copies are different), that's not too bad and very usable.  Just received my 5 x XM-L Trustfire TR-J12, it's 124Hz.  Now that's bad!   

how2
Offline
Last seen: 6 hours 6 min ago
Joined: 06/01/2010 - 10:26
Posts: 2425
Location: london

It's got very poor regulation compare to the Nitecore and Olight triple XML lights.

is this the most expensive DD light?

2100
Offline
Last seen: 2 years 6 months ago
Joined: 05/28/2011 - 10:28
Posts: 4512
Location: SINGAPORE

how2,

It's not DD aka Direct Drive.  I think Selfbuilt chose the wrong word.  If you apply his definition of DD, then each and every Ultrafire, Trustfire, Uniquefire is DD. 

I measured only 6 amps plus at the tail.  It should be approx 15A (good cells) at the tail if it were to be DD.  heh....  Also with Direct Drive, it would never be able to do > 30 mins run, let alone 76 mins till the knee drops.

 

Selfbuilt has provided lots of data in CPF, hence i don't feel like it's nice to mess up his review thread just due to a bit of "semantics". 

 

It certainly is not boosted as the current falls as it's discharged, because the 3 cells are in parallel. TM11 managed to pull it off though. Not the same experience as those with cells in series like TK70, TR-J12, Uniquefire 3900 and the likes.

 

You might wanna check out his graphs.   400 -> 300 in 60 mins.  Drop of 25%.  Can  you see a difference between 75% and 100%, yes you can....try a thrower light that has 40k lux and 30k lux at a distant target.  But in real life usage without A/B, no way.

TM11

asd

XTAR S1

asd

how2
Offline
Last seen: 6 hours 6 min ago
Joined: 06/01/2010 - 10:26
Posts: 2425
Location: london

 

Seems unregulated compared to the other lights.

2100
Offline
Last seen: 2 years 6 months ago
Joined: 05/28/2011 - 10:28
Posts: 4512
Location: SINGAPORE

It is not regulated in a straight line, definitely.  So it is not 100% perfect, that i'd agree.  In fact many 2 x 18650 lights like Fandyfire STL-V6 has a flatter line regulation.   A lot of 1 x 18650 lights like even the Jetbeam RRT-2 XM-L cannot maintain a flat line regulation till the end like the multi-cell lights. All 1 x 18650 light sucks?  Sealed  I just want to put some perspective into things.  Maybe i may seem to be defending the light because I got a "review copy", but allow me to present my case.

This so-called lack of regulation is the least of my worries.  Though the nearly 500Hz PWM is unobtrusive and very usable, it is still mildly visible and i guess to the purist flashaholics that is "not nearly enough" because Olight/Fenix/Jetbeam/Nitecore can do it anyway (which is true). The PWM needs to be pushed up to at least 1kHz....heck while they are at it just push it up to 5kHz (and be done with it) if they can't drive it by varying the current like the Sunwaymans.  Big Smile

This is the difference in throw output at a distant object, the difference is ~ 33.3%.

60k cd

asd

 

90k cd

asd

 

2100
Offline
Last seen: 2 years 6 months ago
Joined: 05/28/2011 - 10:28
Posts: 4512
Location: SINGAPORE

As long as we are talking about higher draw lights like in the 2.5A range (corresponding lumens in the > 600L range), it proves to be very difficult to do with single cell.   Take for example, a Lumintop TD15X.  Check out the 1 x 18650 curve.  (this is Selfbuilt's work)

sdf

And this is the Thrunite TN12 and gang

sdf

 

how2
Offline
Last seen: 6 hours 6 min ago
Joined: 06/01/2010 - 10:26
Posts: 2425
Location: london

2100 how much did you pay for this light?

They are trying to sell this light for $175.

2100
Offline
Last seen: 2 years 6 months ago
Joined: 05/28/2011 - 10:28
Posts: 4512
Location: SINGAPORE

usd175 is delivered i guess, direct from XTAR.  I am not supposed to say, but of course it's easy to guess that i paid a slightly lower price (and taking a caculated risk as well as being an early adopter or rather tester!).  I don't even have lanyard or O-rings! 

The other dealers have not really got stock yet, i have asked.

how2
Offline
Last seen: 6 hours 6 min ago
Joined: 06/01/2010 - 10:26
Posts: 2425
Location: london

It's better to get the Olight and Nitecore or Finex as they all seem to be better regulated.

2100 does the lens have a reflective coating?

how2
Offline
Last seen: 6 hours 6 min ago
Joined: 06/01/2010 - 10:26
Posts: 2425
Location: london

The Xtar has a very inefficent driver as you can see from selfbuilt graphs.

The Xtar produces more light for the first 16mins, then the SR92  whops it's butt from then on.

Run time for Xtar on Max 1hour and 26mins

Run time for SR92 on 1hour 51mins

how2
Offline
Last seen: 6 hours 6 min ago
Joined: 06/01/2010 - 10:26
Posts: 2425
Location: london

2100 why can you not say how much you paid for it?

Many reviewer get there lights for Free and they state this, some just say evaluation copy or review copy.

I'm a straight talker as most of you lot know. I said the Dry was ugly. I have no problems when people review lights that they get for free, I will always take the reviews with a pinch of salt.

What I do not like is when people are not clear whether they got it free or not.

Come on 2100 be a man and tell us.

Chicago X
Chicago X's picture
Offline
Last seen: 7 years 10 months ago
Joined: 07/22/2011 - 16:13
Posts: 4013
Location: See Name

Your challenge of 2100 is unnecessary and unwarranted. 

He has contributed real data to this site, not just a bunch of "that torch sux" comments. 

http://wardogsmakingithome.org/index.html

War Dogs, Making it Home - Rescue Dogs for Returning Vets

xed888
Offline
Last seen: 2 months 3 weeks ago
Joined: 07/16/2011 - 12:22
Posts: 771
Location: BN

Chicago X wrote:

Your challenge of 2100 is unnecessary and unwarranted. 

He has contributed real data to this site, not just a bunch of "that torch sux" comments. 

Agreed.

xed888
Offline
Last seen: 2 months 3 weeks ago
Joined: 07/16/2011 - 12:22
Posts: 771
Location: BN

2100 wrote:

usd175 is delivered i guess, direct from XTAR.  I am not supposed to say, but of course it's easy to guess that i paid a slightly lower price (and taking a caculated risk as well as being an early adopter or rather tester!).  I don't even have lanyard or O-rings! 

The other dealers have not really got stock yet, i have asked.

I think it's safe to assume he paid for it and wasn't given a free sample by Xtar.

RedForest UK
RedForest UK's picture
Offline
Last seen: 1 year 5 months ago
Joined: 06/06/2011 - 11:27
Posts: 1055

I thought the SR92 battery pack contained 6x 18650 cells, this only uses 3 so it's hardly a fair comparison. I do agree however that the lack of regulation is not ideal and is a deal breaker for me, along with the PWM. Not that I need any more lights at the moment or in fact ever.. (I'm sure this won't stop me in the future though!)

 

A little known fact is that PWM actually uses current, (not only as it involves bursting the LED at a higher current and so doesn't take advantage of higher efficiency at lower current inputs) but each switch from on/off of the LED uses a tiny amount of electricity, this means that lower freq PWM is actually higher efficiency then high frequency, and is why zebralight originally used it in L2 mode of the H51 etc as it was more efficient to use low freq PWM than current control as the current to the LED would have been below optimal efficiency level. I'm not sure if this consideration comes into question when manufacturers are deciding what frequency to set PWM but it's an interesting fact.

ILIKEFLASHLIGHTS
ILIKEFLASHLIGHTS's picture
Offline
Last seen: 2 years 8 months ago
Joined: 08/04/2011 - 23:47
Posts: 5693
Location: southeast MO

I for one think it's one hell of a light.  I even told Rick that last week on the phone.  But I also can now say they won't be selling very many to US buyers for $175 shipped to the US.  For that kind of money, I can buy a brand new Fenix TK70 for $164 shipped to my door. 

how2
Offline
Last seen: 6 hours 6 min ago
Joined: 06/01/2010 - 10:26
Posts: 2425
Location: london

Don't worry about 2100. He is a big bad boy, he can handle it.

So can I.Why did he mention single 18650 lights? In response to my question?

That's what is cannot understand. He was not as objective as selfbuilt?

Personally I have never liked selfbuit's reviews as I did not understand it fully but as I got to know more about lights, his reviews have made sense.

2100
Offline
Last seen: 2 years 6 months ago
Joined: 05/28/2011 - 10:28
Posts: 4512
Location: SINGAPORE

I brought in 1 x 18650 sincel-cell lights for comparison because those are just single-cell lights, ie 3.7-4.2V discharge range.  This Xtar S1 is a 4.2V light, just that it is like 9AH in capacity with 3 cells in parallel. But then there are also 3 LEDs and 3 individual drivers.  This is more of a technical/engineering topic rather than a semantics "how many cells the light's got"....   My understanding is that it is not easy to boost a light of this nature to maintain a pretty flat output all the way till the cell is near 100% discharged (driver input voltage is below emitter's Vf). A cell could be at 3.0V under load and you still want to get 700 lumens from it. I am not a driver expert, maybe those into electronics can chime in.

Don't worry how2/xed8888/chicago X....i am not mad or anything, and anyway it's natural to suspect that i got a free light.  The light was sent to me on promise that i don't reveal what i paid because that'd affect the on-sale price....but i certainly did not pay a budget light price.  Still comfortably above 100 bucks.  LOL!

 

Remember, i got the light and and shared measurements before I could see any reports at BLF and CPF (except the folks at shoudian), ie i dived in 100% blind.  And such things are definitely not for fun and laughter, if you know what i mean.  I have actually asked like 5 vendors for availability.

 

I am not sure if i understand the word direct drive correctly.  With DD that it means the light is directly driven by the cell right?  It could be direct hardwire, it could go through a driver's electronics but basically in DD mode the driver does not limit nothing. This is my understanding. 

 But if there is a resistor in there or a driver which limits the current, then it is not direct drive though it could be unregulated (seperate matter).  This is because in the CPF reviews, the direct drive is mentioned multiple times.  Maybe it's coz i have been doing playing with too many DRY 3 x XM-L, now that's DD.  heh...  (ok so the latest driver is still 90% PWM in Turbo).  Perhaps i am wrong and the correct info is something that i can learn too.

 

Here (it's the same as place as the XTAR 18700 group buy) :

asd

 

 

2100
Offline
Last seen: 2 years 6 months ago
Joined: 05/28/2011 - 10:28
Posts: 4512
Location: SINGAPORE

Look bros...it is funny that the word objectivity should even be mentioned.  I thought real-world scenarios sharing is good. LOL!

Ok...I presented all the beamshots + measurements and that is objective data.  But I need to be "subjective" as well because in the real world that is important too.  It's the same in the Hifi/Photography/car mods arena.   Specs and measurements tell you that part of the story only.  As a professional photographer of many years (Chicago X too), we only know that too well.  Same goes for car mods, which X dabbles in as well as for income.  LOL!  Sealed  I am not saying that it's not important, numbers is the lifeline of engineers.

What is the diff of 20% out? It's 20%...nothing more to that.  If you want throw, you'd need heck lots more.  If you have lights that surpass 0.5-1 mil cd you probably can appreciate this more....else it'd always be in that XR-E/XM-L 50-60mm head diameter kind of performance and range and say that hey this XM-L thrower throws very well.  I'm just trying to bring this sense of "scale" into the big picture....

 

PS. I have experience (in events) in which we've got several billions of candlepower comfortably from nearly 50 advertising search lights each of > 5kW (it was a mix), and it threw 10km, but barely only. Of course they were trying to light up the clouds (the wispy ones, cirrus). That's real world sharing for you.... and that's the truth of searchlights in the real world.  LED Search & Rescue lights often advertise that they can do 700m, 1km.  Smile

HKJ
HKJ's picture
Offline
Last seen: 5 days 3 hours ago
Joined: 05/24/2011 - 12:23
Posts: 7377
Location: Copenhagen, Denmark

2100 wrote:
I am not sure if i understand the word direct drive correctly.  With DD that it means the light is directly driven by the cell right?  It could be direct hardwire, it could go through a driver's electronics but basically in DD mode the driver does not limit nothing. This is my understanding. 

 But if there is a resistor in there or a driver which limits the current, then it is not direct drive though it could be unregulated (seperate matter).  This is because in the CPF reviews, the direct drive is mentioned multiple times.  Maybe it's coz i have been doing playing with too many DRY 3 x XM-L, now that's DD.  heh...  (ok so the latest driver is still 90% PWM in Turbo).  Perhaps i am wrong and the correct info is something that i can learn too.

The light is definitely not direct drive, it is more like a 7135 driver as can be seen on the voltage sweeps curves in my review.

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

2100
Offline
Last seen: 2 years 6 months ago
Joined: 05/28/2011 - 10:28
Posts: 4512
Location: SINGAPORE

HKJ wrote:

The light is definitely not direct drive, it is more like a 7135 driver as can be seen on the voltage sweeps curves in my review.

I see, thanks!  heh... i really ought to do more study into drivers.  (electronics is not my forte)

2100
Offline
Last seen: 2 years 6 months ago
Joined: 05/28/2011 - 10:28
Posts: 4512
Location: SINGAPORE

xed888 wrote:

 

I think it's safe to assume he paid for it and wasn't given a free sample by Xtar.

Only saw your reply till now. heh...  Actually in the first post, i already posted this : This is a review copy which I paid for it but the retail copy/packaging and pricing should be firmed up after Chinese New Year (23st Jan). 

Note Review and Retail.   I don't want to use Engineering set, which i have encountered many from Blackberry/Nokia/Samsung/etc as a telco worker....hehe

 

 

Pages