Supfire M6 - resistor mod (Update testing XP-G2s on copper in post 59-60)

133 posts / 0 new
Last post
zelee
zelee's picture
Offline
Last seen: 21 hours 20 min ago
Joined: 06/04/2013 - 06:53
Posts: 1975
Location: Bali, Indonesia

Ouchyfoot wrote:
Werner wrote:
You guys are crazy…
So before I even would think of a higher power, I would definitely try to build a handle for this.

I already know that the handle from my Roche M170 will fit, and pretty sure the one from my EYE40 will too.

glad i’m still keep my M170 handle :bigsmile:
texaspyro
Offline
Last seen: 2 years 6 months ago
Joined: 04/29/2011 - 12:43
Posts: 4593

I put one of my custom 9 amp SRK drivers in one of mine. It starts into the thermal management after a minute or so. An SRK takes about 3 minutes. The driver is measuring the temperature in the driver compartment, not on the flashlight body. The body was only around 10C above ambient when the thermal management kicked in. The thermal management in my driver keeps the light below around 55C. After 30 minutes or so, it has reduced the light to around 900 lumens (around 10 watts) At that point, the temperature has fully stabilized.

Pulsar
Pulsar's picture
Offline
Last seen: 4 months 3 weeks ago
Joined: 07/29/2011 - 00:41
Posts: 5848
Location: Maine

how does that compare to a SRK host? 

Pavithra_uk
Pavithra_uk's picture
Offline
Last seen: 1 week 4 days ago
Joined: 04/24/2012 - 22:47
Posts: 779
Location: SRI LANKA
RaceR86 wrote:

That would depending on how much current you wanted.

When I added 2x R120, i got about 3,2A.

2x R120 should be the same as one R060.

If you add one R056 (the same value as stock on the driver circuit). That would probably give you around 3,3A.

Driver chip is AMC7136. its Current sense feedback voltage is 50mV

Your stock driver setting: 3× 0.056 ohms = 0.018 ohms

Stock drive current = 50mV / 0.018 = 2777 mA

After 2×0.120 ohm resistor added, 0.018 & 0.060 = 0.013 ohm

now drive current will be 50/0.013 = 3846 mA

texaspyro
Offline
Last seen: 2 years 6 months ago
Joined: 04/29/2011 - 12:43
Posts: 4593
Pulsar wrote:

how does that compare to a SRK host? 

The hosts handle the same amount of heat and stabilize at around the same temp/light level. The M6 driver compartment heats a lot fast than the SRK (maybe because the black body radiates heat back into it better).

_the_
_the_'s picture
Offline
Last seen: 2 hours 28 min ago
Joined: 07/08/2011 - 06:22
Posts: 3647
Location: Finland

Nice mod. Thanks for sharing!

Here's my quick calculations of your mod and a hypothetical mod with 1..5 pieces of 0.1 Ohm resistors:

 RaceR86 mod  Planned mod 
Stock resistor 10.056Ohm 0.056Ohm
Stock resistor 20.056Ohm 0.056Ohm
Stock resistor 30.056Ohm 0.056Ohm
Extra resistor 10.120Ohm 0.100Ohm
Extra resistor 20.120Ohm 0.100Ohm
Extra resistor 30.120Ohm 0.100Ohm
Extra resistor 4   0.100Ohm
Extra resistor 5   0.100Ohm
      
Resistance     
Stock0.01867Ohm 0.01867Ohm
w. 1 extra resistor0.01615Ohm 0.01573Ohm
w. 2 extra resistors0.01424Ohm 0.01359Ohm
w. 3 extra resistors0.01273Ohm 0.01197Ohm
w. 4 extra resistors Ohm 0.01069Ohm
w. 5 extra resistors Ohm 0.00966Ohm
      
Feedback voltage (mV)45.5Calculated back from the measured 2.4A emitter current - takes driver losses etc. into account..   
      
Emitter current     
Stock2.44A / emitter 2.44A / emitter
w. 1 extra resistor2.82A / emitter 2.89A / emitter
w. 2 extra resistors3.20A / emitter 3.35A / emitter
w. 3 extra resistors3.58A / emitter 3.80A / emitter
w. 4 extra resistors A / emitter 4.26A / emitter
w. 5 extra resistors A / emitter 4.71A / emitter

=the=

 

ryansoh3
ryansoh3's picture
Offline
Last seen: 1 month 1 week ago
Joined: 08/04/2012 - 08:21
Posts: 3893
Location: US

That’s pretty awesome.

Thanks for the calculations!

BLF ≠ B-grade Flashlight Forum

 

dchomak
Offline
Last seen: 3 years 6 months ago
Joined: 03/17/2012 - 20:10
Posts: 4122
Location: Connecticut

Am I right that the upper limit to this mod is, in effect, direct drive? After all I believe that the cells are in parallel as are the 3 emitters. This is good as there is less danger in wildly over driving the emitter.
The TR-3T6 resistor mod had 2 or 3 cells in series. If I had this light, I wouldn’t hesitate to do this mod.

RaceR86
RaceR86's picture
Offline
Last seen: 5 years 10 months ago
Joined: 02/22/2012 - 08:32
Posts: 3777
Location: Norway

Nice calculations _the_.

Should be good guidelines. Smile

dchomak wrote:
Am I right that the upper limit to this mod is, in effect, direct drive? After all I believe that the cells are in parallel as are the 3 emitters. This is good as there is less danger in wildly over driving the emitter. The TR-3T6 resistor mod had 2 or 3 cells in series. If I had this light, I wouldn't hesitate to do this mod.

Most likely yes. We can not know how far the resistor mod will work until someone tries it though. Might be limits. But as can be seen, medium and low becomes higher too. About 3,6A to each emitter is already at the stupid amount of heat for this light IMO. It will be very useful in the winter though. Not only does it function as a light, but it will be a great hand warmer too! Smile

4 cells in parallel will be able to kill 3 XM-L mounted on aluminium in DD, so DD is not a good idea on stock emitters. Put in XM-L2s mounted on copper and you should be able to output the amount of current you desire, if the driver will do that. But there is really not much point in going beyond the 3,5-3,6A range IMO. Not only due to vf of XM-L2s and the less great medium and low, but mainly due to heat. 

BLF LED database – collaboration spreadsheet and latest news about where to buy LEDs
http://budgetlightforum.com/node/19342

Ouchyfoot
Ouchyfoot's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/01/2012 - 06:15
Posts: 5089
Location: Canada

I’m loving this thread. I’ve been seeing resistor modding in the forums a lot lately, but most of the chatter was between modders with electronics backgrounds. This thread is truly helping me understand what is going on. The before and after pictures by RaceR86 gave me immediate focus of the what and how, and each additional post has helped me with the why. Now I’m looking at all my lights a little differently. I can’t wait to get started.

_the_
_the_'s picture
Offline
Last seen: 2 hours 28 min ago
Joined: 07/08/2011 - 06:22
Posts: 3647
Location: Finland

One of the original emitters failed during the experiment (not sure why), so I don't have directly comparable results, but here we go:

2 x 0.1 ohm resistors did boost the emitter current to about 3.3-3.4A (with XM-L2 3C on Noctigon), quite well in line with my calculations.

Output jumped up to about 3100 lm (OTF, after 30s)

3 x 0.1 ohm resistors gave another 100 lm, up to about 3200 lm of nice neutral white light. Cool

Light
Approximate min lumens
Approximate max lumens
SupFire M6-T6X3 XM-L2 3C (3x0.1ohm mod)3903,196
SupFire M6-T6X3 XM-L2 3C (2x0.1ohm mod)3813,102
BTU Shocker NW XM-L2 Mod523,079
BTU Shocker CW522,820
Solarstorm Warrior XM-L2 T6 1A942,397
Skyray King CW2352,298
Trustfire TR-J123382,195
SupFire M6-T6X33012,190
BTU Shocker NW472,181
Skyray King NW2302,096
Solarstorm Warrior XM-L U2 1A852,077
APEX 5T6 NW1501,941
Blackshadow Darth NW1691,532

Ps. Throw is now about 49kcd (440m)

Edit: Fresh cells -> 3309lm (OTF, after 30s)

=the=

 

RaceR86
RaceR86's picture
Offline
Last seen: 5 years 10 months ago
Joined: 02/22/2012 - 08:32
Posts: 3777
Location: Norway

_the_ that is sweet, and your calculations seemed good too.

Based on the resistor values, and amps, we seemed to get very similar results, so that is always nice to see. Smile

Are you one of the people who have ordered AR coated lens from CNQ? Would be nice to see your lumen comparison on your modded light with AR lens vs stock.

(I know many others have those lenses coming, and some will probably share, but the more numbers we get, the better)

 

Edit: did you measure current with 3 x 0.1 ohm resistors?

BLF LED database – collaboration spreadsheet and latest news about where to buy LEDs
http://budgetlightforum.com/node/19342

_the_
_the_'s picture
Offline
Last seen: 2 hours 28 min ago
Joined: 07/08/2011 - 06:22
Posts: 3647
Location: Finland

RaceR86 wrote:

Are you one of the people who have ordered AR coated lens from CNQ?

Not yet.. But I think I need to order some.. Wink

 

RaceR86 wrote:

Edit: did you measure current with 3 x 0.1 ohm resistors?

Didn't have time for unsoldering the emitter leads again for measurements. Sorry. Might do this later (I know, I should buy a clamp meter)

=the=

 

RaceR86
RaceR86's picture
Offline
Last seen: 5 years 10 months ago
Joined: 02/22/2012 - 08:32
Posts: 3777
Location: Norway

I have a feeling you will order at least one pretty soon. Im still waiting, I want to see some numbers before I pull the trigger.

If you find the time, this day or another, please check emitter current with 3 x 0.1 resistors. Considering that you only saw a 94 lumen gain going from 2x 0.1 to 3x 0.1.  Im wondering if you are not seeing much of that calculated output (3,8A) due to vf and such of the copper mounted XM-L2s.

BLF LED database – collaboration spreadsheet and latest news about where to buy LEDs
http://budgetlightforum.com/node/19342

RaceR86
RaceR86's picture
Offline
Last seen: 5 years 10 months ago
Joined: 02/22/2012 - 08:32
Posts: 3777
Location: Norway

Some may be familiar with my other SRKs (SRK/UV-S5), both modded. A little comparison here.

I tested both of those vs my current Supfire M6 (not supbeam as wrote on one picture). Mouse-overs below.

Camera settings:

1,6s, f5, iso400, 17mm (aps-c), WB was 5000K on first comparison, and 5200K on the second.

What other members have noted, or showed before. The M6 have less flood, and a more defined hotspot with slightly more throw. Otherwise, not a huge difference compared to my other modded king. In real life, there is no doubt M6 appears  brighter due to the more harsh light. Output difference is not that much though. Due to tint,CRI and such, id take the SRK easily. Much nicer in real life. But its impressive to see the output of the M6 with stock emitters (and modded driver circuit).

When it comes to throw, I compared it to my SRK with de-domed emitters and a much lower current. The treeline in the distance are 120 meters away. The de-domed SRK wins fairly easily, not a surprise really. Due to not having that much bright spill, I would say its a better thrower in real life than in the picture. Lots of bright spill and flood kinda limits the ability to see things in a distance. The M6 outputs a lot more light though(!), and still have decent throw despite running close to 3,6A to XM-Ls on aluminium. And when XM-Ls are on aluminium, gain above 3A is not that much. 

BLF LED database – collaboration spreadsheet and latest news about where to buy LEDs
http://budgetlightforum.com/node/19342

zelee
zelee's picture
Offline
Last seen: 21 hours 20 min ago
Joined: 06/04/2013 - 06:53
Posts: 1975
Location: Bali, Indonesia
RaceR86 wrote:

Some may be familiar with my other SRKs (SRK/UV-S5), both modded. A little comparison here.

I tested both of those vs my current Supfire M6 (not supbeam as wrote on one picture). Mouse-overs below.

Camera settings:

1,6s, f5, iso400, 17mm (aps-c), WB was 5000K on first comparison, and 5200K on the second.

What other members have noted, or showed before. The M6 have less flood, and a more defined hotspot with slightly more throw. Otherwise, not a huge difference compared to my other modded king. In real life, there is no doubt M6 appears  brighter due to the more harsh light. Output difference is not that much though. Due to tint,CRI and such, id take the SRK easily. Much nicer in real life. But its impressive to see the output of the M6 with stock emitters (and modded driver circuit).

When it comes to throw, I compared it to my SRK with de-domed emitters and a much lower current. The treeline in the distance are 120 meters away. The de-domed SRK wins fairly easily, not a surprise really. Due to not having that much bright spill, I would say its a better thrower in real life than in the picture. Lots of bright spill and flood kinda limits the ability to see things in a distance. The M6 outputs a lot more light though(!), and still have decent throw despite running close to 3,6A to XM-Ls on aluminium. And when XM-Ls are on aluminium, gain above 3A is not that much. 

it seems there is some mistake on 2nd picture RaceR,it supbeam instead of supfire :bigsmile:

RaceR86
RaceR86's picture
Offline
Last seen: 5 years 10 months ago
Joined: 02/22/2012 - 08:32
Posts: 3777
Location: Norway

Yupp, you are correct. And if you read the 2nd line I also said so. You even quoted me on it.

BLF LED database – collaboration spreadsheet and latest news about where to buy LEDs
http://budgetlightforum.com/node/19342

_the_
_the_'s picture
Offline
Last seen: 2 hours 28 min ago
Joined: 07/08/2011 - 06:22
Posts: 3647
Location: Finland

RaceR86 wrote:

I have a feeling you will order at least one pretty soon.

Good guess. Ordered some. Smile

 

RaceR86 wrote:

If you find the time, this day or another, please check emitter current with 3 x 0.1 resistors. Considering that you only saw a 94 lumen gain going from 2x 0.1 to 3x 0.1.  Im wondering if you are not seeing much of that calculated output (3,8A) due to vf and such of the copper mounted XM-L2s.

I'm not totally sure about this (had some problems measuring), but I got readings of 3.5-3.7A. I think you are correct about the vf and such.

 

Ps. Nice beamshots!

=the=

 

ImA4Wheelr
Offline
Last seen: 4 months 1 week ago
Joined: 02/03/2013 - 14:51
Posts: 7935
Location: SC

There is a formula that one can use to calculate current if the voltage drop across the resistors is known. It can be obtained by putting the DMM leads on each side of one of the resisters. The formula is:

Vdo/[1/{(1/R1)+(1/R2)+(1/R3)..+(1/Rn)}]

Please excuse me if I’m using the wrong types of brackets above. Just trying to make the formula easier to follow. I don’t know if Vdo (voltage drop out) is the correct term either. I don’t have any formal electronics training.

Pulsar
Pulsar's picture
Offline
Last seen: 4 months 3 weeks ago
Joined: 07/29/2011 - 00:41
Posts: 5848
Location: Maine

amazing results _the_  

so the m6 is actually brighter than both of your BTUs? how fast does it heat up?? thats crazy 

_the_
_the_'s picture
Offline
Last seen: 2 hours 28 min ago
Joined: 07/08/2011 - 06:22
Posts: 3647
Location: Finland

Pulsar wrote:

amazing results _the_  

so the m6 is actually brighter than both of your BTUs? how fast does it heat up?? thats crazy 

Thanks!

Yes, more lumens than in either of my BTUs. Much less throw though. Wink

Heats up pretty fast, but I haven't tested longer than 1min bursts yet..

 

BTW / OT: These resistor mods are addictive.. I continued by adding 2 x 0.1 Ohm to my de-domed XP-G2 TN31 -> 440kcd Cool

=the=

 

Pulsar
Pulsar's picture
Offline
Last seen: 4 months 3 weeks ago
Joined: 07/29/2011 - 00:41
Posts: 5848
Location: Maine

how much more output would you expect from a AR lens? just 6% more light, and it will be over 3500 lumens

_the_
_the_'s picture
Offline
Last seen: 2 hours 28 min ago
Joined: 07/08/2011 - 06:22
Posts: 3647
Location: Finland

Pulsar wrote:

how much more output would you expect from a AR lens? just 6% more light, and it will be over 3500 lumens

The stock lens is actually quite clear, so my estimate is about 4%.

=the=

 

Pulsar
Pulsar's picture
Offline
Last seen: 4 months 3 weeks ago
Joined: 07/29/2011 - 00:41
Posts: 5848
Location: Maine

either way, still must be very impressive the amount of light in such a small package... really cant wait to get mine

Ouchyfoot
Ouchyfoot's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/01/2012 - 06:15
Posts: 5089
Location: Canada

RaceR86. I haven’t recieved my M6 yet, and am wondering how easy or difficult it is to remove the driver from this light? Does it need any special technique?

texaspyro
Offline
Last seen: 2 years 6 months ago
Joined: 04/29/2011 - 12:43
Posts: 4593
Ouchyfoot wrote:
RaceR86. I haven’t recieved my M6 yet, and am wondering how easy or difficult it is to remove the driver from this light? Does it need any special technique?

It’s trivial… unscrew retaining ring, remove driver.

RaceR86
RaceR86's picture
Offline
Last seen: 5 years 10 months ago
Joined: 02/22/2012 - 08:32
Posts: 3777
Location: Norway

texaspyro wrote:
Ouchyfoot wrote:
RaceR86. I haven't recieved my M6 yet, and am wondering how easy or difficult it is to remove the driver from this light? Does it need any special technique?
It's trivial... unscrew retaining ring, remove driver.

+1 Does not get better.

No damage or wear and tear to the driver. I just use the same tweezers I use for all flashlights since they can be bent out quite far. Needle nose pliers or something similar can be used too of they are narrow on the tip.

BLF LED database – collaboration spreadsheet and latest news about where to buy LEDs
http://budgetlightforum.com/node/19342

Ouchyfoot
Ouchyfoot's picture
Offline
Last seen: 10 months 2 weeks ago
Joined: 09/01/2012 - 06:15
Posts: 5089
Location: Canada

Great. This light is looking like a modders dream…even beginners. I wanted to see if something similar could be done to my Warrior, but the driver looks like a struggle to get out without damage.

RaceR86
RaceR86's picture
Offline
Last seen: 5 years 10 months ago
Joined: 02/22/2012 - 08:32
Posts: 3777
Location: Norway

Let the battle of the kings continue..

I put on some XP-G2 R5 3Cs on copper. XM-L centering rings were connected with fujick. I put some kapton tape on the back of the reflector (short circuit saftey). Solder connections were low.

The MCPCBs were a bit loose so I only checked emitter current for 5 sec. Results:

3,55A... (!!??) What, that high, on  XP-G2?? That looks promising. Did not expect that..  Smile

2,61A.. what, that low, that was lower than anticipated..

 2,71A.. Low on that on too..

That is 8,87amps total, or an average of 2,95. NCR18650B @ 4,05V+ were used. They were not freshly charged and slightly used. I double checked the 3,55A and got the same number twice. Dont ask me...

 

Supfire M6 XP-G2 R5 3C on copper @ 2,95 average vs SRK  XM-L2 T4 5B1 @ 3,09 on copper

Sure it out-throw the XM-L, but not by much. The XM-L light "throws" a much more wider beam, and it puts out a lot more light. To me, the XP-G2s were more of a let down in real life than in the pictures. 

Supfire M6 XP-G2 R5 3C on copper @ 2,95 average vs SRK  XM-L CW-dedomed @ 2,3A on copper

The de-domed XM-L at lower current (2,3A) kills the XP-G2 light at 3A. No contest...

For this particular test, I had rings and one ugly ring in the XP-G2 beam. So question is, would there be a significant improvement if the reflector sat lower?

After I took the beamshots I removed the XM-L centering rings, did another attempt at getting the soldered wires even lower. That should make the reflector sit marginally lower. Result was pretty much the same. I know I could make the reflector sit lower by using some pieces of brass plate on the MCPCBs instead of wires, or just modify the underside of the reflector. Instead of messing around with the M6 reflector I changed to SRK reflector since that was easier.

Why?

M6 reflector on the left, SRK on the right. As you can see, the place where the wires on a 16mm mcpcb sits crashed with the flat surface on the M6 reflector . Changing the stock MCPCB with 16 mm stars is no issue when using XM-Ls. But XP-Gs need to sit further inside the reflector. That is easier to achieve with the SRK reflector..

BTW, kapton tape is not a good insulator as it can easily be broken, I just sometimes use it instead of a proper insulation pad/plastic thingy when height is an issue. 

Supfire M6 with SRK reflector. The reflector sits directly on the XP-G2s mcpcb now. Focus is much better and no ugly ring..

Temporary I planned on putting the M6 reflector on the SRK, but due to the MCPCBs sitting further outwards in that reflector I would have to cut about 1 mm of the MCPCBs when used in a  SRK in order to get them further out to the edge.

 

To be continued..

PS: Warning once you think you have cycled through all the modes, you point your eyes towards the reflector in order to check something.. "poff, and there goes you vision. D%¤# SOS! 

Beer

BLF LED database – collaboration spreadsheet and latest news about where to buy LEDs
http://budgetlightforum.com/node/19342

RaceR86
RaceR86's picture
Offline
Last seen: 5 years 10 months ago
Joined: 02/22/2012 - 08:32
Posts: 3777
Location: Norway

...continued..

 

Supfire M6 XP-G2 R5 3C on copper with SRK reflector  @ 2,95 average vs SRK  XM-L2 T4 5B1 @ 3,09 on copper

Same lights, longer distance:

I think the reflector change made a huge difference in real life. Much nicer beam, no rings or anything. I think the comparison is interesting too. We are basically looking at XP-G2 vs XM-L2 in similar reflector at quite similar current.

De-domed XM-L would still win when it comes to throw, but the beam is very useful with the XP-G2. Ill keep this light like this and see how I like it once I have used it more.

I have a feeling I will stick to my beloved XM-Ls, like I do in most lights. But I have not tried de-domed XP-G2s....

To be continued.. some day...


BLF LED database – collaboration spreadsheet and latest news about where to buy LEDs
http://budgetlightforum.com/node/19342

Pages