You can also check the SST-20 reviews done by the forum testers, captainkirk: Test of a Luminus SST-20-W 3000K 95CRI led (djozz), Luminus SST-20-W 4000 K CRI95 color and output test (maukka) and Luminus SST-20-W 6500 K CRI70 color and output test (maukka); with peaks at ≈6.15A, ≈6.8A and ≈6.7A, and dieing soon after these figures in every test but the 6500K one as maukka didn't want to kill such emitter.

So yes, up to 5A can be considered a great turbo mode. In any case, due to the high emitter Vf achieving such current with a linear or even plain MOSFET driver will only be possible under optimal conditions (high battery voltage, low component resistances and etc.).

P.S.: this or that, these are max 5A linear drivers.

Thanks very much for the advice, I may order two just in case I fumble…

Kirk

Molek wrote:

It is normal that S21B XhP50 dont have spring from driver side?
https://imgur.com/Wa4n20i

I have sst40 and there is this spring but in XHP – i have 2 of this and there is no spring

I wonder which of Convoy’s flashlights now use “button” on the head (driver) side instead of “spring”.

I have the S21B with SST40 and S21B with CULPM1 (6Amp driver), both use a spring on the head/driver side.

I have the Convoy L21A with XHP35-Hi, and it uses a spring on the head/driver side.
But on the Convoy L21A with CULPM1, it uses a button on the head/driver side.

Simon Mao wrote:

latkovicn wrote:

Hey Simon!

What is the plan with SBT90.2? Don’t leave us in the dark! What host you will use?

SBT90.2 will be installed in L6, 4×18A, 3×21A,
but Currently the drivers are still under design

Simon Mao wrote:

JaredM wrote:

Simon has already told us that he’s planning a 3×21700 SBT90.2 version of the 4×18a. I suspect the L21a and/or L6 might get it eventually as well, since many will ask for it.

@Simon

I do have a request for product improvement. I have two of the latest versions of the M2, and while they are great in so many ways, there are a few things I’d love to see in the future..

Supply ‘pan head’ screws in place of the flat head type for clamping the mcpcb. The original setup forced the LED off center when tightened down and made it impossible to get a good beam. I swapped in some pan head screws and was finally able to clamp the board and have it centered at the same time.

Also, I’d love to see a forward clicky switch and maybe a rubber tactical grip ring for the light. The lack of knurling really makes it a slippery light, especially when combined with the deeply recessed switch that forces your thumb to bend tightly. I might be a minority here but I love the sharp knurling on the C8+ and would like to see more of it, as well as forward clicky options in these larger thrower/tactical lights. Maybe we get an M2+ ?

This needs to be redesigned, maybe I need to set up a series of tactical flashlights.

MFW

Simon Mao wrote:

JaredM wrote:

Simon has already told us that he’s planning a 3×21700 SBT90.2 version of the 4×18a. I suspect the L21a and/or L6 might get it eventually as well, since many will ask for it.

@Simon

I do have a request for product improvement. I have two of the latest versions of the M2, and while they are great in so many ways, there are a few things I’d love to see in the future..

Supply ‘pan head’ screws in place of the flat head type for clamping the mcpcb. The original setup forced the LED off center when tightened down and made it impossible to get a good beam. I swapped in some pan head screws and was finally able to clamp the board and have it centered at the same time.

Also, I’d love to see a forward clicky switch and maybe a rubber tactical grip ring for the light. The lack of knurling really makes it a slippery light, especially when combined with the deeply recessed switch that forces your thumb to bend tightly. I might be a minority here but I love the sharp knurling on the C8+ and would like to see more of it, as well as forward clicky options in these larger thrower/tactical lights. Maybe we get an M2+ ?

This needs to be redesigned, maybe I need to set up a series of tactical flashlights.

I would love to see a forward click option for all S21 and M21 series lights.

Caleb wrote:

I would love to see a forward click option for all S21 and M21 series lights.

Do these flashlights use the standard omten switch or maybe the square one?

Caleb wrote:

Simon Mao wrote:

JaredM wrote:

Simon has already told us that he’s planning a 3×21700 SBT90.2 version of the 4×18a. I suspect the L21a and/or L6 might get it eventually as well, since many will ask for it.

@Simon

I do have a request for product improvement. I have two of the latest versions of the M2, and while they are great in so many ways, there are a few things I’d love to see in the future..

Supply ‘pan head’ screws in place of the flat head type for clamping the mcpcb. The original setup forced the LED off center when tightened down and made it impossible to get a good beam. I swapped in some pan head screws and was finally able to clamp the board and have it centered at the same time.

Also, I’d love to see a forward clicky switch and maybe a rubber tactical grip ring for the light. The lack of knurling really makes it a slippery light, especially when combined with the deeply recessed switch that forces your thumb to bend tightly. I might be a minority here but I love the sharp knurling on the C8+ and would like to see more of it, as well as forward clicky options in these larger thrower/tactical lights. Maybe we get an M2+ ?

This needs to be redesigned, maybe I need to set up a series of tactical flashlights.

I would love to see a forward click option for all S21 and M21 series lights.

I suppose it got confusing whether people like reverse or forward clicky..

The earlier batches of Convoy M21A (SST40) and Convoy S11 (XHP50.2) used a forward-clicky tailswitch (I still have these 2 models with forward-clicky version).
I have since then tried another Convoy S11 (SST40) and this S11 now uses a reverse-clicky tailswitch. (I could interchange the tailswitch of the other S11 to make one forward-clicky or reverse-clicky type.)

I’d like to change my M21A (SST40) from forward-clicky to reverse-clicky, but am not sure if it’s available (the reverse-clicky tailswitch that fits the M21A).
I suppose I’m more used to using reverse-clicky type tailswitch flashlights.

Can you share a photo of forward clicky switches of M21A & S11?

For multi-mode lights, reverse switch makes more sense. And for 95%+ of people it does as well. I understand why it’s the standard. There is however a distinct market and application for forward switches..

Ideally,a forward “tactical” switch should have a secondary method to change modes. Either with head twists or an e switch that only functions as mode switching.

Convoy used to make the L4, which is a form factor very close to what I desire.

If the grip ring was removed from the tailcap and offered as a removable rubber version, and it had a more ‘proud’ tail switch, then the rest is mostly cosmetic preferences.

On that point, less is more, and I rreaally miss the styling of old eagletac, nitecore, 4sevens, and even fenix (original tk11) tactical lights. Sleek, aggressive, attractive, and functional.

Lastly, if a 32mm – 44mm – 60mm head diameter series with deep, quality reflectors and thin bezels (maximizing reflector area per head diameter — like the C8 and M2 exceed at (m21c not so much)) and taking cues from Amutorch designs that keep overall length in check, we could have a winner in my book.

Problem is, you make the bezel too thin (especially Al and not SS), and any good solid bang can bust the glass and ruin the thread. A thicker bezel, especially made of SS, makes the front end of the light almost a hammer-head in ruggedness.

Even in a padded case, I accidentally dropped my camera maybe 2’ (if that) onto the sidewalk, and the oblique-angle hit just the right way not only shattered the clear (“UV”) protective filter, but also bent the flange enough that I can’t even get off the outer ring from the filter.

Well, so much for padded cases…

In a flashlight, if you want it to be “tactical”, you need to be able to crack open someone’s melon with a good hit. So you don’t also want to bust the front glass and bend the bezel, etc.

There are clearly limits to thinness and the tradeoff with ruggedness. I think the m2 has it right though. I think thinner would be an issue but don’t see much practical advantage going much thicker. If you have to chisel your way out of a jail cell then maybe you have bigger problems anyway. I’m personally not willing to experiment with my sample, but I bet the impact would have to be quite severe to render it inoperable.

I thought the same about a hard-shell padded case, but I guessed wrong.

Crazy bad luck there. I know a lot of things are possible, but in a situation where I need to smash a ss bezel into a
live melon, the last thing I’ma be worried about is if I can later unthread it lol

SFT40 showed up yesterday In Convoy FB group... tempting...

id30209 wrote:

SFT40 showed up yesterday In Convoy FB group… tempting…

Interesting. I had not heard of the SFT40 before. There is some discussion about it on the German forum. I guess it’s a stock domeless SST40.

SFT40 is 5x5mm and Osram CULPM1.TG is 4x4mm so i guess it could be stock dedommed SST40

SFT is Luminus speak for domeless 4mm², so very likely a minimally modified sst-40 optimized for no dome. Likely additional silicon pour around the perimeter to protect bond wires and such and maybe a slightly different phosphor mix. I’m very, very excited for this emitter. Just heard of it in TLF yesterday. I knew about the SFT70 from this thread, and I think I’m looking forward to that one more. I think it could be the sweet spot for compact high lumen throwers. 1/3 the cost of the 90.2 and in a 5050 footprint that’ll fit existing mcpcbs, reflectors, and gaskets. Win win win. Bring on the next gen boost drivers please!

Hopefully they offer some better tints and lower CCTs. The SST40 has fantastic efficiency and performance but terrible tints.

SFT40? Without (preferably ultra) high colour rendering, another one I'll for sure pass on.  

Would it be possible to get the SFT-40 and SFT-70 datasheet?
Because I found the Design Guidelines for SFT Chipsets Assembly (SFT-10, SFT-16 and SFT-20).
In it, it says that the thermal pad is electrically active, it is common with the cathode.

.

Just seeing that sbt90.2 on 20mm mcpcb drop in the AX store, Simon you are the King of flashlight modification supplies and straightforward , elegant flashlight design. This is exciting, I have been recently researching what to do for an mcpcb for a sbt90.2, and now I can just order one from you. Amazing. Thank you!

Simon, would be nice Quad S21A/B style side switch
with variable tint and Nichia (E21A?) Leds.
An alternative to IF25 with smaller head size

I have a question- I have the yootoo (amutorch) sd3 small 21700 thrower and I’ve done a couple led swaps and now it’s just not working. I think the driver is shot. I pulled out the driver and now I need to replace it. I want to run either an xhp50.2 (3v) in this or perhaps even a sbt90 later at some point (just for fun). So I am thinking I should get a FET or direct drive driver. It is a mechanical tail switch (not sure if it’s forward or reverse). What would be a driver on simons store that would fit this application? As much as a 6amp or 8amp driver is probably better, I want to be able to have open access to turbo in the even I do put that sbt90 in it some day.

Thanks in advance.

Any news about GT driver?

Here's a fragment of a PM I had with another forum member. Here it is in case someone else can benefit from it:

myself wrote:

…

The Osram CSLPM1.TG or CULPM1.TG (2mm² white flat) runs better at 8A? 8A is about the maximum it can take, according to the few available tests in the OSRAM CSLNM1.TG & CULNM1.TG 1mm², CSLPM1.TG & CULPM1.TG 2mm² thread: CSLPM1.TG peaked at ≈7.6A in test by djozz and peaked at ≈8.2A in EasyB test. The CSLPM1.TG is the 3030 footprint emitter, and so the CULPM1.TG (4040 footprint) may tolerate a little more. Still, as a rule I wouldn't go as high. I did three CSLPM1.TG builds, using this weird firmware linear driver set at ≈6.5A max (in all but the first build) by modifying the sense resistor (stock it drives at up to 8A, with enough input voltage of course).

If you don't want 8A, but something slightly lower, you need to modify the sense resistor. Let's for example say you wanna try the above-mentioned weird firmware 8A linear driver:

Its maximum sense voltage is V = I × R = 8A × 0.005Ω =40mV. Thence, with this voltage you can now how much current will go through (each) sense resistor when in high mode. For 6mΩ sense resistors (if you can find them): I = V / R = 40mV / 6mΩ = 6.667A. You can also stack resistors, resistors stacked are sort of like multiple parallel roads to the same path, increasing current flow, this is what I did in my builds. If, for example, you stack 10mΩ plus 20mΩ plus 50mΩ, you get 6.8A: I = (V / R₁) + (V / R₂) + (V / R₃) = (40mV  / 10mΩ) + (40mV / 20mΩ) + (40mV / 50mΩ) = 4A + 2A + 0.8A = 6.8A. With this math you can calculate the current which goes through whatever sense resistor: for 7.2A, an R010 plus two R025s, and so on…

The weird firmware 8A linear driver also doesn't has thermal regulation, this is a plus for power users.

According to feedback, the Convoy 8A buck driver “for CULPM1.TG” emitters uses a 10mΩ sense resistor, for a peak 80mV sense voltage, which already is a tad gross considering the amount of current it can output… 

All of the Convoy SST-40 drivers feature thermal management, but it can be removed: Convoy 17/22 mm SST40 Driver temperature protection removal (the thread is valid for all SST40 linear drivers with dual MOSFETs, like this ∅20mm driver).

P.S.: the thermal management stuff also involves most of the newer Convoy drivers, it is linked to the newer 10-pin MCU. If I recall correctly, the negative thermal coefficient (NTC) resistor is connected to pin 8 in the MCU.