Supbeam K50 V2 Mod Thread

Thanks again ryansoh3. I guess I have seen images of that side of the driver. Even in this thread at Post 154. Just didn't realize it. Looks exactly the same as far as I can tell.

Look forward to seeing the other side. Maybe it will help us with this driver. Thank you again.

Here are some shots of the driver. I couldn’t find my lens extension tube, so the quality isn’t too great. :~
They are very high res, so click on the link for fullscreen+ sizes.


Sweet. Thank you very much ryansoh3.

So that driver drives 3 emitters in series in the TN30?

Yup, 3 XM-L2’s in series for a total Vf of 9-10.5V. Current is 3.0A at level 6 I think.
It’s a very nicely regulated driver that can drive a laser diode at 5.0V @ 2.4A.

Visually, it's strikingly similar to the K50 driver. The obvious visible difference is the 2 major capacitors are huge and there is that diode on the TN30 driver. There are a couple extra small caps at 5 o'clock. Didn't compare resister values. Here is a mouse over.

Wish I knew enough about electronics to tell if the K50 driver could be easily moded to higher voltages.

Can someone make a single post with the various similar drivers, what their input config is, and what their LED config is? I don't know any of these lights well enough to go back and track down each different version and keep them straight well enough to figure any of it out.

Single Emitter Lights
table(table#posts).
|Light|Input|Output|
|K40|3x li-ion|1x XM-L2|
|TN31|3x li-ion|1x XM-L2|
|TN32|3x li-ion|1x XM-L2|
|K50 V2|4x li-ion|1x XM-L2|
|TN35|3x li-ion|1x MT-G2|
|K40M|3x li-ion|1x MT-G2|

Multi Emitter Lights
table(table#posts).
|Light|Input|Output|
|TN30|3x li-ion|3x XM-L2|
|X40|3x li-ion|3x XM-L2|

Here's a better aligned mouseover:

Are there good pics of all those drivers to help nail down exactly what's different between the ones that will output more than ~4 volts and the ones that won't?

Not that I'm aware of, but I have compared some pics of the TN31 and K50 V2 and they appear identical (didn't compare individual component values) except for the mod to the sense resistor bank.

Nice mouse over enhancement CC. It appears that the traces are the same.

I’ve sent an email to Bella to forward the driver conundrum to the engineers at Supbeam/Acebeam.

We’ll see what she says.

Is there a buck controller on the board anywhere? Or is it all done in the PIC? Could it be a firmware thing? TN30 & K50 both use the same 16F616. I can't see the markings on the little 5-leg things well enough to figure out what they are, those are the only other possibility for being a buck controller I can see on either side.

Thanks ryansoh3.

In you table above. Is the TN32 an xml2 light?

EDIT: It is. It's the TN35 that is MT-G2.

It reads C2KV on my TN30 driver, but my Google-fu isn’t strong with this one.

What’s the package format of the 5-pin component?

Yup, XM-L2.
I forgot to add the TN35, which has a MT-G2 emitter.

Single Emitter Lights
table(table#posts).
|Light|Input|Output|
|K40|3x li-ion|1x XM-L2|
|TN31|3x li-ion|1x XM-L2|
|TN32|3x li-ion|1x XM-L2|
|K50 V2|4x li-ion|1x XM-L2|
|TN35|3x li-ion|1x MT-G2|
|K40M|3x li-ion|1x MT-G2|

Multi Emitter Lights
table(table#posts).
|Light|Input|Output|
|TN30|3x li-ion|3x XM-L2|
|X40|3x li-ion|3x XM-L2|

I’ve grouped the drivers that I think are identical except for current sense resistors (0.5-1.0A difference):
XM-L2 low current K40/TN31
XM-L2 high current TN32/K50 V2
TM-G2 TN35/K40M
The TN30 and X40 are identical except the fact that the X40 is infinitely variable.

These two:

One at the left clearly goes to the FET gate, so that's most likely (though not necessarily) a gate driver; the other one is the main suspect, I think.

Package size... SOT-23-5? They're basically same size as the mag sensor chips and the tiny little FETs right? So smaller than a 7135 (SOT-89).

Whoops, I was looking at the wrong side.

My TN30 reads 8NDE on the left and SB6Y on the right.

Thanks Ryan. I Goggled and edit cross posted with you.

I did some Goggling, but I can't find any good images of the TN35 and K40M drivers.

EDIT: Thankfully, it appears that the answers may be in the images that Ryan provided. Well, that combined with people smarter than me.

I love the efficiencies gained by these large properly designed reflectors. Here’s another fine example of efficiency versus what pleases the human eyes. While the differences are shown to be significant with a lux meter while moving from level 5 to level 6 (max), the eyes perceive it only as a slight increase. 2.7A versus +5A is rather sobering for a comparatively small perceived increase in apparent brightness. I went through a set of cells while hiking through the hills last night switching between L5 & 6 to confirm (mostly max range spotting). IMO, (and not surprising to most of us) it does nearly as well at 2.7A as at 5A.

Sorry if this has been mentioned before, but does anyone know if the mcpcb is a direct thermal design? If not, could I dismount the emitter and expose the copper for the center pad and reflow?

I think it’s been confirmed as a copper MCPCB, but I don’t think anyone has actually removed the emitter and tested the thermal pad with an ohmmeter, to confirm whether or not there is a dielectric layer.