As long as you mentioned it… I recently ran into an issue with strobe too. Strobe modes were actually making the driver reboot.
I thought it must be a flaky diode, or maybe C1 was getting overloaded, or maybe I had a weak solder connection… but no.
It turns out the issue was a weak spring. I guess going from zero to 100% FET in a single clock cycle was enough to literally knock the spring off the cell, for just long enough to interrupt power and cause a reboot.
I have used several different capacitors depending on availability. Regardless, you want a 10V+ at least X6S capacitor. X7R or X7S is even better. Preference should also be given to 10% instead of 20%. The ones I sent you are 16V X6S 10% Murata. I usually use Murata (Japanese made) but there are lots of good manufacturers; just don't buy Y5V capacitors or capacitors from unknown sources. The cheaper capacitors suffer from more DC bias derating, temperature derating, and their lifespan isn't as good. When you're building an expensive driver or light it's worth it to spend the extra pennies and get a capacitor that will go the distance.
Here are a few suitable capacitors.
I also use the PMSN3R0-YLDX and also the PSMN4R0-YLDX depending on the board and what I can get. There was a while when the 3R0 was unavailable so I had to buy some 4R0.
Thanks RMM. X7Rs on the way..
A usefull mosfet update: i got my hands on AP1A003GMT-HF-3TR MOSFET lately (PMPAK®5x6, INFO HERE)
Using this one vs the default recommended i got steady 5-8% better high amps on both XPG2/XPL emitters
thanks for letting us know. Any hint where you got them from? I recently ordered the board vestureofblood posted above and all I am missing is the FET. not sure which ones/where to get them from.
Since you are in the US try MNT or Mouser( Element 14 aka Newark NEVER- bloody thieves they are), or Ali for the default ones
Those that i mentioned came from a local EU vendor but i am sure Mouser have those too, bought them mainly for their price…with the vat( we do love vat in EU, 22) and tax( 4.5) 52 cents/pc :bigsmile: …and they turn out good ones
Thanks Mitko! I'll be checking those out too. Sounds like a real nice bump for those high Vf LED's, I assume better than the SIR800DP.
Thanks Mitko 
Can't locate those MOSFET's anywhere. The manufacturer, APEC, doesn't seem to be popular. Can't find them on Mouser or DigiKey.
Hm… strange, i posted those cause in Europe they are cheap( 52cents with our huge taxes) and good, was thining they would be an easy find in the US
At least we have something that cant be found in US!
P.S
I am on a night swifth making some of those atm btw, hunters got pretty active lately, maybe cause of the Xmass
Tom i have a question, do you also like me recieve every time those with a different minus contact area thickness?
Those on the photo got thick minus contact rings, sometimes my order comes with thin ones, sometimes even the contact circle is covered completely and i have to rework them
Boy, I'm confused... I'm only using FET+1 drivers - wights v009 version, same one that the BLF A6 driver was based on. I really love having the single 7135 to have very efficient low modes.
Think what you (VOB) and Mitko have are FET only boards - I never ordered that specific board - I used to use the ones in the OP though. OSHPark support told me they've had these weird masking problems with wight's boards (a few) intermittently - they dunno what's causing it, couldn't explain it. I got a batch in once with pads covered, asked them, and they rushed out a new set free of charge, even though I could scrape off the contacts and salvage the bad ones -- OSHPark has great support, try them.
For the masking of the ground (I think you mean the outer ring?), don't think I've seen that specifically, accept the set of boards that came in that one time completely covered, but it wasn't this board.
Actualy that masking isnt that big issue….just asking, OSHPARK are quite good to me anyway
Yeah i am using mostly v44 cause most of my stuff simply dont need fet+1 drivers, plus i `ve discovered something interesting: some XPG2 emitters dont like FET+1 driver, for instance S2 1A from FT- i have absolutely no idea why
Hunters flashlights usualy dont have that low mode 
700ma, max, battery check, things like that…ppl got annoyed with too many modes
But ofc i am using Fet +1 too, just not that often, like 6-7 monthly
Ordered Attiny 85s btw, have to try our your new firmaware
Thanks for developing it TOm, greatly appreciated
Thanks 
. Ahh - true, good point bout the simpler mode sets. For tail power switch lights, the high mode count and advanced features can be frustrating - for me it's more awkward to deal with than a side switch., though I'd still like to have 2 e-switch's.
Can't recall - was there any advantage to this PCB alternate layout vs. the one in the OP?
VOB - sorry, what exactly is the 1S and 2S driver, and what is a Cznr pad? Think something to a with a cap and/or zener - not sure. Couldn't answer your question cause couldn't understand it.
Thanks Tom, I will clarify. By 1S and 2S I mean can the board be configured to use 1 cell or 2. On my drivers and on RMMs there are 2 separate places to put the 10uf cap. One place is used if you make the driver for a single cell ( C1). The other place is for putting the 10uf cap if the zener diode is use for the 2S board (CZNR).
On this board I see only one place show for the 10uf cap so I am wandering if the board works in both configurations or only when used with a zener diode.
Ok - think everything is there - the C1 pad, the D1 pad to the right of C1 (unmarked though, but indicated by the arrow), then the zener pad to the right of that. Should be the same # of pads. I don't know off hand how to assemble a zener driver (for 2S). Can't find a pic...
Oops, this is one I did for a Roche LS01:
So the 200 ohm resistor goes on D1, and the cap stays where it normally is on C1, and the zener of course on the zener pad. At least this is how I did it, what I thought...
The more I look at the v043 design/layout, the more I like it. Simple but everything is there. Main problem I have is pin #3 and pin #5 are not wired, and would like to use them for LED indicators, so even if they went to a small pad, it would be great.
Update:
Ok - now I understand... You are talking about Richard's Mtn-17DD driver v1.11 here: http://www.mtnelectronics.com/index.php?route=product/product&path=25_122&product_id=272
That board has a CZNR pad that the 10 uF cap is moved to when in the zener (2S) configuration. Interesting, so he moves the cap after the resistor...... Hhmm... I'm thinking Richard had a good reason for moving the cap so it's after the 200 ohm resistor, and not before. If it's before, I'm thinking it's getting 8V from the 2 cells, but if it's after, it's at 4V, like the MCU... Oh boy.
Dunno - Richard, or someone more knowledgeable needs to address this.
For the zener mod. configurations it is best to have the capacitor after the resistor and next to the zener diode. If you don't, you will get a voltage blip on a lot of startup events that will wig out the MCU. Wight acknowledged that this is a potential issue in the OP, and I found out with my very first revision boards that it doesn't work reliably unless you change the circuit, so in my second revision boards I added the second pad so that you could zener mod without having to stack anything.
The only reason we don't always run the capacitor closer to the MCU is because driving the FET causes an MCU over voltage boost if you have the capacitor after the diode---the inductance from charging the gate charges the capacitor to 6V+ and it has nowhere to go with the diode blocking the backwards flow. It was comfychair who first scoped this out, and I have since verified his results.
Good to know Richard, thanks! I couldn't find it mentioned in the OP though. So, my wight driver in the Roche LS01 pictured in post #627 has it wrong because the cap is still on the C1 pad, before the resistor, but wight's boards don't have the extra pad your driver has. So probably best thing to do on his boards is stack the cap with the zener?
The LS01 seems to work fine, but probably because it happens to be tolerating the voltage blip?
It really depends on the light and components. I built some that worked, then some wouldn't work right. Alex then tried some and found pretty much the same thing. Moving the components closer together solved most, but not all, issues. Stacking or changing the layout will work 100% of the time, which is why I went that route.