A bleeder resistor is needed on the driver. It’s the same concept as using a lighted tail switch.
If someone else is interested in producing these in another place they can go for it. I’m going to make these in the USA, either all myself or with domestic manufacturing. The price is in line with what Lexel used to sell his boards without the need for a button cell.
I’m waiting on Oshpark to deliver these pcbs. Hoping to finalize this by the 24th of May and ship by June 4.
This is a game changer to me except if it is not piece of cake to solder lets say from spring side of the driver? I would have to see few pics to visualize that modification…
IF ANYONE knows where to get THIS switch— this IS THE ONE for FET pressure switches on guns. The on/off button acts like a forward clicky— so don’t even need to press the long side of the switch.
Sofirn USED to sell it, and I’m waiting to hear back from Lan/Barry re. availability. You don’t need the tail-cap or plastic wire exit piece (replacing the silicon switch cover) for most lights, if you’re willing to pot whatever switch your light comes with. For the Convoy C8+ : punch a hole in the rubber switch cover, remove the old switch, replace with the FET switch (with any spacers needed) and add silicon— done! (oh and make sure you add the resistor to your driver— more to come on that here… I’m wondering what the best method is for that mod myself)
Have no idea… Sofirn sold it to me last May (with the tailcap for a C8A) and I rebuilt it into one of the C8+ lights— it’s absolutely awesome (after 10 days hunting in the field— using four other different models which all sucked mostly… this one SHINED!). Simon can’t get it, Sofirn hasn’t replied yet, and so I thought I’d throw it out there so someone may be able to ID it— AND provide a source!
Richfire looks like the source to me. Contact them. They also have a 21700 tube style LEP with USB-C charging… linking that in the LEP thread… but anyway they look like they blatantly clone some Surefire and Streamlight designs. Heck, you never know, they could even OEM those two. It’s China after all.
Thanks for the lead… I’ll contact them for an OEM opportunity… but it will be a while to get any info as I’m out for a week starting here in a few days (offline and out of the country)
I looked again and although the switch itself is identical, the wire isn’t coiled like the Sofirn version. But I’d prefer no coiling actually as it gets in the way of things.
Made some prototype . Led directly connected to switch. Full DD power output for 4min , latter starts to drop and in 30sec goes off at Vin 2.8V, at 3V - moonlight, at 4V - 1A. So its like batt indicator and LVP ;)) To recharge you need to turn off for second or few . It works like 1 mode DD. Main driver is unnecessary ;))
So I got my parts in today. Don’t yet have the PCBs so I ‘air wired’ everything. First run with a 1500mah cell @ 3.7V starting lasted 13 minutes before hitting the LVP flash. Retesting with charged cell but thinking something in the 15min is going to be where it lands. I started with a 4.7Mohm gate resistor and the turn off response is just slow enough to cause mode switching to be effected ( off time needs to be a few hundred ms longer than normal ). I’m going to try doubling them up and seeing how 2.35Mohm behaves.
Important note: the 22mm four mode sst40 Convoy driver that I received a few weeks ago works without any additional modification.
EDIT: 2.35Mohm works perfect. Starting the runtime test, but it should be right around 6-7 min.
Good News! I hope the ramping driver used in the C8+ and L21B also come ready to use. But good to know what will be needed if not. Excellent work JaredM!
So the runtime is going to depend on the output level and when it hits LVP on the driver it seems. Since as the capacitor drains, the Vgs decays. Rds-on then climbs an exponential curve as Vcap approaches G-S threshold voltage. On mode 2 of 4 the setup ran over an hour before I cut it short. On mode 4 of 4 (~6A) it ran about 3.5 min. Mode 3 = ~4min. I feel like there is going to be dependencies on the Vf of the emitter too, so I’ll test something high (was using LH351D).