Well that’s interesting. I got an S41 today. I popped in an 18350 did some quick measurements for modes. All was well.
Put my BLF A6 tube and a 30Q into the S41 head and ran a 30sec test to measure turbo and all seemed fine.
After letting it cool back down to room temp I cycled up to get a high mode measurement and nothing over mode 2 functions properly. Moon and low is okay, but mid is measuring low and everything above mid measures about the same. Lots of flickering here and there.
Now the interesting part is that a high drain 18350 (WindyFire) still works perfectly fine and measures very close to the 30Q numbers. No 18650 cell works.
I think the battery-tube of the A6 is a bit shorter and does noch give a good contact to the driver.
Screw the tube hard in the head and test it again. To spare the ano, you can take a paper-clip and bend it to a circle and put it in the head before screwing in the tube. Be carefull to not shorten something.
I have similar issues if I use the A6 tube instead of the long S41/S1 tube.
I did bypass the tailcap, but I completely forgot about the tube issues from way back.
Indeed it is the tube itself that is the issue. Amazing. That’s crazy that it works perfectly for a whole test and doing nothing other than waiting, it just starts malfunctioning.
What do you mean, what kind of problems? You mean because of the higher current? Actually my plan was using a X6 bistro driver from banggood on the S41S which is on its way from china, do you know if that driver has problems?
Drivers using wight’s design tend to have a voltage spike to the MCU when the FET gets shut off. This is fine on the attiny13, but the attiny25 MCU is more sensitive and bistro needs a tiny25 or bigger. So it may have a tendency to reboot itself in modes which turn the FET on and off, especially if using an older driver layout like what Manker used.
If I understand correctly, this is why the S41 used a BLF-A6 driver instead of the BLF X6v2 driver. They probably tried the X6 driver first and found it didn’t work in a quad.
Newer drivers (like Texas Avenger, BLF Q8, and RMM’s FET+1) don’t have this issue.
The low half of the modes use only a 7135 chip, which is a current regulation chip, so … yes. Basically. The runtime is relatively easy to calculate by measuring the current with a DMM. Like, if you have a 3500mAh cell and a mode runs at 15mA, then it would run for 3500/15 = 233 hours, or about 9.7 days on that mode.
Thanks for the reply. Currently I only have a ZTS battery tester. What is an accurate enough DMM for these purposes? I don’t know how to use a DMM on a clicky. Is there a good guide on the forum? Sorry for all the questions. When you say the low half of the modes, that includes the first 3 then?
Pretty much any cheap DMM should work fine. Put it in DC mA mode, take off the tailcap, and touch the leads to the cell and a non-anodized part of the host. It should tell you the mA rating for estimating runtime.
Be careful not to get the light into turbo mode though, or make sure the DMM is in a 10A or higher measurement mode first, or you could blow a fuse in the DMM. The 10A mode won’t be very precise for measuring small amounts of current though.
I forget exactly how the modes are spread out, but … well, let’s see. The source code says this:
#define NUM_MODES1 7
// PWM levels for the big circuit (FET or Nx7135)
#define MODESNx1 0,0,0,7,56,137,255
// PWM levels for the small circuit (1x7135)
#define MODES1x1 2,20,110,255,255,255,0
So, the first 3 levels should be regulated. Level 4-6 should be partially regulated and partially direct-drive. Level 7 is direct-drive only.
Awesome info. Thanks for taking the time to teach me. I’ll get a DMM off of Amazon and see how it goes. I’m still confused on how I switch between modes, or even turn the flashlight on, when the tailcap is off: Maybe it will become apparent when I try it out. Super cool to see the source code and the direct drive vs 7135. Thanks again!
When the switch / tailcap is in place, the switch simply connect or disconnects the tail of the cell to the body of the light. When you connect the leads and the meter, you do the same thing, now with the current meter in the circuit. Which means that when everything is properly connected and the meter is in the right mode, it will show how much current is flowing through the meter and the LED. Changing modes will happen with brief interruptions in the circuit, so be deliberate and firm when applying the leads. I would get the fiddly one on the un-anodized aluminum body first, while you have a second hand, then apply the one to the cell. Having a small vise or a minion to hold the light in place and still be able to see the light come on would also be handy.
Thanks. I think I’m starting to get it. So basically I need to enable mode memory. Then whatever mode I turn the flashlight off in, once I unscrew the tailcap and complete a circuit using the DMM, it will be on in that mode? Or, do I deliberately interrupt the circuit and make educated guesses about what mode the flashlight is in? Sorry I’m so clueless about this stuff.
If you hold one lead on the light body, you can literally just tap the battery with the other lead and it’ll act like you’re pressing the button on the tailcap. The flashlight doesn’t know the difference. It’s pretty easy to operate the light when using a DMM instead of a tailcap.
