Two Flashlights with similar issues (OFF = moonlight)

Scallywag · 2018-05-10T05:11:59.000Z

I am having issues with two of my flashlights. I have a convoy L6 with XHP70.2 and this driver, as well as a Supfire M6 with Lexel's TA-based 46mm Fet+1+15. Anyway, they are both doing the following:

Whenever there is power to the light, the LEDs are faintly illuminated. For the Supfire M6, I get a trio of dimly glowing Luxeon Vs. For the L6, I get a glowing XHP70.2. Both drivers are otherwise acting normally - I can double-click to MAX, I can click and hold to ramp, and then click to off - except when the driver is "off" on either flashlight, I'm getting moonlight from the LEDs.

The Convoy L6 was working great until today, when I went to install a lighted side switch. However, the switch isn't the issue - I've disconnected it and the problem persists. Does anyone have any suggestions for what I should check? I do own a DMM. I've cleaned both board thoroughly with isopropyl alcohol and examined both closely to rule out any visible shorts. The real common ground between the two drivers is the use of ATTiny85 on both. If anyone can explain to me what has to happen for the lights to even behave this way, I can try to rule out the possible causes. Currently, I'm stumped. Thanks in advance for any help.

TBone · 2018-05-10T07:13:53.000Z

7135 linear current regulators are rated for 6 V. Running them on two fully loaded cells in series only works with a few types. Some 7135 deteriorate over time.

Measure with your DMM if the single 7135 on the L6 driver is leaking.

HarleyQuin · 2018-05-10T07:27:39.000Z

Please state the firmware both driver are using.

I remember that HKJ had found in his driver tests, that under certain conditions PWM 255/255 is not completely on and PWM 0/255 is not completely off (scope did not show a flat line but in case of PWM 0/255 spikes upwards).

This does only occur under certain conditions, but I don’t remember which they were (FET or 7135 / frequency / fast PWM or phase correct PWM). But that could fire up your LEDs although turned off.

Flashy_Mike · 2018-05-10T09:56:30.000Z

I am having issues with two of my flashlights. I have a convoy L6 with XHP70.2 and this driver, as well as a Supfire M6 with Lexel’s TA-based 46mm Fet+1+15. Anyway, they are both doing the following:

With the L6 this has been reported before:

Convoy L2/L6 modding thread Flashlight Modding and DIY Parts

Anyone with Texas Avenger in a L6, and Narsil experience know if this is normal. When side switch is turned off, but not off at the tail switch, the emitter has a very low glow, to it. Its almost like having an illuminated switch at the head. Just wondering if I did something wrong. [image] [image]

HarleyQuin:

I remember that HKJ had found in his driver tests, that under certain conditions PWM 255/255 is not completely on and PWM 0/255 is not completely off (scope did not show a flat line but in case of PWM 0/255 spikes upwards).

This does only occur under certain conditions, but I don’t remember which they were (FET or 7135 / frequency / fast PWM or phase correct PWM). But that could fire up your LEDs although turned off.

This happens with FAST PWM due to the way it is realized but its handled in firmware usually. And in OFF mode the mcu is usually sent to sleep mode with pwm deactivated and output disabled.

The Luxeon V in the M6 have a very low forward voltage, so there might be the same cause for the glow in both lights.

Lick · 2018-05-10T17:40:10.000Z

For a temporary fix, you could just lock them out with the tailcap. Don’t have a solution for you but that was my plan with a lit tailcap anyway. Interested in the reason why this happens!

Scallywag · 2018-05-11T01:18:05.000Z

HarleyQuin:

Please state the firmware both driver are using. I remember that HKJ had found in his driver tests, that under certain conditions PWM 255/255 is not completely on and PWM 0/255 is not completely off (scope did not show a flat line but in case of PWM 0/255 spikes upwards). This does only occur under certain conditions, but I don’t remember which they were (FET or 7135 / frequency / fast PWM or phase correct PWM). But that could fire up your LEDs although turned off.

The M6 has NarsilM 1.2, the L6 has the Emisar D4 firmware option, where the tail seitch is just a lockout. The L6 worked fine until last night, when I added the lighted switch. So I doubt the 7135 went bad suddenly. The switch was supposed to kind if serve as a power indicator for when the light was off but not locked out.

The M6 doesnt lockout well so it's not practical there. The M6 is a much dimmer glow than with the l6. The l6 is bright enough when off that the "power on" two flashes can't be distinguished, whereas that's clearly visible with the m6. I'll likely dis- and re- assemble both lights Sunday again and see if I can figure anythijg out further .

Scallywag · 2018-05-13T05:30:57.000Z

To update:

I've been working on the L6. The LED has a suspiciously constant 4.95v across it when lit in the "off" mode. The draw varies in the 3-12 mA range.

I also noticed that, compared to the picture on mtnelectronics, the driver I have is slightly different. It has an ATtiny85 and lacks the "AEC1" component by the "TEMP" label (re: this image from the driver page)

Also, "RT1" is not present.

What's driving me nuts is that the light was functioning as intended before I dis- and re-assembled it to install a lighted side-switch.

TBone · 2018-05-13T06:17:14.000Z

TBone:

Measure with your DMM if the single 7135 on the L6 driver is leaking.