Yet another Arc4 anomoly?

Ok here is another one for ya'll. I haven't seen this one or read anything like it. My light lives attached to a 2x123 battery tube and I haven't tried it on the 1x123 config. While sitting idol and off it will on it's own come on at a very low dim glow. It's not flashing in any way. The RFS and locator flash are off. I tried turning these on and no difference. So turned them back off again. Tried a reset, no change. While the battery was off I also tried pressing and holding the switch for a few seconds in an effort to simulate light on momentary to possibly drain anything that might be in any capicitors. Reninstalled battery no change.
The switch seems to feel and operate the same as it ever did (Perfect). Nice pronounced click, good tactile. Other wise the light operates as normal. Just this dim glow. Some times it's there sometimes it's not. No rime or reason that I can figure. Can anybody shed any light on what is going on.

Should also note that mine is a rev.1 if that makes any difference.

cheers,
zwf

Sounds like something mildly electrocunductive is shorting your switch. Try disassembling the switch, cleaning and drying all parts and reassembling. /ubbthreads/images/graemlins/thinking.gif

[ QUOTE ]
Dr_Joe said:
Sounds like something mildly electrocunductive is shorting your switch. Try disassembling the switch, cleaning and drying all parts and reassembling.

[/ QUOTE ]

Assuming you are talking about the actual mechanical switch, why should it have to do anything with this phenomenon?
As much as I know about the light (very little), the only possibility is that an electronic part does not switch off completely (for what reason either) or the part which opens the circuit is shunted in some way.

ZWF, does it depend in any way on temperature? Does it get more or less when cooled down or warmed up (within reason)?

PeLu said:
ZWF, does it depend in any way on temperature? Does it get more or less when cooled down or warmed up (within reason)?


Not that I've noticed so far. However the jury is still out on that one.
zwf

You found an odd one ZWF. I doubt seriously the problem is conduction through the switch since that is connected to the digital part of the circuit. It is either on or off. Sufficient partial conduction would cause the switch to appear stuck on.

I would be interested in your observations of the light with a 1x123 pack. Also, how much current does the light draw when the light is set to "off".

With the RFS turned off and the light in off mode, there should be almost zero voltage to the LED regardless of how you sweep the voltage across the input range. With the RFS on, you will see a ramping charge/discharge voltage to the LED that can cause a faint flash with low Vf parts. I think in some cases, this is helped by increased leakage across the FETs on the board.

It is unlikely that this conduction was caused by customer introduced moisture because those parts are on the board that is furthest from the rear of the head. That would place it deeply within the epoxy used to seal the head.

However, moisture could have been introduced during manufacturing (the boards were washed by the CM) and cause various symptoms including this. I would have like to have conducted more research on this.

I would suspect that captive moisture would dissipate through the layers of the board and other boundaries via capillary action. If a symptom was moisture related, it would seem to me that is would follow a bell curve of influence over time as the pool of moisture sought to balance itself evenly throughout the entire cavity. This means it would get worse before it got better and at some point it would seem to not get any better.

Another possibility I looked at is that excessive current may have caused a part to fail. However, the circuit is fairly durable. The failure modes I have seen don't cause this type of symptom.

So, I really don't know what it is. I am curious though if you want to delve into this a little further. We may find that the only fix is total replacement however.

Peter

Well now, My light would seem to have a multiple-personality disorder. Tonight as I write this the light is acting normal. Can't replicate the problem. I guess for now I'll have to let it sit in the corner and think about what it did wrong /ubbthreads/images/graemlins/thinking.gif Hopefully we shall never see this re-occur. I'll keep you posted.

cheers,
zwf

ZuluWhiskeyFox,

If I understand your problem statement correctly, the light functions normally in every way except the light will not turn completely off - being very dim when it should be off. Further, a reset had no effect on the problem.

Since the light is otherwise functioning correctly, we can eliminate everything related to the switch. That leaves electronics and software. A reset puts everything back into a known state - both hardware and software - and since this had no effect, the problem is most likely hardware. The discussion about leakage and moisture in a previous post are erroneous and should be ignored.

The most likely cause of this problem is that the light is failing to fall asleep when it is turned off. Whenever the light is awake, it is generating power - even when "dark" - hence the very dim light output. This is why the light is never completely dark when it is turned on - such as when displaying the menu flash sequences or waiting for a command sequence to complete. This guarantees there is sufficient power to allow the light to operate at voltages down to 1.5V once it is turned on. Another example is the "dark" flash, which is used to discharge the capacitors to ensure RFS works reliably when the batteries are changed - the very faint flash that is sometimes visible when RFS is enabled - especially when the light is cold.

Anyway, the inability to fall asleep can be caused by several problems, all of which I can positively identify causing this problem lead back to non-fixable hardware problems. None are related to software.

The good news is that the light will run for a few hundred hours on the "dark" setting so you can mostly ignore the problem - especially if you are using rechargeable CR123A batteries. Worst case is you have to replace the batteries every week or unscrew the battery case (1/4 turn) when the light will not be needed for long periods.

Hope that helps.

Henry.

Thank you both Peter and Henry. Henry yes you understood my problem exactly. It won't always go to sleep. A flashlight with insomnia. Bless it's little heart. It doesn't want to stop making light.
Peter you were curious about how it behaved with a 1x123 instead of my normal config of 2x123. Well it seems to behave better when on 1 battery. It seems to go to sleep better. On the 2 battery pack is when it doesn't sleep so well.

Humph, So the 2 battery pack is supposed to extend battery life. But that gets somewhat negated by the insomnia problem.

I wonder what's the delivery time on HDS lights these days? The Ultimate-60 is starting to look better and better all the time. /ubbthreads/images/graemlins/smile.gif

Maybe it could be galvanic corrosion, the electrical potential produced by a different alloy composition, in the 2x123 tubes that where offered later(?) as an option. Any metal in the head touching the side body would complete the circuit. The mositure in the air could supply the medium.

But, then again, I don't believe L.H.O. was guilty. So, what do I know.

Fin

OK tonight I have another observation for everybody to ponder. While attached to a 2 battery pack, it draws .034ma when it's supposed to be sleeping. While attached to 1 battery pack it draws .005ma when it's supposed be sleeping. At the .005ma current draw there is no visible light. However at the .034ma draw there is the very dim almost invisible glow I mentioned earlier. So what do y'all think about them apples? /ubbthreads/images/graemlins/duh2.gif

I think tha light is possessed by deebils /ubbthreads/images/graemlins/evilgrin07.gif /ubbthreads/images/graemlins/ooo.gif /ubbthreads/images/graemlins/aaa.gif
and ya oughta send it ta me so's I can depossess it/ubbthreads/images/graemlins/twakfl.gif

Sadly /ubbthreads/images/graemlins/mecry.gif the light will not be able to be returned
the depossessin' process /ubbthreads/images/graemlins/touche.gif is rather messy /ubbthreads/images/graemlins/dedhorse.gif /ubbthreads/images/graemlins/eeew.gif /ubbthreads/images/graemlins/str.gif

But, you will not be bothered by any more of them flashlight deebils /ubbthreads/images/graemlins/evilgrin07.gif and can get yor life back in order /ubbthreads/images/graemlins/sleepy.gif /ubbthreads/images/graemlins/clap.gif

Fin /ubbthreads/images/graemlins/grinser2.gif

ZuluWhiskeyFox,

Same prognosis. The light only generates light when it is awake - it is impossible to generate light when not generating power. The 5uA idle current is the current when sleeping. The 34uA is the current for the "dark" level.

However, this does point out that my calculations for battery run time on the dark setting were wrong - since I just used the "minimum" setting power for the calculations. At this current draw (35uA), your batteries should last over 3 years and therefore you don't need to be concerned about the light failing to sleep.

Henry.

Henry, So if I understand you correctly the 5uA current draw when sleeping is normal for Arc4's in general. And the 34uA current draw is in fact awake and making light at the lowest possible setting that the driver can run at ie. "dark" level. So when my light is attached to the 2 battery pack it is incapable of sleeping. Kind of like having to much coffee. And as you pointed out 35uA draw the batteries will still last many years. I guess not so bad. Could be worst.