Armytek Wizard Pro Prototype Long-Term Testing

[Trevtrain]

IIt seems to me, that when you first reproduce it, you somehow enable it to happen often.

Yep, even though this makes no sense at all it does seem to be a pattern.

It also happens on normal use (while not using any strobe or voltage meter)

+1

i have the indicator led turn off - maybe this is a factor.

Great observation! I'd totally forgotten that I'd turned this off around the same time I started testing the strobe and getting the problem.

Also i think that it is much more probable to happen if you turn the light off in the time it is off - to explain it: when you have strobe or voltage meter, there is a pulse (or a pause) when the light is not on - if you turn the light off in this time you increase the probability of it locking up.

Ummm.... I tried to experiment with this in line with theories about current pulses and residual charge but despite quite a few attempts I couldn't draw any conclusions.

 

[cronus]

I have done several cycles with the indicator light on, and no lockup so far. Enough tests that I would be able to reproduce it several times.
That would be one strange bug!
Edit: I disabled the indicator LED and it locked up on my first attempt.

 

[paja-jojo]

Ummm.... I tried to experiment with this in line with theories about current pulses and residual charge but despite quite a few attempts I couldn't draw any conclusions.

It's just a feeling - when trying to replicate it i think it was more probable, but as it is nearly random, wher this problem occurs it is hard to prove it factually.
I tryied to replicate now and got 2x lockup in last 10 minutes both on powering off when the light is off, i'll now try to replicate it in times, when the light is on - which is happening as well because it happens also on the normal (continual) usage.

 

[cronus]

I reported the indicator LED component of the issue to Armytek. We'll see how they deal with this now that there is a 'workaround' of sorts...
For me, I don't know what is more annoying; the indicator LED or the lockup. :D
Edit: I was too hasty to call this a workaround. I cannot use the light with the indicator enabled. See my comment below.

 

[paja-jojo]

I reported the indicator LED component of the issue to Armytek. We'll see how they deal with this now that there is a workaround...
For me, I don't know what is more annoying; the indicator LED or the lockup. :D

I consider this not a workaround - i need the indicator turned off as the intensity of the indicator led is to high, would it be just 1/4 the intensity i would consider it usable. Unless there is some paint that can be aplyed to the button rubber and is save and does not void waranty.

 

[cronus]

I agree completely, it's an unacceptable workaround for me too. Also, I was hasty to say it is a workaround, because it may still occur in that mode, just not occurred yet.
Since I EDC this, I have to have the green LED off. The green flash is too bright for use at night as a bedside light.

 

[RedForest UK]

I have done several cycles with the indicator light on, and no lockup so far. Enough tests that I would be able to reproduce it several times.
That would be one strange bug!
Edit: I disabled the indicator LED and it locked up on my first attempt.

Wow yeah, I hadn't switched the indicator light off since I received the light. But sure enough after I turned it off and completed the process previously described it locked up on around the 3rd attempt.. and then again on the fourth.

Could it be that there is some residual power in the circuit which would normally be used to power the state indication LED but when that is switched off somehow interferes with sensing the button press?

Maybe there could also be a link between the fact that the indicator LED flashes every four seconds, whilst the lock-up seems to require the light having been switched off for around five?

 

[Dhobi]

I played with my Wizard Pro two days with indicator ON with no lockout.
Then I switched the indicator OFF, accessed specialty (decimal/strobe) mode about 20 times and the lock out occured in almost all cases. Sometimes immediately after jumping out from decimal/strobe, sometimes after several switching through main/firefly modes.
Guess I need to send it back to Armytek for replacing but may be it would better to wait a while to be sure they fixed the bug in production. Dont want to receive replacement unit with the same problem...

EDIT: Above mentioned test was conducted with Armytek protected cell. Then I used Panasonic 3100 unprotected (supplied with the light) and during about 5 minutes playing, including accessing specialty modes, I noticed just one lockout. (Wish no one would occure because beyond this issue I really like the light and looking forward to use it on mountain bike...)

 

[bradthebold]

That's weird. I've always had the indicator light off and never used strobe mode and mine locks out. I'll have to play with the strobe and then turn the indicator light on and see if it still does it.

Edit: With the indicator light off, I was able to replicate the lockout within 3 tries testing with the strobe mode. I wasn't able to replicate the issue with the indicator light on while testing for a few minutes.

 

[Trevtrain]

Thank goodness for the Interwebs

Thank goodness for the Interwebs and the collaborative powers of the flashlight forums brains trust!

Looks like we've made some proper progress at last.

Only a short time to play this morning as I have to leave for work but the LED indicator seems to be the deciding factor. The strobe may or may not be involved.

My results the same as cronus and others - with the Indicator turned on again I haven't been able to get the lockup. I will continue testing this evening just to be sure.

My unqualified opinion is that whatever "residual charge" is causing the problem is drained by the Indicator circuit just as it would be by loosening the tailcap and pressing the switch or waiting for it to decay.

Knight_Light - If you are about would you care to test this theory on your pre-production sample for us as well?

 

[tony613]

Re: Thank goodness for the Interwebs

So far I seem to be the only one that has had the problem happen with the switch indicator enabled. As I posted above, it happened to me once. I noticed from across the room that the switch was blinking red about once a second - low voltage indication, but the battery voltage was 4.16 volts (DMM). I have not been able to reproduce the problem with the switch led's enabled using the strobe method, but I HAVE reproduced it at least 10 times with the switch led's disabled.

It is currently in the locked out condition right now, and I'm going to leave it that way because I have to go out. I'll see if the light heats up or shows any other reportable behavior.

 

[cronus]

Re: Thank goodness for the Interwebs

I would not be surprised if the issue occurs with the indicator enabled. Maybe it is less likely, who knows. All we can do is guess what might be happening. Let's hope the Armytek engineers get this one figured out so we can have reliable lights. That's the Armytek claim to fame, right?
For now, I'll keep the indicator on and turn it off at night. Nowhere near ideal, but I can deal with it for a little while.

 

[RedForest UK]

Re: Thank goodness for the Interwebs

I just unscrew the tailcap a quarter turn at night if the flash is bothering me. It's not my first choice night light anyway with the PWM on the lowest mode.

 

[cronus]

Re: Thank goodness for the Interwebs

Good idea RedForest_UK!

 

[Kaban]

Re: Thank goodness for the Interwebs

Is it safe to assume that 18650 headlamps (in general) can be safely run on two CR123 batteries?

 

[Trevtrain]

Re: Thank goodness for the Interwebs

Is it safe to assume that 18650 headlamps (in general) can be safely run on two CR123 batteries?

No.

It depends on the voltage supported. Armytek specifically says 1 x 18650 for this light.
The Armytek Predator can run on 2 x CR123 or 2 x 16340 but not this headlamp.

Also, some high-powered lights designed for 1 x 18650 may draw too much current to safely use CR123s.

 

[bradthebold]

Re: Thank goodness for the Interwebs

This is their latest response about the stepdown and lockout:

All runtime test by all manufacturers and reviewers are done in
conditions of good air-cooling. That's why the numbers
correspond such conditions. If you ride a bike, for example and
the headlamp is able to loose heat faster, there will be no
step down in the output and it will work for an hour at Max
mode.

We are very sorry for these misunderstandings caused by a
mistake in the manual. Here's the corrected information:

Warning Indication. In ON-state shows the battery voltage and the temperature inside the headlamp. When the voltage drops, the headlamp main LED flashes 3 times, and when the temperature rises – 5 times. Additionally, the color LED shows the hazard level: voltage – 1 flash, temperature – 3 flashes. Warning level indicated once a second in corresponding color (light output decreases to Main2 mode). Critical level indicated twice a second in corresponding color: voltage – 1 flash (light output decreases by 25%), temperature – 3 flashes (light output decreases to Firefly1 mode). See the figures for the detailed correlation of indication LED color and pauses between flashes to the voltage and temperature.
For example: In environmental temperature equal to +27C the headlamp delivers light in Maximal Mode for about 5 minutes and then the brightness steps down to the Main Mode 2 level. 5-20 seconds afterwards the brightness increases to the Maximal level again. This stepping goes cyclically to maintain the user's safety and the headlamp's functionality). In the conditions of good air-cooling the headlamp produces constant light even in Maximal Mode.

The possible changes in the indication mechanism are now discussed and
we plan to change some things.

I'll contact you as soon I get new information.

------------------------------

So with their new system, there is no warning. Warning drops the output, which would prevent the light from getting any hotter. So it would never progress through colors or ever reach the critical state.

 

[ArcticHighlander]

Re: Thank goodness for the Interwebs

I just unscrew the tailcap a quarter turn at night if the flash is bothering me. It's not my first choice night light anyway with the PWM on the lowest mode.

+1

 

[ArcticHighlander]

Here's what I do to reproduce it, and what I sent to Armytek:
a. Turn on the light.
b. Click four times to activate fast strobe.
c. Wait between 5 and 10 seconds, then turn the light off.
d. Wait another 5-10 seconds and press the button and see if the light turns on
About 1 in 10 attempts will cause the light to lock up.

The odds of occurrence seems more likely than 1 in 10 now, for me anyway.

Tried the above. On the 2nd try it locked up. I had the State Indicator turned off. I tried turning the State Indicator on to see if it mattered but it locked up when I tried to do so four times in a row. On the 5th try the State Indicator turned back on. I then tried the above again but after 3 attempts it would not go to strobe at all, but shut off instead each time. On the fourth attempt it finally went to strobe and I repeated the above steps 10 times in a row without a lock up (or a failure to go to strobe). During these tests I was using the original unprotected Panasonic 3100 that came with the light.

 

[ArcticHighlander]

Thanks tony613, glad to hear a similar (and even more direct) tone from Armytek regarding replacing the light.
My light started doing it the first day I received it, others seem to have it occurring later on, but once it occurs it seems to occur more often thereafter. Curious behaviour. :thinking:
Edit: Armytek are claiming this is limited to a few units that have been reported to them, which may be true. Anyone who is experiencing this and are concerned about it should make sure to report it to Armytek so they are fully aware of the scope of the failures, and to get your name on the list for replacement/repaired lights.

I just reported my lock up problem to Armytek and mentioned that this seems to be a common problem based on the responses here. I also mentioned the problem with the defective lens coating and that it has been reported by a couple other people as well.