Quark AA2 S2 Failure

beamis

Newly Enlightened
Joined
Dec 13, 2009
Messages
171
Location
Southern California
I've had a 4Sevens Quark AA2 S2 for about three years, and it has performed well up until today. Today I noticed that the moonlight mode, low, and medium no longer work. Instead I just get the preflash and then the LED goes dark. Turbo, High, SOS, beacon, and strobe work fine. I cleaned all the contacts until they looked like mirrors and I cleaned the threads until they looked like the day I got the light. A fresh pair of Eneloops off the C9000 and still no luck. I contacted 4Sevens, but I'd never heard of this kind of failure before so I thought I'd check to see what anybody here knew.

EDIT - I double-checked that the tailcap is tight. I'm going to be really bummed if I can't get a replacement S2 or R5 head. I don't want an XML Quark.
 
Last edited:

reppans

Flashlight Enthusiast
Joined
Mar 25, 2007
Messages
4,873
By pass the clicky just to see if it's the head or tailcap. With the tailcap off touch the bare tube edge to the neg. batt. terminal with a piece of tinfoil to test the head functions and modes. If it works then try tightening the clicky retaining ring with needle nose pliers (silver disc with 2 dimples surrounding the spring), and/or the clip's screwdown collar.

If you still have problems, send it in... You're still under warranty and they'll take care of you.
 

beamis

Newly Enlightened
Joined
Dec 13, 2009
Messages
171
Location
Southern California
Thanks for the tip. I used a paper clip to make the connection at the tail. Same story :( I really hope 4Sevens won't replace it with an XML head or a cool white XP-G2.
 
Last edited:

pjandyho

Flashlight Enthusiast
Joined
Oct 29, 2003
Messages
5,500
Location
Singapore
What's wrong with the XP-G2? I am a neutral and HCRI guy but I am perfectly happy with my XP-G2 Quarks. They may be cool white but the tint is almost a nice creamy white on mine.
 

pjandyho

Flashlight Enthusiast
Joined
Oct 29, 2003
Messages
5,500
Location
Singapore
I understand. I dislike bluish as well, but my sample of XP-G2 is pretty creamy white than bluish. In fact I am very pleased with the tint. That said, mine is only one sample out of the many thousands made so my experience does not speak for the entire lot of the XP-G2 Quarks out there in the market.
 

monkeyboy

Flashlight Enthusiast
Joined
Mar 7, 2006
Messages
2,327
Location
UK
I've had a 4Sevens Quark AA2 S2 for about three years, and it has performed well up until today. Today I noticed that the moonlight mode, low, and medium no longer work. Instead I just get the preflash and then the LED goes dark. Turbo, High, SOS, beacon, and strobe work fine. I cleaned all the contacts until they looked like mirrors and I cleaned the threads until they looked like the day I got the light. A fresh pair of Eneloops off the C9000 and still no luck. I contacted 4Sevens, but I'd never heard of this kind of failure before so I thought I'd check to see what anybody here knew.

EDIT - I double-checked that the tailcap is tight. I'm going to be really bummed if I can't get a replacement S2 or R5 head. I don't want an XML Quark.

I have a similar problem with a quark 2x123 xp-g except mine was acting up from day 1. To begin with, moon mode would occasionally not work, but now I'm completely missing the first two modes, moon and low. I know their customer service is great but I really can't be bothered sending it back. I've bought 3 lights from 4sevens in total and ALL have had minor issues. I've returned one, sold another, and just kept the 3rd. I won't buy another 4sevens light.
 

Rod911

Enlightened
Joined
Sep 16, 2009
Messages
302
Location
Melbourne, Australia
What's wrong with the XP-G2? I am a neutral and HCRI guy but I am perfectly happy with my XP-G2 Quarks. They may be cool white but the tint is almost a nice creamy white on mine.
Maybe because he doesn't want to play the tint lottery? If you buy neutral, you're going to get neutral. That neutral might be a little bit yellow or green, but it certainly wouldn't have any blue/purple to it which, as you know, the cool whites have.
 

Etsu

Enlightened
Joined
Jul 1, 2013
Messages
783
What's wrong with the XP-G2? I am a neutral and HCRI guy but I am perfectly happy with my XP-G2 Quarks. They may be cool white but the tint is almost a nice creamy white on mine.

Yes, my Quark XP-G2 (cool white) is actually a nice white. It measures slightly cooler than my Quark XM-L2, but to the eye it looks slightly warmer. I'm not sure why, but perhaps because it also measures to have less magenta tint in it, so looks more pure?

In any case, while I prefer a neutral white, the current Quark XM-L2 and XP-G2 emitters are on the warm end of cool white. Around 5500K-5800K. They're not the angry blues that older emitter designs tend to have. I'd love to get a Quark with a neutral Nichia 219, but they only come in cool whites now.
 

Cypher_Aod

Enlightened
Joined
Apr 5, 2010
Messages
265
Location
London, United Kingdom
It sounds like one of the components on the driver has failed, probably the PIC12F629 controller - all the high modes work and none of the low modes do.

I'm not familiar with the specific driver design that 47s use, but given that the PWM modes (max, strobe and SOS) work but the current-regulated low-brightness modes don't, I'm more inclined to suggest that the microcontroller works fine but the current regulating part of the circuit has failed instead?
 
Joined
Jul 2, 2008
Messages
553
Location
Gainesville, Florida
I'm not familiar with the specific driver design that 47s use, but given that the PWM modes (max, strobe and SOS) work but the current-regulated low-brightness modes don't, I'm more inclined to suggest that the microcontroller works fine but the current regulating part of the circuit has failed instead?

You know, you are likely right. The LED- is connected to a power transistor, the legs of which are connected to a 6-leg IC marked 01-717 and a 5-leg IC marked SA which looks like a voltage regulator. I guess the 6-leg IC is a logic processor of unknown type. One of those could have gone out as well, but as the high-power modes still work, it is something with logic, rather than power, so I'll guess the logic IC rather than the regulator. None of this would be user serviceable, so it should go to 4-sevens for a new driver.

I'm actually trying to reverse-engineer the quark driver right now to understand how it works, so this is very interesting to me!
 

beamis

Newly Enlightened
Joined
Dec 13, 2009
Messages
171
Location
Southern California
You know, you are likely right. The LED- is connected to a power transistor, the legs of which are connected to a 6-leg IC marked 01-717 and a 5-leg IC marked SA which looks like a voltage regulator. I guess the 6-leg IC is a logic processor of unknown type. One of those could have gone out as well, but as the high-power modes still work, it is something with logic, rather than power, so I'll guess the logic IC rather than the regulator. None of this would be user serviceable, so it should go to 4-sevens for a new driver.

I'm actually trying to reverse-engineer the quark driver right now to understand how it works, so this is very interesting to me!

What I find really interesting about it is that the preflash is still there, just the actual mode doesn't work. Having no light after the preflash gives the opportunity to see what normally can't be seen. The preflash for each mode is brighter than the preflash for the mode before. In other words, the preflash for moonlight is dimmer than the preflash for low, etc. I wonder if the preflash is a brief lighting of the emitter at the next highest output and then stepping down.

EDIT - That wouldn't make any sense because if it were just starting at the higher level and stepping down, that would mean the next level should work.
 
Last edited:

beamis

Newly Enlightened
Joined
Dec 13, 2009
Messages
171
Location
Southern California
The preflash for each mode is brighter than the preflash for the mode before. In other words, the preflash for moonlight is dimmer than the preflash for low, etc.

I take that back. Looking at the light right now, the preflash situation is more complicated than I thought.

When turning on the light "cold" (meaning the light is switched on from being off for a few seconds), the preflash for moonlight is brighter than the preflash for low, and the preflash for medium is brighter than the previous two. When cycling through the modes without turning the light off, the preflash for moonlight and low are the same and the preflash for medium is brighter than both.
 
Joined
Jul 2, 2008
Messages
553
Location
Gainesville, Florida
Preflash is the discharge of a capacitor on the driver board, so it sounds again like the power components are turning on, but the logic chip is not activating the current through the low-mode path. It is probably the high-mode path that is charging the capacitor, as both paths run through this component.
 

beamis

Newly Enlightened
Joined
Dec 13, 2009
Messages
171
Location
Southern California
4Sevens is not very helpful. They flat out refused to give me even a ballpark turnaround time. All they would offer is that they are currently experiencing a delayed turn around time for repairs. I am not interested in purchasing lights - especially at the huge premium 4Sevens charges - from a manufacturer that can't even be bothered to give their customers an approximate turnaround time for warranty repairs. 4Sevens just wants me to ship my light to them with no expectation of when I might receive a repaired light. Maybe in a few weeks, maybe in August ... who knows? That's not a very good warranty if you ask me.

Any recommendations for a good AA or 2xAA light with moonlight mode and no PWM from a manufacturer with a better customer service?
 
Last edited:
Top