Misbehaving Luxeon Zetex 300 circuit

dwtowner

Newly Enlightened
Joined
Feb 29, 2004
Messages
4
Location
Bristol, UK
Hi all,

I have now built three 1watt Luxeon drivers using the famous Zetex 300 circuit. The first two of these worked perfectly, but the third behaves strangely. For the first hour or so of operation, the LED will be brightly lit. It will then occasionally start to flicker - just a momentary dimming every 2-3 seconds. After a few minutes of flickering, it will revert back to a steady light, and then at some seemingly random point thereafter, go back to the flickering behaviour. Eventually, the LED will go out completely. At that point, switching the circuit on and off will reward you with another 20-30 seconds of light, before the LED goes out completely again. At this point, the only cure is to replace the batteries. My other Luxeon circuits don't behave like this at all; they gradually dim over the life of the battery until the light becomes too weak to be useable.

Any ideas what might be causing the third circuit to behave differently?

thanks,

dan.
 

d'mo

Enlightened
Joined
May 9, 2002
Messages
937
Location
Rochester, NY
What volatage are you using? if it's dropped below about .8 VDC, it might be momentarily dropping out of regulation, the battey recovers and drops out of regulation again.

I've also had circuits do this if the sense resistor is too small or if they get over heated. Overheating will cause odd even when it's cooled down afterwards. If this is the case, the chip will need to be replaced.

Hope this helps.
 

pbarrette

Enlightened
Joined
May 5, 2004
Messages
346
Location
Huntsville, AL
Hi Dan,

I ordered the parts for 20 of these things so I know exactly what you are talking about. Personally, I'd have to agree with d'mo here.

The flickering sounds like exactly what d'mo states: The circuit is dropping out of regulation which allows the battery to recover a bit, which puts it back into regulation, which drains the battery more, which drops it back out of regulation, etc..

The reason that ony 1 circuit is doing this could be caused by a number of reasons. I'm guessing that you are using a length of wire for the Isense like most of us and those things are hard to get measured exactly the same and soldered with the same tolerances from one circuit to the next. It could also be that one or more of the components are slightly out of spec, requiring the high-end voltage level listed on the datasheet as opposed to the lower average. This could make the circuit operation require a higher on-state voltage and could result in a higher drop-out (off-state) voltage.

Finally, it could also just be a difference in the battery or brand of battery itself. You should take a "flickering" battery from that circuit and test it out in one of the other two circuits to see what happens.

I'd also be interested in knowing what voltage you are feeding the circuit and how much current you can supply to the Lux. I can usually get ~145mA out of 1xAAA and ~340mA out of 2xAAA batts.

I've also noticed that if the Isense resistor is just barely too low a value, that fresh batteries will cause the flicker as well. I'm guessing that it's due to the Isense pin being just barely overloaded which causes the transistor to fire erratically as minor fluctuations in the circuit vary the current levels above and below the overload point.

pb
 

dwtowner

Newly Enlightened
Joined
Feb 29, 2004
Messages
4
Location
Bristol, UK
Hi,

Thanks for your suggestions. I think that you are right that there are two factors here. Firstly, the behaviour does seem to be linked to one particular set of batteries. If I swap the flickering batteries to another circuit, that circuit flickers as well. I haven't tried a non-flickering set of batteries in the `bad' circuit yet though. The bad batteries are the same make and model as the good batteries (Hama NiMH 2000mAh) so I'm not sure why they behave differently; I'll mark them in some way to remind me not to use them for this application. Secondly, increasing the length of the isense resistor also appears to cure the problem, although obviously to the detriment to the brightness.

In answer to your question, with 2*AA I get ~350mA. I don't have any AAA's around to try out though.

thanks,

dan.
 

Latest posts

Top