Faulty Driver?

Ggmesquita

Newly Enlightened
Joined
Jul 23, 2009
Messages
48
Location
Sao Paulo, Brazil
Hello,
I`m new to this addiction, but I`ve always liked electronics and PC hardware. The community and the forum are awesome! Congratulations! lovecpf
The thing is I recently bought 2 pieces of the Alpha G10 Dropin to use with the G2 Led (one for me and the other one for my dad).
I`m using 3V 900mAH RCR 123 from Tenergy and I noticed a strange thing going on. One of the modules runs fine for about 40 mins before shutting down, but the other (Which should be identical) can`t rum at all, due to the battery`s protection. It runs for 1 or 2 seconds and shuts down, making it clear that the second module is drawing more juice.
When I use primaries, both modules run fine. The hungriest one runs for 15 minutes less than the other and is of course, brighter and much hotter.
My question is, is this difference normal? Could this indicate a faulty driver on the hungriest one?
I`m waiting for my 3.6v AW`s to arrive together with an Ultrafire charger to do futher tests.
Thank you!
Gg
 
1st of all WELCOME TO CPF!!! I am rather new here too but I quickly found out that using their google search on ANYTHING light related before purchasing it usually gave me a pretty good feel for how well it would work. A little too late i know...it was for me too but the below link is a rundown of what you bought :sick2: So anyway...its definitely the equipment, the drop-in should not be pulling enough to shut off the batteries that fast. Its not even an R2 LED so the current should be less than 350ma. What you have is a bad dropin that probably isn't repairable. I would contact arctic traders about it and see if they will send you a new one before you leave ebay feedback. And I am so sorry your first CPF experience was identical to mine:welcome:


Alpha G series Lights
 
Hmmmm....

I would use a multi-meter to measure the current to each module on a pair of regular CR123s. Account for some mild voltages losses under a load and calculate a rough total power consumption. They *should* run ~3.0-4.5W total power consumption if they are typical cree LED drop-in modules.... he variation in performance and behavior from one module to another speaks volumes about the quality control.

The interesting thing to me, is that, you are able to actually SEE a difference in performance from one module to the other. It usually takes AT LEAST about a 30% difference in output to be able to detect a difference with human eyes.... This suggests to me that there may be more than one thing going on here.... The lower powered module may be underperforming for some other reason (less efficient driver?) while maybe the higher powered module is actually the one working correctly... Or they may both have problems that are totally separate... ORRR>...

Another possibility, is that these units do not have drivers at all, but instead, just a resistor, and the folks who slapped these together for re-sale didn't take into account variations in LED Vfs, and just put the same resistor value in all modules.. This would result in significant differences in performance from one module to the next.

Were these modules sold as being compatible with the "3/9" series of SF style lights as well? (like the G3/9P/C3)... or did they include any literature about their input voltage compatibilities? I just want to make sure you aren't going to fry your led drop-ins with the pair of 3.7V cells... If there is nothing that indicates to you that the module is compatible with higher input voltages, then the theory I have proposed above about the lack of regulated (direct drive resistor instead) may in fact be true, and may be how they are getting 10K lux out of some of the better performing ones, and would also explain the massive heat buildup you describe on the higher powered module (the LED could be over-driven, which would make it pretty bright, but at the same time, make a lot of heat, combined with the huge amount of heat from the resistor).

If it were me, and I were there, I would be taking measurements of the current consumption with different voltage power sources to see what happens. If the current consumption rises with increased input voltage, then they are just using a resistor, and probably will not be compatible with a pair of 3.7V cells, if they use proper regulation, then they will consume less current at higher input voltages.

There are lots of possibilities, and unfortunately, I don't think very many if any CPF members have that specific module to help solve your issues as I think most CPF members that are in the market for a cheap LED drop-in just pick em up of DX, since ebay seems to be taking a lot of those same cheap modules, slapping fancy names and claims on them, and reselling them for ginourmous markup.

-Eric
 
Hey Erick, Thank you very much for your time.

I`ll do that, I`ll measure both currents to see what is going on with each module. I hope the higher powered one is not too over-driven. I live in Brazil and it would be too expensive to RMA it and also it would take too long.
As for the specs, The seller said the following:

"...Alpha has a very advanced voltage regulator circuit built inside each unit which allows their bulbs to run at one of the widest ranges in the industry and make them of great value to the military, 3.6 to 15 + volts...."

Though the validity if the above is questionable, I am assuming it is OK to run it on 2X 3.7V RCR 123. And I hope it will run stable on that setup, as it should draw less current.
My 3.7V batteries were not yet delivered, so for now I am stuck on 3V Tenergy Rechargeables and Surefire primaries. I`ll measure the currents with both.

It is really interesting that I am able to actually SEE a difference in performance from one module to the other, but that is OK as far as this is not an indication of a faulty module.

Thanks to CPF, I already ordered a M30 and a 6P, together with a 18650 rechargeable from AW (the new model). I will bore the tube and let it shine!:naughty:

Thanks! I will post the current measurements when I have them.

Gg
 
hey ggmesquita, sorry my first post was not so helpful, mdocod has come excellent suggestions, and as i am not that familiar with the terminology and trouble shooting I can't be that helpful. I have; however, run G10 on the 3.7V RCR123s without any issue of overheating or anything of the like. I didn't notice any boost in output, but definitely did not notice any detrimental heat buildup in my G2 head either. I left my multi-meter at a friends place after i finished fixing their pool, so i don't have measurements about the current with rechargeables vs the primaries...sorry.

One other possibility for these drop-ins:

I orginally ordered a G9 and got shipped the G10 as a "free upgrade" I wonder if they put whatever LED they have available at the time into the drop, It may be possible you have 2 different lamps, hence the different output? The pills are pretty hard to open to actually determine the LED type, but it is a possibility.
 
Last edited:
Did you eliminate the batteries as the culprit? I don't think you mentioned troubleshooting them, or taking any measurements of them. You might just have one bad cell.
 
SFG2Lman ----> Thanks man, all the information I received was very helpful. I was able to open both modules and the LEDs look exactly the same. (Cree, R2 bin I think). So I am assuming the difference is due to something else (like quality control as said by mdocod).

Marduke ----> Yes, I am sure the problem is not the batteries, cause I have 4 pieces of them and this happens regardless of what combination I try. Also, as I said earlier, both, the weakest G10 module and the original G2 LED run fine on them.

I still didn't have time to measure the currents, but I will do it tonight when I'm back from work. Then I am going to post the results here.

Thanks everyone!

Gg
 
you go the actual pill open? if you post some comparative pics of the drivers themselves we may be able to see a difference in the circuit, if not, resistors are always manufactured with a tolerance, some cheaper ones are +/- 10% so if one of your drivers used the high end of the 10% and the other had the low end, it would make up to 20% difference in output, not saying that this IS the cause, just another possibility. so yea poor QC
 
No, not the actual pill. I simply removed the reflector so that the business end of the LED would be exposed. It was not possible to see the circuit. Will post pictures tonight together with the current measurements.
 
OK, here we go. I`ve taken pictures as I measured the current used by both modules.

DSC01741.jpg

Those are the modules side by side.

DSC01742.jpg

You can see they look like your regular R2...

DSC01745.jpg

Here you can see the battery`s voltage.

DSC01747.jpg

This is the higher powered one.:eek:oo:

DSC01748.jpg

Now for the lower powered one.:eek:oo:

If this is correct, the higher powered one is sucking 22.2W:eek: while the lower powered one is getting 12.6W. Is this even possible in the first place?

Another thing is that I can confirm that it is totally possible to actually SEE the brightness difference between both units. The brightest one throws further and has a greater spill. The same unit also gets VERY hot. While the other one only gets warm.

So what you guys think on all that?


Thanks!
Gg
 
:confused: thats a good one, it honestly looks like one is running direct drive (no driver) and the other has a constant current regulator...perhaps an internal short on the high powered one is causing it? I'll recommend my Ford repair kit
:)xyxgun: )for those modules. Without ripping the pill apart and testing each component on the driver individualy...its just a guessing game...and there may be no way to put it back together again... Trust me, if you buy a different drop-in you will be amazed at the better output...Sorry I keep coming to the "spend more money" conclusion, but without a complete teardown its impossible for me to guess
 
Something seems very wrong with your multimeter readings, if the meter is set to read amps neither module should draw that much current.
I just can't make out the settings on the meter in your photos.
Norm
 
Yeah, very odd.

The meter is set to the 10A range. the reading should be correct.
I forgot to measure the original SF module as reference. I'll do that tonight.
At this voltage (let's say 6V) what current reading should I expect on the SF module?

Gg
 
Just guessing (I've never had a SF module) maybe around 500mA at 6 volts. Should be less than 1A.
Norm
 
OK, I`ve just measured the current with the original SF P60L-WH module (G2`s original). The result was:

0.82A @ 5.84V = 4.7W

What do you guys think, does this sound OK to you or my multimeter is going nuts?

Can anyone test this module to see how it goes? Please, or I am the one who is going nuts here!

Gg
 
Looks like your P60L is operating correctly, or close to it... Both of your G10s are operating VERY incorrectly if those readings are true, but since your P60L module is reading at least close to what it should be, I think we can say for sure that both of your G10s are WAY overdriven...

I'm not surprised you are talking about a heat problem, what does surprise me, is that the modules still work at all, they should have self-destructed the LED long ago, unless those readings are incorrect.

Keep in mind that the CR123s you are testing with appear to be running a little on the low side (2.9V open circuit indicates that they are probably fairly discharged, no surprising since your G10s are eating them for lunch it would seem).. You can't take the open circuit voltage and multiply by the running current to figure out power consumption, because the cells will sag under a load, measuring voltage while the circuit is running would be required to get a true total power reading, this can be tricky to say the least :) . So more realistically, your P60LED module is probably running closer to ~3W give or take.. (which is where it should be)... The 0.8A draw of the P60L, provided this is an accurate reading, tells us that the input voltage has dropped enough for it to start drawing more current to offset the voltage loss. Under normal circumstances I think it's more like 500-600mA on a pair of CR123s (this from memory, I could be a bit off here)... Since it appears that your G10s are just operating in direct drive (as I had suspected before, even though they aren't supposed to be) we know that the power supply us sagging to the Vf of the emitter at that current, which is probably around 4V, the variation from module to module in this case is a result of the variation in LED Vf, so your actual power consumption is more like 15W and 9W respectively. Regardless, this type of heat and load is all around not safe. The fact that your 3.0V cells won't light up the higher power consuming module is no surprise, as the voltage would immediately sag low enough to kick in the over-discharge protection.

So...

I have some bad news:

* You should not use those 3.0V rechargeable cells anymore. If they have been used to drive even the lower powered module, they have undoubtedly been over-taxed far too much. Even a few discharges at those rates (~4C) would be enough to make them a liability. You should dispose of them properly.

* You also can't use your new 3.7V cells when they come in with either module, as that would be very unsafe as well. More than likely they would trip their current limit and not work on either module anyways, but either way, don't do it.

* You have experienced illumination from severely over-driven LEDs, it's likely that when you get correctly working modules that you may not be all that impressed anymore... I have a hunch about how they are getting that "10K lux" figure now... On a pair of CR123s in the 2+ amp range, an R2 would produce about 10K lux at 1 meter in a 26mm reflector. properly performing R2 modules in this size class will perform ~5-7.5K lux depending on drive level.

*The company that sold you these modules has put you and your home at risk of fire/explosion and hydrofluoric acid poisoning. Discharging CR123s at nearly 4 amps, discharging lithium cobalt's at 4C.... Lots of danger here, you are lucky you haven't had an incident. I would be writing a really nasty email to them, and informing ebay of the dangerous products they are selling.

A few comments:

*When you get your 3.7V cells, it's very important that you do NOT use the charger that you used for your 3.0V cells to charge them. It will over-charge the 3.7V cells to dangerous levels. (the 3.0V cells you have are actually 3.7V cells with voltage "bucking" circuits/diodes in them, charging them requires higher than the normal 4.2V cutoff to over-come the voltage bucking circuit in reverse, so the charger actually terminates at 4.4-4.5V, very dangerous for other cells).

* you may be able to use your G10 modules in a configuration where you have a single 3.7V cell (like an 18650 or the like) driving the module. You would want to take a current measurement on a fresh cell and make sure it isn't any higher than about 1.2A to make sure it can run like this safely, but this is the only way that I would suggest using these modules without some sort of fix...

*If you're handy with a soldering iron, you might consider gutting the LED modules you have there and installing a proper driver. Probably would be best to install new LEDs as well, as they are very likely nearly toast at this point anyways...

-Eric
 
Last edited:
Hello Eric! Thank you again for your time and effort. Very useful info as always.

First of all, despite the reading with the P60 being close to what it should be, I will measure everything again using my friend`s pro multimeter. Mine is cheap stuff. Just to eliminate it from the equation.

Second, thank you for your concern. I will definitely not use any of these batteries again, once I finish testing. Nor will I use the new 3.7V batteries on these modules, or charge them using the same charger. I already ordered an Ultrafire charger for them.

I will write an email to the seller and another one for ebay, but only after I am 100% sure about what is going on here.

Regarding the fact itself, in another thread ( http://www.candlepowerforums.com/vb/showthread.php?t=222550 ) there is a user with the exact same problem (900mAH Tenergy batteries shutting down when using a M60). Based on this, it is feasible to say that my less powered module is drawing less current than a M60, correct? If it is OK to say that, there is still a chance that I have a bad reading here.

Anyway, instead of guessing, I will measure it again tomorrow using a GOOD multimeter. We`ll see. Oh, and I will use fresh primaries this time.

I hope you are wrong about me being disappointed when my M30 arrive... :(
 
hey gg I just got my meter back and measured the current, i'm pulling 860ma at the tail in a surefire G2 with metal bezel and the Alpha G10 Supra dropin from arctic traders on ebay (I just wanted to make absolutely sure we have the same setup) I can't understand why yours would be so much higher unless there is a short somewhere in the driver, or if its just a big resistor in yours, sorry once again that I couldn't be more helpful :shakehead
 
Top