1.4W Laser functioning, but not as expected?

Hey all. I've got a problem with a 1.4W laser I bought from weidmark a while back. It runs off of 1 18500 and weidmark had it clocked at 5 minutes of runtime before the battery was giving out.

I never got to see it in its full glory because due to unfortunate circumstances the laser was turned on with reverse polarity for a split second, and then turned off again as soon as it was noticed that nothing was coming out. The good news is, on a full battery the laser still works and is VERY impressive and powerful... for about 20 seconds. After 20 seconds it drops to about half brightness and starts flashing. It continues this flashing for about 15 seconds, and continually grows dimmer until it just shuts off. This has happened probably 10 times and every time the battery takes a good 4-5 hours to charge, so it would appear that all the juice in the battery is indeed getting used in about 20 seconds.

Things I know:
The diode and driver don't show any signs of physical damage.
The driver has more than ample heatsinking and hardly heats up at all during this process.
The battery I'm using is an UltraFire rechargeable battery rated at 1600mAh, 3.7V, and is a protected cell.

I would like to figure out where the problem lies (battery, diode, or driver?) and possibly even fix it myself, if it is fixable. How can I test which component is damaged without further damaging the laser?

Thanks for your help guys.

Edit:
Please let me know if theres any other info you need to diagnose the problem.

Also: How is the entire battery being used if it is only running for 20 seconds at full power and doesn't heat up much. Where does that energy go?

Sounds like your battery cannot provide enough current or the driver is overheating..

Check the connections to the driver, and possibly the switch.

Protection circuits can also cause issues.

I'm assuming this is using flexdrives. The flashing is the thermal protection for the driver(s) kicking in, it's perfectly normal. You can add a heatsink to the driver (I think DTR had a thread on doing this, the coin heatsink mod or whatever) and that should help. Hope this helps.

It actually is already coinmodded, and doesn't heat up at all during the process. I think weidmark just followed DTR's guide on it.

Also: Weidmark had it running for 5 minutes off the same kind of battery. Do you think it is just the fact that it is a protected cell?

It could be the batteries protection cell kicking in, but this sounds much more like the flexdrive's protection circuit than anything else.

Right, that was what I figured due to the flashing. It seemed like a heat problem, but there wouldn't be a heat problem, its got a great heatsink and is coinmodded. I can confirm that it is a flex drive, its got the letters FLX on it. I'll post some pictures in a couple minutes. Thanks for your quick responses!

Would there be an easy test I could do to find out if it was a problem with the battery or a problem with the driver? I can probably borrow a regulated power supply from my college, or at the very least an ammeter/volt meter.

Does it blink slowly or does if flash rapidly?

If you do get your hands on a bench supply say 3A and over and simulate a 4.2V battery it should tell you if it is a battery issue.

It flashes rapidly.

So if I hook up the whole laser to the supply as is and give it 4.2 volts, and then see whether or not the same thing happens it would tell me if it were a battery issue or not. Right, that makes sense. So I would hook up the positive to the terminal inside, but where would I touch the negative pin to? Just anywhere on the case?

Also, how much would I need to worry about power supply fluctuations during the experiment? Is there a way to limit them, or will any scientific lab power supply be fine to use?

Sorry about all the questions, I just don't want to damage it any more than it already is, because it DOES still work to some extent, and I'm a noob when it comes to these things. Eager to learn though.

If it is flashing rapidly then my money is on a current issue. Most likely the battery or the connection inside the host like the spring for the battery positive contact if it is an 18500. You could upsize to a 22600 and get a little more capacity and it will not be as loose in the host. In my experience when the flex starts going into thermal protection mode it blinks slowly and when there is a current issue it flashes rapidly.

If it is a current and voltage limiting supply you would just set the current to max and the voltage to 4.2V. Just make sure it is rated around 3A and over.

Also try this. Sometimes with the 18500 you may not be getting good pressure against the contact points. Take the tailcap out and complete the circuit with a piece of metal putting a little force against the battery and see if it makes a more solid connection.

Awesome, so we've got it pinned down to a connection issue.

The battery is always very tight against both springs when I screw it on. It is actually kind of tough to get it in, so the two connections are very tight.

If i was going to do this test, though, when you say complete the circuit you mean just get a paperclip or something and bridge the gap from the negative terminal to anywhere on the case? I'm a little confused about how the negative side is connected in the circuit. Does it just flow through the entire case, across the coin, and directly into the driver somewhere? I see only one wire leading from the battery compartment/switch area to the driver (presumably for the positive terminal of the battery), so it seems like thats the only possibility...

Yep maybe use something a little thicker than a paperclip but yes the negative flows through the host and if it is set up like I normally do all my hotlights it goes from the case through the heasink into the diode and the case pin carries it back to the driver.:beer:

Ahhh, that makes sense. Alright, well I'm waiting for the battery to charge so I can do this test.

I'm pretty sure this is pretty much exactly like your hotlights coinmod build. Looking at the laser I only found one real difference which is that there are only two wires coming out of the siode itself, whereas yours showed three. Don't know if that is anything worthy of note, because apparently it is still a complete circuit. I wish I had a sautering iron so I could get in there, find the problem, and fix this thing...

Finally got the pics working:

Update: I did the test replacing the endcap with a wire and it ran fine for about 10 seconds (battery wasn't fully charged), but then entered the flickering mode and stayed there for another 5 seconds before I shut it off.

I think I'll try the regulated power supply test tomorrow, to see if its the driver or battery.

Crossing my fingers for a battery issue so that I can just order a new one...

Thanks for your help tonight guys, I'll update you tomorrow.

Just be careful when hooking to a P/S..

I work with my own boost driver and have little experience with the FLX..
So I cant say for certain.

You may want to double check with Dr. Lava for any known issue when
used this way.

I remember something with an old version.. They didnt like certain
bench supplies, even good ones. You'd be better off buying a newer
better brand of battery, or even 2 x 18650's in parallel to drudge up
some current.

If you try this blindly. you may be risking the driver and LD..

At least replace the LD with a test load to trouble shoot.

Meebs said:

Update: I did the test replacing the endcap with a wire and it ran fine for about 10 seconds (battery wasn't fully charged), but then entered the flickering mode and stayed there for another 5 seconds before I shut it off.

I think I'll try the regulated power supply test tomorrow, to see if its the driver or battery.

Crossing my fingers for a battery issue so that I can just order a new one...

Thanks for your help tonight guys, I'll update you tomorrow.

Click to expand...

Thanks to your warning I decided against doing the PS test. I just ordered another battery (at $3 per battery, its really not worth it to risk the laser hooking it up to a PS, haha).

It should come within a few weeks. I'll let you guys know how everything turns out. Crossing my fingers that I don't have to replace the driver.