Death of another BDR-209

[WizardG]

So, I was testing the water cooled module I flanged up for my wife's fabric cutter. Driver was @ 600mA as I received it from Jordan. All was well, but in my innate monkey curiosity I thought, "these GaN diodes tend to put out quite a bit more power when they're cold, I'll add some ice to the coolant."

At first it was working great. I don't have a calibrated LPM @ home but my improvised solar cell rig was indicating the output of the diode going up & up as the module got cooler. Then, COD. The diode LED'ed then went completely dark.

I think maybe I'll try an M-140 for cutting fabric. Hey Jordan.........

 

[DTR]

Which driver was it? The nano that I offer with the BDR-209 is a fixed 500mA driver. Also what was the power source for the driver? Can you attach a picture of your setup.

 

[WizardG]

Hi Jordan, I'll take a couple of pics when I get home from work (11M or so local time) The driver was whatever you used when I bought the module from you. I have no complaints whatsoever with the hardware. The driver is outputting 11 volts into the dead diode while consuming ~10mA. I think the diode went LED then died completely to an open condition. Power source was a NiMH pack that measured 5.5 volts open circuit. I ran the diode for quite a few hours this way cutting fabric. I didn't have any problems until I started cooling the diode close to freezing. (I wanted more power for cutting heavy velvet.) According to my crude solar cell setup (measuring current from the solar cell) the output from the diode nearly doubled before it died. LOL, I may have very briefly set a new record for single mode 405 output power.

 

[DTR]

I am guessing this may have been picked up before I started offering the nano and it is using an X-boost since you said it was set to 600mA. Depending on how it was being cooled if there was any condensation allowed to collect on the optics that could cause a back reflection which could kill the diode. Don't think low temps should have any detrimental effects on blueray diodes. Most diodes seem to like it. As for the driver measuring 11V means it is not on a load and over boosting the voltage to max which can damage the driver. I would probably not try to use that driver again. The way to test the driver would be to use a test load to test the current output as switching drivers require a proper load in this case with a boost a load that is higher than the voltage of the power source.

 

[WizardG]

If I recall correctly it says x-boost on the driver board. I'll test it with a high power violet LED and if it still regulates current I'll use it in a future build. There was no condensation on the optics. I've got them protected behind a sealed AR window to keep smoke from the fabric from getting to them (dear ghod cutting silk this way stinks! I knew wool would be bad but silk is the WORST!)

And yeah, the diode liked being chilled down. Liked it a little too much I think. If I take the results of my crude solar cell measurement at face value the output of the diode nearly doubled before it suffered COD. Live and learn. I'll be placing an order for a 2 watt M140 and driver with you shortly, I want to re-use the module. I've got the means to press a new diode in so no problem there.

Thanks Jordan!

 

[DTR]

The BDR-209 is a 3.8mm diode and the M140 is a 5.6mm diode so you would at least need a new copper mount but you could reuse the lens, spring, focus ring and back half. If you do try another BDR-209 the new nano is the way to go. It is specifically designed to be gentle to the 405nm diodes.:beer:

 

[WizardG]

All right then, I'll order another '209 + a nano. Thanks Jordan. Do you happen to know what the soft-start delay on the nano is? I tied into the 'knife down' signal on the cutter to gate the power to the laser on the x-boost but if the soft start delay is too long this might be an issue.

 

[DTR]

I always opt for the shortest delay. It should be the same as the X-boost you got from me.:beer:

 

[Benm]

At first it was working great. I don't have a calibrated LPM @ home but my improvised solar cell rig was indicating the output of the diode going up & up as the module got cooler. Then, COD. The diode LED'ed then went completely dark

This is to be expected though: The failure is due too to much optical power, not electrical power. Optical power increases as temperature drops when you supply constant current, and this fried your diode.

It can be a real problem when driving diodes that hard: It can also kill them when used in portable lasers that have been sitting in a car over night in winter or something like that, as long as the batteries don't poop out first.

 

[Lifetime17]

Hi,
I have two BDR-209's with a Nano set to 500mA and they are awesome. personally i have to respect the 405 x 16x they been put through some torture tests for some time.
if you want to play with current then buy a sled and pull the diode its a lot cheaper. jordan sells top notch items he is very reliable and stands behind all his products. And thank yo Benn for elaborating on the subject. Folks have been looking for the Holy Grail of output on these diodes forever . Some have been lucky and some not, as jordan said at one time its luck of the draw. now a binned diode is a little different .. The Nano diode is perfect for the BDR-209 16X diode. it can be set to 275mA,500mA or 1.1A by resistors bridged And not to mention its a buck or boost as well so go figure. I hope all ends well with your 405build but these guys are rally fussy and get pissed very easy.. I guess we will all have to wait till there is a 405 capable to do over 1w + with no worry nuff said... i love the 405 WL its awesome..

Rich

 

[WizardG]

I used the BDR-209 in my wife's fabric cutter for its tight focus. Even at less than a watt it's awesome! I've also read many of the stories of woe from those who have pushed the diode with current trying to get every last microwatt outta the darn things. I thought I'd try my luck chilling the diode to get a few more mW for cutting heavy velvet. As the case temperature dropped I was indeed seeing impressive power gain but the poor diode just can't handle that much light and.....COD. Oh well. After my conversation with Jordan (above) I'm gonna try another '209 at a slightly lower current and I'll keep the temp' above 10C. It's OK if it won't slice through the fabric quite as fast; it's a CNC tool and I can lower the cutting speed and just let it run all night and hand my wife the finished pieces in the morning if that's what it takes. It STILL beats the hell outta dicking around with the tiny and expensive knives the tool vendor sells. Google "Silver Bullet cutter" if you wanna take a look at what I'm using the laser for. I'll post a couple of pics of the water cooled module I flanged up (probably tomorrow morning, I work late) too.

 

[Lifetime17]

Hi,
That would be great with some pics still the diode is only a 3.8mm tiny so to speak . this might work at a lower setting and al so cooled like to see the results your getting .

rich

 

[WizardG]

Here's a couple shots of the water cooled module I flanged up for my wife's fabric cutter. The hardest part was getting the copper tubing wrapped around the module without kinking. Anneal, flex, anneal, flex, anneal, flex. When it fit the module correctly I just glued it in place with arctic silver epoxy. The extension tube on the front of the module let me load it into the cutter as a drop in replacement for the knife holders the tool came with. I tapped into the 'knife down' signal of the cutter to gate the laser power

 

[Benm]

There really is no need to watercool these things, unless you want to reduce the weight of the heatsink in a moving-laser application.

One thing to understand is that you are -grossly- overdriving these diodes even with 600 mA. I think the maximum rated -pulse- current is 550 mA for these things as they operate in the blu ray writer with fairly limited heatsinking.

In the drive mean time before failure is probably 10.000 hours or something like that, but running it at 600 mA CW might already reduce that a lot. By the time you run it at 900 mA CW this could reduce to only a few hours, could be okay for a pointer but not for any type of engraver/cutter/etc that runs for fairly long periods at a time.

 

[WizardG]

It ran for many hours @ 600mA with no issues. Only when I lowered the temperature below 10C did the diode suffer COD. And yeah, water cooling for this application is an absolute must. The weight needs to be kept to a minimum due to the design of the cutter, and with no coolant flowing the module would heat up too much for comfort in about a minute. I ordered a replacement diode from Jordan with a driver set to 500mA. We'll see how long it lasts this time! I'm hopeful that as long as I don't get greedy for mW and try super cooling the diode again it will run for a year or so.

 

[AngelG]

I wrote about some possible death reasons here: http://laserpointerforums.com/f38/first-casualty-dead-bdr-209-diode-90121-2.html#post1448816
Probably instead of cooling it, you may want to combine the power of 2 diodes. I remember seeing a guy who did it by precisely tunning the beams to be point in/over point at the surface, thus effectively doubling the power density. He had been using 2 separate modules. His DIY machine was "scanning" above the surface.
Moreover, I defend my opinion that at these currents you heavily over*** these diodes above their specifications here: http://laserpointerforums.com/f38/pioneer-16x-bdr-209dbk-88124-9.html#post1448818

Whenever I try to focus, I use 1% PWM and current near the Ith (so power is about 1% of 60mW) to be able to watch the points..