24X Liteon iHAS324-32Y - Successful harvest and implementation

[siqueirah]

Well, with this whole 445nm fever going on, and red is still a beautiful color. I really can't resist stopping and looking at these pics. Really good, by the way.

Nice build !

 

[Bionic-Badger]

That's some great work, and I really appreciate the fine photography.

 

[flogged]

Excellent writeup.. and even better photos!!

There seems to be a trend, and it's not good for us DIYers.. manufacturers are making it harder to extract these diodes. It used to be most of these sleds had closed can diodes that were press-fit into a heatsink. They could usually be easily be extracted. Now more and more of the diodes are 'open can' (saves cost, but not good for DIY builds.. you don't want dust accumulating on the facets of the diode IMO) and are housed in armor-like heatsinks, making extraction much more difficult.

You wisely choose to file the heatsink down rather than attempt to extract the diode proper. A tedious bit of work no doubt

I doubt the diode perished because of your power supply.. you were driving it too hard (450mA). The power of the red diodes in these DVD drives has not increased the past few years. Most of the development is now on t'BluRay' diodes. From what I've seen the majority of these red diodes are not good for more than 1/5 watt power (~200mW), at least if you want a reasonable diode life. Yes you can boost the power, but in nearly all cases this will result in very short diode life. Your initial 200mA drive current was more reasonable.

One way to determine the max safe power is to calculate the 'slope' of the diode at various power levels. You need a DMM and LPM to do this.

For example, Hemlock_Mike did some testing on a DVD diode being offered by StoneTek:
LPC-815 20X DVD burner diode out of sled [LPC-815_1] - $12.99 : StoneTek Electronics, Your Laser Parts Store

See this thread:
http://laserpointerforums.com/f50/new-loc-laser-diode-review-52240.html

Anyway, he determined these numbers for the diode: (pretty typical number for these diodes)
Drive Current (mA): Power out (mW)
100: 56
150: 92
200: 134
250: 180
300: 210
350: 226
400: 240
450: 240
500: 237
550: 215

To determine the max safe power, look for the most efficient point of operation. Divide current consumed by power out..

Current (mA): mA/mW (mW per mA)
100: 0.5
150: 0.61
200: 0.67
250: 0.72
300: 0.7
350: 0.64
400: 0.6
450: 0.53
500: 0.47
550: 0.39

Now you can see the most efficient point of operation is between 200-300 mA of current. The lower the better for long life.

 

[jupiter8]

Hmm... thanks flogged, good info and it makes sense. I reckon maybe I did push it too hard. I'm learning and I'm already cringing when I think back to the way I treated diodes in earlier builds (attempts).

I've got 3 LPC-815 diodes on order from stonetek as well as some modules from aixiz, and 2 drivers here ready and waiting for them, so I'm looking forward to playing with those. I'll be much more realistic with what I expect from them and won't push them so hard, based on the info above.
Cheers.

 

[Benm]

Sad news.... RIP 24X diode. Gone. Dead. Kaput. I suspect I killed it by driving it with a STUPID little multi-select transformer power supply. I'm angry with myself... this is a crappy non-regulated power supply. I'm certain that's what did it, a spike from that. I'm still grieving the loss of my baby... :cryyy:

I'm going to try again. The project was so successful up to this point. I'm SURE I had adequate thermal contact, enough heat sinking and was driving it at a conservative current... the only explanation is a voltage spike from the transformer... (?)

Go for better current sources.. good desings will keep current to what you selected regardless of input voltage - at least until you manage to thermally blow out transistors

 

[qumefox]

There seems to be a trend, and it's not good for us DIYers.. manufacturers are making it harder to extract these diodes. It used to be most of these sleds had closed can diodes that were press-fit into a heatsink. They could usually be easily be extracted. Now more and more of the diodes are 'open can' (saves cost, but not good for DIY builds.. you don't want dust accumulating on the facets of the diode IMO) and are housed in armor-like heatsinks, making extraction much more difficult.

Or they just plain switch to packages that are hard for DIYers to make use of. For example I have some diodes on my bench at work right now that I pulled out of an A-open burner I think it was. It has the photodetectors integrated inside the can with the diode. The freaking things are rectangular, about a half inch wide and have 12 pins! I ended up decanning one just to figure out the pinout. I had a 50/50 shot of it being the IR since both were identical other than the markings, and google gave me nothing on those. Turns out the one I decanned was the 650. It needed it anyway, as it had some kind of diffraction grating integrated in the can. I had it lasing earlier today at about 150ma and holding up a collimation lens to it but in the weird package, i'm not sure what good even messing with it will do.

In all honesty, these packaging changes have nothing to do with us, and everything to do with profit margins. As long the 5.6mm TO packaging stays cheapest, that's what they'll use. Otherwise they'll switch to the cheapest way to manufacture their product to extract the highest profit margins.

 

[Benm]

I dont think they care about us at all - there are quite a bunch of people playing with lasers, but compared to the number of laser sleds and drives sold its all negligible.

5.6mm is a conventient form factor mainly because that its what fits in aixiz modules, which are cheaply available. If the industry decides on using different packages from now on, there will be different modules to mount them in. I suppose the 3.5mm round diode housings are a good candidate to begin with.

Open can is something we should get used to - in an optical drive the diode is installed such that there is little chance of dust etc becoming a problem. It should be much of a problem for builders either - the only thing is to get rid of the plastic lens threads and springs.

I used aixiz glass lenses with a layer of teflon (gastec) tape around them to make a nice and secure fit.

Pieces of plastic coming off due to the spring are a problem for diodes in a can too btw - they can land on the output window and melt to it. The diode might still work, but its use is gone.

 

[LASERBOTICs]

Can you provide the Pin out for this particular laser ???

Thanks,
Russ

 

[Boosted98gsx]

Sorry to bump, any luck with running these recently? I picked up 2 of these same diodes recently, and I want to make some good strong hand held lasers with them.

 

[jupiter8]

Boosted - no I haven't been inclined to try it again. I have 3 LOCs sitting here waiting for homes and I don't think the 24x are much (if any) more powerful than a LOC, but much more annoying to mount.

 

[DTR]

Got one of these flat diodes mounted in an Aixiz module. I just ground off the back of the Aixiz module and rounded the edges of the rectangular heatsink they are mounted in and used thermal adhesive to secure it in place. It was actually a very simple process if you have access to a grinder.

I had an acrylic lens screwed in to keep it straight while the adhesive set and it worked out well. Then I hooked it up to a groove driver set to 350mA and put it in a host. It ran for about 1 min giving me a max of 196mW with an acrylic lens. Then it went LED. :cryyy:

I am wondering if anyone thinks it died from too much current or because the rectangular heatsink it come in does not provide very good heat displacement causing it to overheat?