2D square barcodes on LD cans

[ogoun]

I've ID'ed the 2d barcode type used on the BDR-205 diode. It's Datamatrix (ECC200), format 1 (12 x 12).
This should actually be easier to read than the micro QR codes, since more readers support it. Again, it looks highly likely that it is a 10 numeric digit code, with heaps of error correction. These look a lot like batch numbers, though the different barcode styles suggest different manufacturers...

The marking style looks like laser marking, which has a kind of symmetry to it, since the mark is being lasered onto a laser

BTW.. Anything with features bigger than about 800 nanometres (0.8 micron), I can image at very high resolution Might be interesting to look at a laser diode chip after it has died from overpower to analyze the failure mode...

I am considering depackaging my dead Nichia diode and photomicrographing the diode die itself in the hope of finding a manufacturer's marking of some kind.

Any thoughts on all of this?

Pete.

 

[ogoun]

OK, found a tool to manually create a monochrome matrix that represents the 2d barcodes, and I have found that all of them have this 10 digit format number.

The earliest diode (pre bluray/hd-dvd diodes) is the Nichia 25mw out of the Coherent Radius.
It's number is 1000692694

The next diode is the PHR-803T diode.
Its number is: 1101039184

The next diode is the unknown (supposed 8x) kendall diode.
Its number is: 1101313481

The next diode is the supposed 120mW Nichia (new, not harvested from a drive), purchased off epay july 2009.
Its number is: 1102322470

The next diode is a BDR-205.
It's number is: 1105550324



Certainly looks like a pattern is developing... This also implies one single manufacturer, and I think we know who that is


Pete.

 

[lazerov]

Hopefully laserov will come by again else we might just have to bug him by pm. :san: No idea what he used to make the conversion.

Well, I'm a bit late but I'm excused since I live in different time zone
Anyway, the method I used is quite simple:
1. Crop only the barcode.
2. Play a bit with brightness/contrast and levels until you achieve decent contrast.
3. Convert to black and white. Some manual editing may be necessary.
4. Resize to 17x17 pixels (this is the barcode size) using no filtering.
5. Resize to some bigger size (again - no filtering). Add some whitespace if needed.
I used GIMP to do this.

I am considering depackaging my dead Nichia diode and photomicrographing the diode die itself in the hope of finding a manufacturer's marking of some kind.

I'll be surprised if there are marking on the diode die but it's certainly not impossible. But, even seeing the die from up-close would be interesting.

The earliest diode (pre bluray/hd-dvd diodes) is the Nichia 25mw out of the Coherent Radius.
It's number is 1000692694

The next diode is the PHR-803T diode.
Its number is: 1101039184

The next diode is the supposed 120mW Nichia (new, not harvested from a drive), purchased off epay july 2009.
Its number is: 1102322470

The next diode is a BDR-205.
It's number is: 1105550324

Good work! No one has done this before, as far as I can remember.

I admit, I was hoping for something more useful than just a number ... maybe CW and pulsed power output LOL :san:
But it may be possible (only possible) to ask Nichia about one of the numbers (the BDR-205 of course) and maybe they will care enough to provide some data about the diode .

 

[sk8er4514]

You guys should check this out, http://laserpointerforums.com/f38/8x-sled-diode-47634-2.html#post657703

would be cool to read that bar code, see if it is the same as a GGWH20L barcode.

supposedly from a LG BH08 sled but data is telling us otherwise. Diode appears to test like a strong 6x blu ray.

-Kendall

 

[HaloBlu]

But, even seeing the die from up-close would be interesting.

Die photos ....

 

[lazerov]

You guys should check this out, http://laserpointerforums.com/f38/8x-sled-diode-47634-2.html#post657703

would be cool to read that bar code, see if it is the same as a GGWH20L barcode.

supposedly from a LG BH08 sled but data is telling us otherwise. Diode appears to test like a strong 6x blu ray.

-Kendall

This is the barcode from the supposed 8x diode (clickable picture):

It may have 1-2 errors since the starting picture is a bit unfocused.

Die photos ....

I missed this thread. Thanks.

 

[sk8er4514]

This is the barcode from the supposed 8x diode (clickable picture):

It may have 1-2 errors since the starting picture is a bit unfocused.


I missed this thread. Thanks.

nice thanks. I'll try taking some macro shots of a GGWH20L diode barcode when I get some more in (currently out of stock, more are in the air on the way though).

-Kendall

 

[BShanahan14rulz]

This thread is going just where I was hoping for it to go! Hopefully I can crack open my cheapo blu-ray and see if even that might have a barcode on it under all the glue...

 

[LarryDFW]

laserov;

Here is another photo of Kendall's "8X" barcode which is focused better at the top:

I think your picture is correct, with the exception of the adjoining outer lines that touch the square target. I think the top line is generally alternating black dots with spaces.

Great job on decoding these barcodes.

Thanks for your effort.

LarryDFW

 

[BShanahan14rulz]

 

[lazerov]

I think your picture is correct, with the exception of the adjoining outer lines that touch the square target. I think the top line is generally alternating black dots with no dot.

It would be great if we could decode these barcodes.

Are you using a program to generate the images ?

Yes this is the dot I had some doubts about. Here is the picture with that dot changed:


I already explained how I convert the pictures: http://laserpointerforums.com/f38/2d-square-barcodes-ld-cans-48101.html#post660510

 

[ogoun]

If people can get this quality of monochrome image, I can decode them.

The number on the supposed 8x diode is: 1101313481

I have edited my numbers list (earlier post in this thread) to add this one. A clear pattern is emerging.


I am wondering if I should start a new thread to track these numbers, possibly in conjunction with the 8x and 12x murder/testing threads...

Pete.

 

[Dvorhagen]

This is really very cool, interesting stuff! Noob question, though: aren't the data sheets on these diodes readily available?

 

[ogoun]

Unfortunately, there is no definite way to ID which of the Nichia commonly/publicly available datasheets (if any) apply to these harvested diodes. There are no part number markings of any kind (apart from the barcode), either inside or out on these diodes.


Pete.

 

[sk8er4514]

GGWH20Lbarcode.tif

Here's a close up shot of the GGWH20L code. I had to clean off the stupid green dot that LG put on there, looks like with a sharpie. A dab of alcohol got it off.

You'll have to click the link to see the pic, it's pretty huge and I already cropped the shot to be of the diode only.

So my question is if this is the same code as the LG BH08 code?
Edit: nevermind it is obviously not. You can tell by just eyeing it.

-Kendall

 

[ogoun]

After a bit of investigation, it appears that the 10 digit number is not a batch number, but a unique serial number. I am fairly sure the more significant digits indicate a product line (e.g. their standard part numbers, proprietary builds, etc), and the leas significant 5 or so digits are the serial digits within that group.

This info should help us identify LDs with similar characteristics, especially when looked at in conjunction with the various testing regimes completed and currently underway here on this board.

Pete.