DIY red burning laser (DVD diode)
b...@fractalfreak.com
These really are easy and fun to build.
I'm using a Sony 16x DVD burner LD. These can be reliably
pushed to between 200-300 mW output (how long they will
last is another story, they are rated at 80mW CW, this
is using about 140 mA of current. Most people are driving these
with between 200-300 mA of current.
You can get the diodes here for about $20.00/each:
http://sales.stonetek.org/index.htm
They also sell inexpensive driver kits and housing for these
diodes. Assuming you've got some basic electronic skills,
know how to solder and use a multimeter, you can build
one of these for $35-40 in parts.
See this thread for more info on a simple DIY driver
circuit:
http://www.laserpointerforums.com/forums/YaBB.pl?num=1185701612
I built the driver using radioshack parts- the parts came to about
$20.00. I'm using a Meredith LDC1 collimator and mounting block
for the optics and heatsink. I'm using the acrylic lens which
makes for a tight focus. These parts cams to about $30.00
see:
http://www.mi-lasers.com/cgi-bin/shopper.cgi?search=action&keywords=diode_optics
Finally the diode was another $20.00. So the parts came to
$70.00, plus the work of throwing it together.
The Meredith housing is meant for 9mm diode cans (the Sony is 5mm)
so it's necessary to press fit the sony diode into a little brass
adapter. However I found that while focusing the Meredith
assembly the diode would squirm and rotate, not staying centered
in the optical path. So I ordered some thermal epoxy to fix this:
http://www.arcticsilver.com/arctic_silver_thermal_adhesive.htm
After getting the diode snug and flush in the adapter I applied
a few dabs of the thermal epoxy to get assure good heat flow
from the diode the the adapter. The adapter in turn I then
thermal glued to the back of the case housing, centered. This
worked perfectly, I can now focus the diode and it stays put.
The Meredith lens will focus from infinity to less than 8 inches
with about 180 degrees of turning.
I metered this one up to 240mA. The consensus at LPF is to
keep the current at or below 250 mA unless you've active cooling
or a large heatsink. 240 mA of current will likely put the optical
power over 150 mW.
The pointer will pop balloons out to over 40 feet and light wooden
safety matches out to 7-8 feet. With the focus adjusted up
close you can burn a lot of things. Very impressively bright
at night.
Some pictures...
Testing the regulator circuit in the breadboard. I pushed the
diode up to spec (~140 mA of current) which puts it around
80 mW output.
http://www.fractalfreak.com/wx/DIY_red2.jpg
Completed laser brick:
http://www.fractalfreak.com/wx/DIY_red2_5.jpg
Metering the current while adjusting the rheostat after
getting things working. Unfortunately I was only able
to meter up to 200mA of current before the pot maxed
out. Another 10 ohm resistor allowed me to get to
240mA. I stopped here. I'm interested to see how long
the laser will last 100mA over spec.
http://www.fractalfreak.com/wx/DIY_red2_1.jpg
Closeup of the innards. Radioshack parts + box,
4 AA nimh batteries, diode in Meredith housing+
optics.
http://www.fractalfreak.com/wx/DIY_red2_4.jpg
First task, burning a hole through the top of the
case from 8 inches.,
http://www.fractalfreak.com/wx/DIY_red2_3.jpg
Projecting the spot onto the ceiling of the bathroom.
It's that bright.
http://www.fractalfreak.com/wx/DIY_red2_6.jpg
Overall I'm quite happy with the pointer. Curious
to see how long it hold up before the diode pops.
-Eric
Lostgallifreyan
62058a...@q39g2000hsf.googlegroups.com:
> Overall I'm quite happy with the pointer. Curious
> to see how long it hold up before the diode pops.
>
Not long, if you're seriously trying for between 200 and 300 mW out of a
diode rated for 80 mW CW. People have tried for months and failed. I've
watched their discussions, and tried some diodes myself. Unless the diode
is rated for at least 120 mW CW, you won't reliably get over 220 mW and
you'd be lucky if it lasted a week.
No-one's going to buy on the strength of an overinflated claim, no matter
how well it appears to be backed up with apparent research. After all, they
need ask just ONE question: If you can really do this, why didn't the
makers of DVD drives also notice and start selling 24 speed DVD writers?
They haven't, and there's a damn good reason why not. They can't do it. And
if they and the diode makers can't push this performance, no-one else can,
certainly not with the same diodes, anyway. Unless you have some totally
unique supply, the likes of which the big industries would pay millions to
access, you can't do it, and there's no reason for anyone to believe that
you can.
Lostgallifreyan
news:Xns9A20E1B3565...@140.99.99.130:
> If you can really do this, why didn't the
> makers of DVD drives also notice and start selling 24 speed DVD writers?
Actually, I just checked, they have, just... HP have one now. But if you
can reliably get 200 to 300 mW CW, you could be punting out 32x, maybe even
40x write speed drives. The current 20x drive diodes can barely reach 300
mW in their pulse ratings, so CW is obviously out of the question.
Lasers In Mind
that your $10 laser is going to add up to
about $6,500
Go back to launch pad my freind
b...@fractalfreak.com
> b...@fractalfreak.com wrote in news:28a46d90-4168-4ae0-93dc-
>
> > Overall I'm quite happy with the pointer. Curious
> > to see how long it hold up before the diode pops.
>
> Not long, if you're seriously trying for between 200 and 300 mW out of a
> diode rated for 80 mW CW. People have tried for months and failed. I've
> watched their discussions, and tried some diodes myself. Unless the diode
> is rated for at least 120 mW CW, you won't reliably get over 220 mW and
> you'd be lucky if it lasted a week.
>
lol. We'll see.
The reason these diodes can be pushed to 200-300 mW
w/out getting COD is they're designed to be good up
to 250 mW, but only pulsed, not CW. Apparently if
well heatsinked and not pushed with more than 250 mA
of current (which should give an output of at least 180mW).
they have a reasonable life as a pointer.
Look at the discussion over at LPF. People are reliably
driving these diodes with 300mA of current and using
them for pointers and they are not dropping like flies.
Getting about 200mW while consuming 300mA of current
is not unusual for the Sony 16x diodes.
I'm not going for 200+ mW, I'm hoping for 150+ mW
and using 240mA it should be at least that. I was hoping
by using thermal glue to enhance the thermal path
for heat to exit the diode I might enhance its life a bit.
I don't usually leave it on for more than a couple of minutes.
I'll be keeping track.. even if it lasts a few 10's of hours
that a lot of use from a pointer. Fun while it lasts.
b...@fractalfreak.com
Here's a datasheet for the diode (sony sld1236vl):
http://pdf1.alldatasheet.co.kr/datasheet-pdf/view/197026/ETC/SLD1236VL.html
lava
> Here's a datasheet for the diode (sony sld1236vl):http://pdf1.alldatasheet.co.kr/datasheet-pdf/view/197026/ETC/SLD1236V...
The markings on the diode (JL) lead me to beleive that is may be this
one:
http://www.zet.biz/Laser/data/SLD1239JL.pdf
Bob
>http://www.mi-lasers.com/cgi-bin/shopper.cgi?search=action&keywords=diode_optics
>
>Finally the diode was another $20.00. So the parts came to
>$70.00, plus the work of throwing it together.
>
>The Meredith housing is meant for 9mm diode cans (the Sony is 5mm)
>so it's necessary to press fit the sony diode into a little brass
>adapter. However I found that while focusing the Meredith
>assembly the diode would squirm and rotate, not staying centered
>in the optical path.
Thanks for your post, Eric. I've been accumulating a few parts,
thinking that I'd eventually put together a small red or IR laser for
some experiments (not for burning, but high power if possible). Sounds
like Meredith has some useful mounts and lenses.
The diodes that I found are -supposed- to be 150mw, but I haven't been
able to find any data on them. Part number is J8Y4M4C. Anyone know
anything about it?
The diodes are packaged with a lens that says "Achromatic Lens, 12mm x
6mm, AR Coated."
If this gets to be a pain, I'd just buy a completed assembly
somewhere, if they exist. Looking for something compact with about
150mw or more. Ideas?
Lostgallifreyan
news:084904f0-7d0f-4ef4...@21g2000hsj.googlegroups.com:
>
> Here's a datasheet for the diode (sony sld1236vl):
> http://pdf1.alldatasheet.co.kr/datasheet-pdf/view/197026/ETC/SLD1236VL.
> html
>
You still don't get it. Pulse mode!
http://www.sony.net/Products/SC-HP/cx_news/vol37/lsd1236vl.html
Look there if you still refuse to believe, that's Sony's own.
Lostgallifreyan
21f612...@k39g2000hsf.googlegroups.com:
> Look at the discussion over at LPF. People are reliably
> driving these diodes with 300mA of current and using
> them for pointers and they are not dropping like flies.
>
When people want to boast about their successes, they're not going to shout
about their failures. I don't care about a forum of people playing, if this
were a real claim, it would be made my the industry that makes diodes and
drives! They don't leave the big stuff for newbies to find. All we get are
crumbs from the great table, and don't ever think otherwise.
An analogy: amateur astronomers sometimes find things the big scope
operators don't. But they do it by a lot of persistence, and probably a
great deal of experience at using their limited stuff. NO-ONE who just
comes to it with a few parts scrounged off some related optics is going to
do it just because it seems new to them. They wouldn't even understand
enough about what they saw to know it in the extremely unlikely event that
they did discover something new.
When something is new to us, we can enjoy its newness, but it takes serious
folly to assume that no-one else has seen it, or that we have something
truly new. All this playing with diodes is meaningless except to establish
the best conditions for prolonged life at high power, and all that wouyld
ultimately do is sharpen the accuracy of the manufacturer's MTTF data,
which VERY few people even get to see. And even that depends on rigorous
recording and controlling of conditions, none of which is happening in
these kinds of tests. It's like a fantasy world where people come back from
the edges of the oceam screaming about the wonderful new monsters they've
found, while not realising that that world has already been thoroughly
mapped.
Lostgallifreyan
news:9lmbo3psq800p1e4h...@4ax.com:
> If this gets to be a pain, I'd just buy a completed assembly
> somewhere, if they exist. Looking for something compact with about
> 150mw or more. Ideas?
>
eBay. Cheap too. So cheap it's not worth trying to build any to compete
with the low cost, unless you want the fast lane to a miserable life.
Far better sense to sell diodes in small quantity from bulk stock to
hobbyists. I know, I've tried both, and the raw diode selling was a better
ride, by an order of maginitude at least, once I'd had all the fun I wanted
from trying to build modules.
I'll still try, I have a couple of ideas whose cost to build is so low it
could be viable, but it's not that likely. Unless you're aiming for top
spec performamnce, to supply a market willing to pay $$$ for it, forgot it.
That cuts both ways, just as it's not economical to make modules for
cheap sales unless you have access to a large scale Chinese manufacturing
base, it's becoming harder to find any because makers are realising this.
It's easier to find raw diodes. People are waking upo to this too. The only
thing I have, is I woke up to it a few months sooner than most.
b...@fractalfreak.com
Thanks for the heads up. Where'd you get the diodes
from?
These 5mm cans are tiny, but looking at the side
of my two spares I see, two lines, "5HJL" "0058"
So you may be right.
Eric
Lostgallifreyan
news:7df9fbd4-b25c-4cf5...@q77g2000hsh.googlegroups.com:
Neither of those diodes will do what you claim. They're both exceeded by
the Rohm diodes that overwhelmed eBay for a while, and the Rohm's are
exceeded by the newer Opnext diodes. And none of these can do what you
claimed. Ignore all this if you like, but it won't go away.
b...@fractalfreak.com
> b...@fractalfreak.com wrote in news:cc899bcc-0dde-4694-a1a8-
>
> > Look at the discussion over at LPF. People are reliably
> > driving these diodes with 300mA of current and using
> > them for pointers and they are not dropping like flies.
>
> When people want to boast about their successes, they're not going to shout
> about their failures. I don't care about a forum of people playing, if this
> were a real claim, it would be made my the industry that makes diodes and
> drives! They don't leave the big stuff for newbies to find. All we get are
> crumbs from the great table, and don't ever think otherwise.
>
Good grief what's your problem? You clearly have some sort
of chip on your shoulder.
Look, NOBODY is saying these parts are going to last anywhere
close to their MTTF driven the way they are. Who cares, it's
a *pointer*. Generally, as a pointer, it's not on for more than a
minute of two at a time, if that. Even if the part only lasts 10 hours
that's hundreds of "uses".
If I had purchased one of the new Hitachi diodes (640nm @ 150mW)...
http://www.photonic-products.com/products/laserdiodes_visible/hitachi_visible_ds/hl6385dg.pdf
(for example) I would make every effort to make certain the
part was driven at spec. But then for the price of one of
these cans I could literally order dozens of the Sony diodes
at $20.00 each. The parts for my pointer cost more than
the diode. If the diode pops.. <shrug> I can just put another
one in the Meredith assembly.
Where else are you going to find a visble single mode
diode that can be pushed this hard for $20.00?? For the
average experimenter, nowhere. I'd love one of the new
high powered 640nm Hitachi cans, but there's no way
I'm going to drop hundreds to thousands on a single
diode.
Guess my point is, lighten up. It's a $20.00 diode, even if
it lasts 10 hours I think you get at least $20.00 worth
of value, YMMV.
Eric
Lostgallifreyan
news:e6b0a301-ca3d-4a7d...@e4g2000hsg.googlegroups.com:
> On Jan 10, 6:25 am, Lostgallifreyan <no-...@nowhere.net> wrote:
>> b...@fractalfreak.com wrote in news:cc899bcc-0dde-4694-a1a8-
>> 21f612565...@k39g2000hsf.googlegroups.com:
>>
>> > Look at the discussion over at LPF. People are reliably
>> > driving these diodes with 300mA of current and using
>> > them for pointers and they are not dropping like flies.
>>
>> When people want to boast about their successes, they're not going to
>> shout about their failures. I don't care about a forum of people
>> playing, if this were a real claim, it would be made my the industry
>> that makes diodes and drives! They don't leave the big stuff for
>> newbies to find. All we get are crumbs from the great table, and
>> don't ever think otherwise.
>>
> (...)
>
> Good grief what's your problem? You clearly have some sort
> of chip on your shoulder.
>
Just urging a bit of reality. The net is already full of unjustifiable
claims without me standing by and letting people fall into traps made by
wishful thinking. The only 'chip on my shoulder' that exists is the
annoyance of having been there already, seeing people making again, and
again, the same mistakes. If people beleive that crap the only way I can
compete to sell stuff is to start lying! As I have NO intention of going
there, I will defend what I know of the truth.
b...@fractalfreak.com
> On Wed, 9 Jan 2008 13:03:42 -0800 (PST), b...@fractalfreak.com wrote:
>
> >Finally the diode was another $20.00. So the parts came to
> >$70.00, plus the work of throwing it together.
>
> >The Meredith housing is meant for 9mm diode cans (the Sony is 5mm)
> >so it's necessary to press fit the sony diode into a little brass
> >adapter. However I found that while focusing the Meredith
> >assembly the diode would squirm and rotate, not staying centered
> >in the optical path.
>
> Thanks for your post, Eric. I've been accumulating a few parts,
> thinking that I'd eventually put together a small red or IR laser for
> some experiments (not for burning, but high power if possible). Sounds
> like Meredith has some useful mounts and lenses.
>
> The diodes that I found are -supposed- to be 150mw, but I haven't been
> able to find any data on them. Part number is J8Y4M4C. Anyone know
> anything about it?
>
> The diodes are packaged with a lens that says "Achromatic Lens, 12mm x
> 6mm, AR Coated."
>
> If this gets to be a pain, I'd just buy a completed assembly
> somewhere, if they exist. Looking for something compact with about
> 150mw or more. Ideas?
I've had good luck with the Meredith assemblies. I've got some
pointers I made a couple of years ago (15-20 mW, 635nm) that
are still going after hundreds of hours of use, using the Meredith
assemblies.
Other than that I'm not familiar with your part. "lostgallifreyan"
does have a good point -- if you expect to get a reasonable life
out of the diode you need to take care not to drive it over
spec.
Eric
b...@fractalfreak.com
Ooops, however now that I look at the product code
it's listed as "xxKL".. not "xxJL"
lava
> it's listed as "xxKL".. not "xxJL"
Yes, I have seen this, but you you think the 'VL' diode would have
marking 'JL' and the JL diode have marking 'KL'? They do make a KL
diode and it's possible that figure was simply copied from that
datasheet, but not changed to JL.
Either way, lostgallifreyan is right. I still have about 90 Rohms
that are trickling away on ebay and once they are gone I won't lament
their passage into the hands of experimenters that burn the diodes as
much as their electrical tape and matches. It's even possible some
may end up seeing real service.. one can only hope.
b...@fractalfreak.com
real service" lol
Don't see what the problem is. Adjust the rheostat
accordingly. Sounds like at about 150mA these diodes
would last. That's still a 80-100mW laser - not bad
for $30-40.
It's amazing they can be pushed so far overspec and
survive at all.
Eric
Lasers In Mind
> Other than that I'm not familiar with your part. "lostgallifreyan"
> does have a good point -- if you expect to get a reasonable life
> out of the diode you need to take care not to drive it over
> spec.
This is an area I recently had a dabble in. Some diodes are
analogue and will keep increasing output depending on how
much voltage/current is shoved up them to the point when they
will of course die.
Others have a quasi digital flavour and after a certain amount
of current is put through them they refuse to take any more.
There is a definite 'point' when this happens and easily recognised
as the (Rohm) diode dims slightly.
The former I have seen in Optnext and the later in Rohm.
I am NOT experienced in any way with diodes. Just my observations
after blowing a few up. To this end I'd have thought that a dabbler
would be better off playing with Rohms as they are so much harder to kill.
lava
> This is an area I recently had a dabble in. Some diodes are
> analogue and will keep increasing output depending on how
> much voltage/current is shoved up them to the point when they
> will of course die.
>
> Others have a quasi digital flavour and after a certain amount
> of current is put through them they refuse to take any more.
> There is a definite 'point' when this happens and easily recognised
> as the (Rohm) diode dims slightly.
>
> The former I have seen in Optnext and the later in Rohm.
>
> I am NOT experienced in any way with diodes. Just my observations
> after blowing a few up. To this end I'd have thought that a dabbler
> would be better off playing with Rohms as they are so much harder to kill.
do you mean like this:
http://hacylon.case.edu/laser/SLD1239JL.jpg
this is one of the diodes from the OP's link.
Lostgallifreyan
08b958...@f10g2000hsf.googlegroups.com:
> It's even possible some
>> may end up seeing real service.. one can only hope.
>
> real service" lol
>
> Don't see what the problem is. Adjust the rheostat
> accordingly. Sounds like at about 150mA these diodes
> would last. That's still a 80-100mW laser - not bad
> for $30-40.
>
Try 125 mW for the Rohm's (in a clean supply, and perhaps some TEC if you
really want the long haul and a slight wavelength shortening, useful as
apparently they're already around 655 (most makes tend to be longer than
658 rather than shorter)). But you're right, that's by far the best way to
go. Most people will want them adjustable anyway, and won't want the full
brightness all the time. I'm thinking of using a simple external switch, to
allow a very tiny module to be built. Too small for fancy drive, but still
with very good protection added.
> It's amazing they can be pushed so far overspec and
> survive at all.
>
True but you enter the chaotic realm the moment you push beyond around 125
mW CW, on ANY of the current DVD diodes, no exceptions. Even the Rohm's
might do 300 mW for a short time but most won't, and I could burn the rest
so fast I'd have no meaningful plot of the expected performance even after
getting through 200 of them. The later Opnext diodes seem to be cleaner in
their modes at mid to high power, but even those will be chaotic at above
150 mW. That's why Marconi (on PL, with the MaxYZ modules) can keep a good
position by using narrow stripe multimode diodes. I don't know where
else those diodes will start to appear from wherever he gets them from, but
that's the place to look for brighter diodes for show gear. He won't give
up the detail, I can't compete with that, so it's like a chess game in
which I can't get out of check, even though it's not checkmate. If I was
limited to a chessboard the only thing to do would be resign, but being
real life, there's always the hope of turning resources some other way. I
decided to turn my diodes into parts for drivers, and tooling to design and
make PCB's. Much harder to be caught on a new stalemate that way.
Lostgallifreyan
True that Rohm's are harder to kill outright. Several buyers commented
favourably on it, to the point where I might be able to slightly extend a
claim for them. I won't though. They mode hop frantically at over 125 mW
CW, starting anywhere on the scale, That's what accounts for the
discontinuous output scaling. The best we can get is a fit between the
application and the edge of the chaos. This is difficult, and is probably
the main reason why the Rohm diodes flooded the market. They are good, but
just too unpredictable to base any mass marketed device on. And Rohm have
plenty of other diodes in their arsenal too.
At the top end of the scale you won't see then dim, they'll just die. :)
And they won't always do it at times and in ways you can predict. One might
die hard, another will wimp out during a long and modest uptime. Opnexts
around 2 years ago could be pushed more consistently to max, probably still
true. The telltale was a loss of speckle, a gentle breakdown of coherence
even at close range. You could see it by eye, if projected across a matt
black surface. I'm not sure that newer ones reveal that same clue though.
If so, you can back off on the drive on seeing that effect, and be sure
you're at peak CW drive. I don't know why that effect existed though, I
only know it did exist, having pushed a few to destruction carefully to
verify it. Might have gone on with them but they were EXPENSIVE to me at
the time, £50 a go, and I discovered retroreflection damage the hard way.
Thought it was static, eleiminated everything by careful design and
construction, before proving that even low level reflections could kill
them during overdrive. No-one was talking about that, not the seller, not
the maker, and I don't know if there was even anything in the LaserFAQ
about it then. I think that put a lot of people off making modules,
especially those that combined beams. Either diodes were consistent and too
costly, or cheap and too inconsistent. Not sure if anyone's really cracked
that yet.
Lostgallifreyan
news:Xns9A21DF6B251...@140.99.99.130:
> which I can't get out of check, even though it's not checkmate.
That looks daft, not what I meant. You know, the kind of end game where you
can GET out but never STAY out...
Lasers In Mind
at £20 a go and almost chucked the whole lot in.
Got a set of 5 Rohms from LG three months ago
expecting to blow a few more up but still haven't
managed to kill one yet despite severe abuse and
runs of up to 4 hours CW.
b...@fractalfreak.com
>
> True but you enter the chaotic realm the moment you push beyond around 125
> mW CW, on ANY of the current DVD diodes, no exceptions. Even the Rohm's
> might do 300 mW for a short time but most won't, and I could burn the rest
> so fast I'd have no meaningful plot of the expected performance even after
> getting through 200 of them. The later Opnext diodes seem to be cleaner in
> their modes at mid to high power, but even those will be chaotic at above
> 150 mW.
How about DVD burner LDs harvested from 20x drives ..
would it possible to get reliable operation at 200mW?
(by reliable I mean at 100+ hours for use in a pointer)
I wish I knew more about how these LDs are constructed,
and what the limiting factors are to getting higher power
single mode LDs. You are right that it's very hard to find
any single mode LDs above about 150mW. I believe much
of the early work and trade secrets to getting reliable higher
power LDs revolved around preventing COD to the end facets.
Is the limiting factor a fundamental one, that is the optical
power density at the facets inevitably causes it to degrade,
or is it more a manufacturing problem - reliably creating the
die with very few crystal defects ??
Eric
Lostgallifreyan
4aa138...@k39g2000hsf.googlegroups.com:
> How about DVD burner LDs harvested from 20x drives ..
> would it possible to get reliable operation at 200mW?
> (by reliable I mean at 100+ hours for use in a pointer)
>
Doubtful, because it's so hard to get the small extra power beyond current
levels, that makers have resorted to leaving out the cans and the windows
to prevent losses (and presumably save money. The loss of physical
protection for the die means there's just more that can go wrong. With
drives people will likely update within weeks or months, they don't care
much.
> I wish I knew more about how these LDs are constructed,
> and what the limiting factors are to getting higher power
> single mode LDs. You are right that it's very hard to find
> any single mode LDs above about 150mW. I believe much
> of the early work and trade secrets to getting reliable higher
> power LDs revolved around preventing COD to the end facets.
>
The facets are cleaved and polished. There's no coating, it's difficult to
put one there, so the surface of the crystal is the weak point. It's not
very durable, and the lack of a can and window on the 20x writer diodes
doesn't help at all... Increased purity of materials is about all they can
do, and even then, dopants don't make precise crystal lattice fills,
there's always some flaw there. To change that there would have to be some
way to precisely build it up like a lego model, atom by atom. That is
doable, but not feasible. Not really doable either, it would take too long,
some better way to mass produce them will be found before one diode could
be built that way.
> Is the limiting factor a fundamental one, that is the optical
> power density at the facets inevitably causes it to degrade,
> or is it more a manufacturing problem - reliably creating the
> die with very few crystal defects ??
>
That will take more knowledge than mine to asnwer well. I tried, but it's
not a very useful answer.
Lostgallifreyan
Nice. I nearly gave up, when I spent £500 (which was very hard to get for
me, let alone risk spending on 10 Opnext 80 mW diodes, only to learn the
hard way about the retroreflection, which cost me 4 of them before I'd
proven it wasn't some other cause). Fortunately I could get enough for just
TWO modules built at that time, to recover all my losses. Next big diode
spend was >£500 on 200 Rohm diodes. Luck break, last the guy had, I just
missed an offer of 500, a German guy bought 2000. I suspect he realises he
might have bought a few too many by now. By the time he shifts them, no
matter how he does it, their value will fall low enough to be a problem for
him. I expect them to hold their value for another year, at most, and he's
certainly not punting out lasers or diodes at a rate that will sell out in
that time.
If you or anyone else want some more Rohm's, I'll sell a few more. Got my
sights on some deep-cycle batteries, I want to add them to some solar
panels I just bought... I also have some polarised beam splitter cubes, 5
mm, AR coated for red wavelengths, not as efficient as Edmund Optic's
cubes, but they won't cost a couple of hundred quid, either. I also have
several small efficient TEC's with metallised pre-tinned surfaces, to run
off 3.3V and pump up to about 3.5 watts.
Lasers In Mind
> I just bought... I also have some polarised beam splitter cubes, 5
> mm, AR coated for red wavelengths,
Tell me. What makes the cubes frequency specific?
Lostgallifreyan
Nothing. That's why you optimise the transmission for one wavelength, at
the expense of others. At least this way, red light gets through cheaply,
with little loss. Without the coating, other clours might get through
better, but all will lose more that red would on these. These PBS cubes
were made specifically for combining red diodes.
b...@fractalfreak.com
>
> Doubtful, because it's so hard to get the small extra power beyond current
> levels, that makers have resorted to leaving out the cans and the windows
> to prevent losses (and presumably save money. The loss of physical
> protection for the die means there's just more that can go wrong. With
> drives people will likely update within weeks or months, they don't care
> much.
Yes I imagine with the 'open can' design dust might be
a problem if it were to accumulate.
>
> The facets are cleaved and polished. There's no coating, it's difficult to
> put one there, so the surface of the crystal is the weak point. It's not
> very durable, and the lack of a can and window on the 20x writer diodes
> doesn't help at all... Increased purity of materials is about all they can
>
Quite certain you're wrong on this. Facet passivation is done to all
high power LDs to slow/prevent COD. It's likely one of the many
proprietary manufacturing steps we know nothing about
You wrote earlier:
(...)
>but you enter the chaotic realm the moment you push beyond around 125
>mW CW, on ANY of the current DVD diodes, no exceptions. Even the Rohm's
>might do 300 mW for a short time but most won't, and I could burn the rest
>so fast I'd have no meaningful plot of the expected performance even after
>getting through 200 of them. The later Opnext diodes seem to be cleaner in
>their modes at mid to high power, but even those will be chaotic at above
>150 mW.
You talk here about entering chaotic modes/realm once the
power gets above about 150mW. Can you explain what you
mean by this and how you're evaluating this?
Eric
Lostgallifreyan
news:8648a425-9fcd-4a75...@e23g2000prf.googlegroups.com:
>> The facets are cleaved and polished. There's no coating, it's
>> difficult to put one there, so the surface of the crystal is the weak
>> point. It's not very durable, and the lack of a can and window on the
>> 20x writer diodes doesn't help at all... Increased purity of
>> materials is about all they can
>>
>
> Quite certain you're wrong on this. Facet passivation is done to all
> high power LDs to slow/prevent COD. It's likely one of the many
> proprietary manufacturing steps we know nothing about
>
Didn't know about that. Last I heard, there was no way to prevent COD with
a coating, only to prevent oxidisation. Google does show COD related
results for Facet passivation though.
http://www.freepatentsonline.com/4337443.html
I tried reading some of that but it goes beyond my grasp quickly. It looks
like there's a way to set the thickness and the refractive index to prevent
an AR coating forming, in turn preventing a rise in threshold current. That
allows effeciency and higher maximum output. Probably what Opnext have been
doing.
> You wrote earlier:
> (...)
>>but you enter the chaotic realm the moment you push beyond around 125
>>mW CW, on ANY of the current DVD diodes, no exceptions. Even the
>>Rohm's might do 300 mW for a short time but most won't, and I could
>>burn the rest so fast I'd have no meaningful plot of the expected
>>performance even after getting through 200 of them. The later Opnext
>>diodes seem to be cleaner in their modes at mid to high power, but
>>even those will be chaotic at above 150 mW.
>
> You talk here about entering chaotic modes/realm once the
> power gets above about 150mW. Can you explain what you
> mean by this and how you're evaluating this?
>
Rohm diodes visibly mode hop at anything above 125 mW, some samples lower.
If you google for holography and DVD diodes, you'll find a few pages
discussing this and other diodes. Rohm's were ok below around 50 mW, though
still noisy. Other diodes were better, but I didn't see any report of a
diode doing well beyond about 150 mW, so I think it's safe to assume that
you can't predict performace above those levels for CW in single mode
diodes, especially given that few if any are rated beyond 120 mW CW.
steveb...@swva.net
I'm running a Mitsubishi 636 mM at 250mA (They're rated at 300 max)
and it seems quite happy. (It's 150 mW output) I did pop one though
due to an electrical error (dangling wires) and at 295 USD ea I about
crapped .. but I'm still at it now..(tho more carefully)..
Lostgallifreyan
606148...@e10g2000prf.googlegroups.com:
> I'm running a Mitsubishi 636 mM at 250mA (They're rated at 300 max)
> and it seems quite happy. (It's 150 mW output) I did pop one though
> due to an electrical error (dangling wires) and at 295 USD ea I about
> crapped .. but I'm still at it now..(tho more carefully)..
>
Special diode? No DVD type for sure, maybe for other purposes needing high
visibility. Is it single mode? The price is bearable, just, but I won't try
for one till I have something worth getting them for. Well beyond my budget
for general experiment..