Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

LPF Donation via Stripe | LPF Donation - Other Methods

Links below open in new window

ArcticMyst Security by Avery

400 - 5mW diode configuration questions.

Joined
Jan 29, 2009
Messages
11
Points
0
Hey guys,

I hope this isn't too crazy to ask but this seems to be the place to ask it.  I have already learned quite a bit from sifting through the site for the last day or two and the depth of topics discussed here is great.  I have been going through the threads and searching the site but couldn't really find anything like what I am trying to do... so here goes nothing.

Basically I am trying to construct a 400 laser diode array.  Each of these diodes will be about 3.2 volts, 635nm, and 30 mA.  I have already tested out some configurations so I will try to list what I have tried and then post my questions afterward.  

The diodes we are using right now are the 635nm modules from Aixiz that come with the LD driver circuit and adjustable lens/brass casing.  The first thing that a bunch of us amateurs did was the easiest thing we could think of at the time.  We were running these diodes in configurations up to 200+ in parallel using modified ATX power supplies.  I got a bunch of 250 Watt power supplies and isolated the 3.3V rail which gave us the capability of about 15-17 amps if needed.  I know this sounds ridiculous but it was working  :p.  I figured the ATX power supplies would be a great source of electricity because they are regulated pretty well, have built in heat-sinks and short-circuit protection.

Anyhow where we are now is that we want to forgo the whole module/driver circuit and instead get just the diodes themselves because it is much cheaper.  I would like these diodes to last as long as possible so I was thinking a soft start, voltage/amp regulation would be needed.  So here are a few questions now that you know the background:

Is the ATX Power Supply enough regulation to run diodes without the modules and Aixiz driver circuits (I am pretty sure it is a no but might as well ask)?

If I do need closer regulation then what would be the best way to connect this all together?

Could I connect a bunch of diodes in serial (say 10 - 15) use a higher voltage/lower amperage and just regulate each series of diodes using a modified DDL driver?

Would it ever be efficient or probable to regulate 400 diodes in parallel at an economical price (my guess is no)?


Thanks for reading guys and I hope this isn't out of line posting stuff like this.
 





Joined
Nov 22, 2008
Messages
1,506
Points
48
What a bizzare project! There must be some story behind that, care to explain?

As long as you're careful there should be nothing preventing you using a load of lm317-based regulators but remember 1 false move and you will kill ALL the diodes connected to it. LM317's are safe up to 1.5A total current, provided they're cooled properly. Remember, PC power supplies aren't ideal voltage sources, there will be a surge as it powers on, and fluctuations as load changes, so use smoothing capacitors. Try experimenting with a few test loads in parallel with each other (test loads are silicon diodes in series, around 4 behaves a bit like red diodes, or 6 like blu-ray diodes) and an LM317, and see what works best in your situation.

Is there a specific need to use 400 diodes, however? Why not simply wire up 10 modules or so and split the beams using optics? Or, you could try video editing if it's for a film or something similar. With that many diodes all potentially pointing into your eyes, you'll also need to take safety into consideration.
 
Joined
Jan 29, 2009
Messages
11
Points
0
Bizarre yes... I can agree with that. If you must know ;) it is for the creation of a home use therapeutic laser device. Lasers are at the cutting edge of healthcare these days and there is plenty of research that has been done on their uses. Different wavelengths are used for different applications depending on the depth and type/color of the tissue that needs to be targeted. For this particular application more diodes are better because we are trying to cover a larger surface area. In fact all of these lasers will be diffused to cover more area. I know this will reduce the depth and joules/cm^2 but we are not going for depth in this application. This device is used for wound healing, scars, etc. etc.

We have been tinkering with mediocre designs for the past year or so and have been recently looking into ideas to upgrade our design to possibly make it cheaper to construct in the future. I am pretty decent working with basic electronics, soldering, wiring etc. but I this is the first time dealing with lasers for me. I have learned quite a bit in the recent months but I just want to make sure I have all my basis covered since I know these diodes can be extremely sensitive sometimes.

I found a decent regulator that isn't too expensive that runs anywhere from 3 - 7.5 A. Would this be ok to use in this application?

[Appears I can't post links yet... but there is a line regulator made by LINEAR; LT1083/LT1084/LT1085 Series. You can find it on their site with the data sheet] Sorry for the inconvenience.

Do you know a good place where I can find some silicone diodes?

Thanks for the input... it is nice to hear from people with lots of experience using laser diodes.
 
Joined
Jan 6, 2009
Messages
202
Points
0
Just use a few higher power diodes and don't focus them. Without optics, the beam diverges a good 20-30 degrees, more light bulb than laser.
 
Joined
Jan 29, 2009
Messages
11
Points
0
The problem with using these in this application is that there is a "sweet spot" so to speak in terms of exposure. In this case more is not always better. The bio-stimulation range for tissue/cells is 4-6 joules/cm^2; around 10 j/cm^2 you will get no benefits; and around 16+ j/cm^2 you will actually start to do more damage. We did some calculations a while back with a laser sensor and such and came up with the 5mW diffused at 1.5-2cm from the target area for 20 minutes to get the desired energy.

This is good for bio-stimulation because it actually increases the exposure time. It is theorized that this increases the amount of Nitric Oxide (NO) and SOD that is created, which is part of what we are looking for. A higher mW might reach bio-stimulation quickly, which would be convenient, but not necessarily better. If we can prove that a certain higher mW diode at x distance is equivalent during the same time period then maybe it would be ok. Unfortunately most laser sensors are so small it is very hard to measure accurately without very expensive equipment.
 
Joined
Nov 22, 2008
Messages
1,506
Points
48
As the guy with the name made of numbers said, it would be better to use a powerful diode unfocused, without optics, with a diffuser or underpowered. With this you can cover a larger area more consistently and, by setting the distance or power, tweak the energy concentration to the "sweet spot" exactly. Using hundreds of diodes is almost never a practical solution. If you want, I will be able to get some photos tomorrow with my red laser focused, unfocused, part-focused or "bare" at some different distances, so you can see the difference in brightnesses. If the aim is total energy transferred, you will without doubt be better off with a few powerful diodes unfocused, instead of lots of weaker ones focused.
 
Joined
Jan 29, 2009
Messages
11
Points
0
While in theory it should work just the same... all of the testing was done with this distance and this power output. The design we are using keeps the lasers close to the skin as to minimize power loss. I remember reading somewhere that diffused diodes lose there ability to bio-stimulate at certain distances due to the particles hitting the skin at different more extreme angles. At this distance there is some diffusion but not enough to lose its efficacy.

Charlie can you recommend a place to get silicon diodes?

Thanks for your help.
 
Joined
Nov 22, 2008
Messages
1,506
Points
48
Silicon diodes: any normal, cheap-as-chips diodes will do. 1n4001 or similar diodes are available online through DigiKey and RadioShack, Maplin or sites like those. Don't buy it without checking, but I believe that the diodes here:

http://www.maplin.co.uk/Module.aspx?ModuleNo=19079 are suitable. While you're there you could pick up some voltage regulators, but I'm probably not the best person to ask. Daedal wrote an interesting thread on homemade drivers, so you could PM him.

If the light needs to enter at a direct angle, you can almost certainly get a beam expander, or other optics to get the right spread and angle of entry. They'd be expensive, but in the long run I suspect it will save money in power, diode costs and maintenance.

When you said power loss what were you referring to? The only loss I can think of is reflections form dust ini the air, and that is negligible...
 

Danjoo

0
Joined
Sep 18, 2008
Messages
202
Points
0
hapymann talk about 635nm modules from Aixiz, is these wavelenght required for this medi projekt?
The high power dvd burner reds normally have about 650-660nm.


daniel
 
Joined
Oct 26, 2008
Messages
50
Points
0
Wow what a project :eek: .
I would try to connect as much diodes as possible in series and drive them directly form the rectified and current regulated mains. Here in Europe after rectifying you have 320 Volts so when the diodes have a Voltage drop of say 2V you could connect 160 minus some (because of the voltage drop at the regulator) in series which means that you only have to draw 3 times the current of one diode (roughly 25mA x 3 = 75mA). In that way you can use small linear regulators for current regulation.
One problem I see with this configuration is that most red laserdiodes are connected to the case with their cathode so you have to isolate the diodes from each other.
 
Joined
Nov 22, 2008
Messages
1,506
Points
48
Andy, I'm not sure about that as a solution, it would be too vulnerable to surges and AC fluctuations. AC mains rectified, a capacitor to smooth and a set of regulators would be safer. With so many diodes at stake, it's not worth that type of risk. A fuse won't protect it either, diodes would blow first...
 
Joined
Oct 26, 2008
Messages
50
Points
0
charlie bruce said:
Andy, I'm not sure about that as a solution, it would be too vulnerable to surges and AC fluctuations. AC mains rectified, a capacitor to smooth and a set of regulators would be safer. With so many diodes at stake, it's not worth that type of risk. A fuse won't protect it either, diodes would blow first...


@ charlie: This is exactly what I meant with "rectified and current regulated mains", maybe I didn't describe properly enough. I agree that just hook it up to mains is too dangerous, it definitely needs current regulation. To be really isolated from the mains fluctuations, maybe even put a dc-dc regulator before the current regulation.
 
Joined
Nov 22, 2008
Messages
1,506
Points
48
Hi, sorry about that, I thought you meant just rectified mains without current limter. I had another thought, would a bench-type power supply with a good current limiter work? Sadly, setting the current on one of these will cause problems when (if) one of the diodes breaks - the current would end up increasing through the rest of them. Best design, if you really do need hundreds of diodes, will be small clusters of say 5-10 diodes in series, with a regulator, each cluster then joined to a regulated power supply.
 
Joined
Oct 26, 2008
Messages
50
Points
0
charlie bruce said:
Best design, if you really do need hundreds of diodes, will be small clusters of say 5-10 diodes in series, with a regulator, each cluster then joined to a regulated power supply.

This gonna be a really big box then. For 400 diodes that would be at least 40 power supplies. I know from high power diodes that the preffered connection is all in series.
Here is an example: http://www.roithner-laser.at/htoe/AS081Q.html
The AQ081Q15000 has 150 Bars in series and is operated at 300 Volts. Not comparable with the specs but just a proof that such series connections work.
 
Joined
Nov 22, 2008
Messages
1,506
Points
48
Again sorry if I wasn't clear, I meant that each cluster is connected to the same power supply (provided it has enough ampage to power all of the diodes!). I said this so that the diodes don't drain more voltage than the supply can give, there's the fault protection on the diodes being in series rather than in parallel, and the flexibility of being able to remove a "cluster" with no adverse effects for maintenance.
 
Joined
Jan 29, 2009
Messages
11
Points
0
Wow guys thanks for all the input. Charlie that was kind of what I was thinking about doing. Right now I am using a modified ATX style power supply which can give me access to a lot of different voltages at a pretty high amperage. I was crunching some numbers to see how many diodes I would be able to put in a series and I guess it would depend on what type of voltage drop I would get with the regulator I use. I read through the first 6 pages of Daedal's stickied post when i had some free time (man is that thing long) and I think he said you could expect a 2 volt drop using his regulator.

So I was thinking of using something like this to regulate voltage:

LT1762 - 150mA, Low Noise, LDO Micropower Regulators (sorry can't post urls yet... you can look up the part number at linear dot com)

Would using these connected to diode clusters using modified ATX power supply be enough or would I need another item to regulate the current?

Once again thanks for the input.
 




Top