blue led

I think you need to provide us with some additional information.  What is hight power?  Do you have a a wattage number.  Do you need pulsed or constant illumination?  Do you have a specific wavelength that the blue light needs to meet?  What are the capability of your mains?  I am assuming you are using european current and not US circuitry.

If you tells us better what you need, we can make better recommendations to solve your problem.

FYI, I have a lot of experience in spectral sensors, sources and processing and would be interested in hearing a little more about your overall project.  I may be able to help you come up with a better solution.

But first, you have to help us understand your problem before we can help you.

Just a thought,

DAB

Hi DAB

Thanks for getting back to me. Sorry better give a bit of background. The blue LED is for the use in microscopy to excite an antibody label at 470-490nm.

Here comes the problem, I know the biological side of it but don't have much of electronic background.

I found some articles in journals bellow;

http://http://www.plosone.org/article/info:doi%2F10.1371%2Fjournal.pone.0002146

http://www.ncbi.nlm.nih.gov/pubmed/19772530

http://www.stanford.edu/group/blocklab/GutierrezAJP2010.pdf

Looking at other web sites gave different components which I'm not sure what is the advantage of using those. The Idea is to make it as cheap as possible probably around £20 if possible. Also the price of LED varies so much from around £1-£40 for the same so called high power LED. It is a worth while project if it works hence I need your expertise and suggestions.

Many thanks

Hooman

Ok Hooman,

I did some web searches and from what I can tell, the HPLC technique uses the UV light to measure how much the cells absorb as they pass through the flow column.

I am not sure you can do this with an just an LED, but I am not a specialist in using the equipment.

I could probably help you if you just want to illuminate cells on a slide to see the cells by contrasting the fluid absorbtion verses the cells, but I am not sure if that is what you want.

Hopefully someone else in the community can give you a more concise solution, but I am afraid that this problem is out of my expertise.

Sorry,

DAB

Thanks DAB for going to trouble to looking into it. it is a bit like flowcytometry i guess but in solid state. The way it works is the antibody is conjugated to a light sensitive chemical. what we do is we put this on the sample. if what we are looking for is present in the sample it will bind to it and under right wavelength it will luminance.normally it is done using mercury or similar lights with filters to obtain the right frequency to this to work. 

did you see the board designs? do you think they are ok? or can be improved? is it better if I just put up the board designs?

Not sure if I'm making scene but we need your expertise to solve our problem to make this cheaper and more accessible. This is the problem with science, lack of communication .

Many thanks again

Hooman

Actually reading your reply again that is exactly what I want. cells are fixed to a slide, then stained and observed under microscope. all it needs a constant supply of blue light which does not loose intensity. it is also important that the light source has a long life and doesn't need replacing like mercury lights where it has to be changed after 200 or so hrs to keep the intensity.

I assume you are in states or I would be more than happy to explain in person.

Hi Hooman,

Ok, now I understand.  You need the UV to trigger the binding mechanism.  I didn't see any circuit diagrams in the search I did.

However, I have a suggestion.  Search on E-Bay for an old UV EPROM eraser.  Nobody uses them much anymore so you should be able to get one inexpensively.

The one I have came with a UV bulb and a timer.  If you can get one of those, you simply put the slides inside and set the timer and the UV lamp will illuminate the cells.  You may have to play with the timing a little to find out how long you need to expose them, but that would be the most inexpensive way to get what you need.

Other than that, you can build a simple 555 timing circuit with adjustable frequency and pulse width to drive a UV LED.  The circuit is readily available and you should be able to get by with a low wattage LED.

If the action is like you say it is, then all you need to do is illuminate the sample until the marker sets.  Again, you just need to play with the exposure time.  A higher wattage LED might run faster, but you might need more driver circuitry, which is more complex then the simple timing circuit I propose.  Just search on 555 timers and you will get more than enough tested circuits that will be easy to build.

I think that will solve your problem.

I think good quality blue LEDs are manufactured by Cree: XP-E series:

http://www.cree.com/led-components-and-modules/products/xlamp/discrete-directional/xlamp-xpe

(output power up to 500mW at IF 350mA, IF max 1000mA so output power over 1W can be achieved). These LEDs are available in two versions: royal blue (450-465nm) and blue (465-485nm).

There is also available XT-E (up to 550mW at IF=350mA):

http://www.cree.com/led-components-and-modules/products/xlamp/discrete-nondirectional/xlamp-xte-royal

You need good cooling if you plan to use them at higher power levels. It can be less difficult to work with them if they are soldered onto MCPCB (as on http://www.plosone.org/article/slideshow.action?uri=info:doi/10.1371/journal.pone.0002146&imageURI=info:doi/10.1371/journal.pone.0002146.g001).

As to schematic from:

http://www.plosone.org/article/info:doi%2F10.1371%2Fjournal.pone.0002146

I think it is better to use current source e.g. based on LM317. Best is switching mode curent source (less heat) but is moe complicated and small current variations (hence light intensity variations) are noticable but in kHz range. If exposure time is in ms range, these variations should be unnoticeable. But if you need on/off gating in microseconds range, circuit should be different.

Good update, I was hoping someone would come to my rescue.

Your links are very useful.  I learned a lot about spectral mircrscopy by looking into this post.

I haven't had an interesting project to analyze like this in a while.

Thanks for the education.

DAB

Thank you both for your help. I'm non the wiser as the subject is well over my understanding. But will try your suggestions RWM and google the bits i don't understand. There was also a good link about LED safety which helped a lot. As you'll be staring down the microscope up to 2 hrs at a time or even more safety of the operator is paramount.

Once again I thank you both as this is a worth while project and potentially can save many lives. If and when I get it off the ground I will make sure both your names are on the paper (if any to be published).

meanwhile any more suggestions are more than welcome.

I will keep you updated of progress.

cheers

Hooman

Hi Hooman,

if you have questions please ask, I will try to help.

Hi RWM

Thank you for your help and advice. I'll try to get a plan together very shortly (probably from the web ) and i would like you to check it and see what you think. I will keep in touch and keep you updated of the progress.