LED heat sink questions and help

[tjones96761]

hello all. new to the forum, hope you guys can enlighten me (pun intended). I'm still restricted for being a new member, please excuse the run-on as I can't use enter button for paragraphs for some reason. Recently got into bowfishing, and I mean REALLY got into it. I've been researching LEDs and have purchased a couple from a US vendor that is sourcing from China. The more I research what's available, the more I think the US vendors are making a killing and I want some of that action. Also, the bowfishing lights are in the stone age as far as what's available in my opinion. SO, questions about heat sinks - how hot is too hot? I see 50w chips rated for 60C which is pretty hot. in theory, if a 50w chip is stuck to a heat sink and putting out 4000 lumen @ 40C, how much is the lumen output affected if the temperature goes to 50C? and inversely down to 30C? I do purchasing for an oilfield equipment manufacturer, so I go to China 3-4 times a year. I'd like to get some ideas together and have components made in China and assemble in the US with US made chips. emphasis being specifically on lights for bowfishing, of course... I've done a fair amount of electrical work with VSDs, and one thing keeps coming back to me. VSD manufacturers run water over heat sinks to reduce size and operate where ambient temperature is high. I don't know why this concept hasn't made it into LED lighting, maybe it has and it just didn't work. I have some experiments set up to test heat, but don't have a meter to test light output. Because these lights will be used on lakes, I think I should be able to configure a heat sink that is water cooled. water will be pulled out of the lake at roughly 82F, ran through the heat sink and back into the lake. My thinking is by doing this I can run say 100w chips on a 3x3" heat sink. Let's skip the idea of underwater lights for now. No one seems to high on the idea of cutting holes in their boats for lighting. Up for any ideas, input and help. Thanks.

 

[Harold_B]

How much the output drops as the temperature rises will depend on which LED you spec but that should be called out in the manufacturers data sheet. You will have a lower output and a shorter product life by over heating the LED but most can handle 85C at the die.

 

[DIWdiver]

First, for the return key thing, open a text editor like Notepad, type two line feeds, then select and copy them. Then you can paste into the post.

A number of us are having this problem. Are you running Windows 7?

 

[DIWdiver]

Second, water cooling is very viable, just not practical in most lighting applications. I mean, who wants to keep a lake in their house, car, or flashlight? In your application, where the lake is right at hand, you could easily replace the large finned heatsink with a small block of metal with water running through it. In this case, I would like to see a thermal cutout to protect everything in case the water flow stops for some reason.

Harold B is right that most data sheets give a graph of output vs temperature. This usually normalized, so the output is '1' at 25C, though Cree is starting to normalize to 85C. Then the output would be higher at lower temps, maybe 1.05, and lower at higher temperatures, maybe 0.9, even 0.75 in extreme cases. Typically the difference between 30C and 50C is fairly small, but the difference between 30C and 100C is getting substantial.

You have to remember that the important number is the temperature at the LED die. This can be substantially higher than the temperature at the surface of the heatsink. There can be many thermal interfaces between the die and the air or water that is ultimately your 'infinite heatsink'. If any one of these is poorly designed or executed, the die temperature will soar, while the heatsink temperature won't (usually).

 

[DIWdiver]

Third, you probably already know this, but you have to be very careful having things made by the lowest cost supplier. One dive light maker got really screwed when the vendor he had making his lights started selling knock-offs before his lights even hit the market. I know this can happen anywhere, but it is vastly more likely in areas with poor patent and copyright protection than others with better protection.

 

[RoGuE_StreaK]

I had to do an image search to find out what bowfishing was; seems it mainly involves women in bikinis?

Are the lights intended for boat-mount or bow-mount? I assume the former from the watt figures you are giving? Have you researched which light profiles penetrate the water best?
There have been several fishing light threads pop up here, might be some useful info in them.

As per DIWdiver's comment on suppliers copying your designs, might be worth having a closer look at some US-based small-scale CNC manufacturers, you might be pleasantly surprised at the costs you can get.

 

[tjones96761]

First, for the return key thing, open a text editor like Notepad, type two line feeds, then select and copy them. Then you can paste into the post.

A number of us are having this problem. Are you running Windows 7?

yes, windows 7.

 

[tjones96761]

I had to do an image search to find out what bowfishing was; seems it mainly involves women in bikinis?

I think there would be a much bigger following if that were true...

Are the lights intended for boat-mount or bow-mount? I assume the former from the watt figures you are giving? Have you researched which light profiles penetrate the water best?
There have been several fishing light threads pop up here, might be some useful info in them.

As per DIWdiver's comment on suppliers copying your designs, might be worth having a closer look at some US-based small-scale CNC manufacturers, you might be pleasantly surprised at the costs you can get.

I've searched and read everything that has been postedalready, not much relevant. Most seem to be from bank shooters and floundergiggers. Here are some pictures to give you an idea what we're working with:

This one is mine, rigged with 500w halos. Reflection off thesurface was pretty bad, so rerigged with 300w bulbs.



You can see in this picture how it looks on the water. All thelight shining up on the bank is reflection off the surface. Wish it didn't dothat, but it seems to just be the only way.



This is what it would look like with the 50w LEDs thatpeople are using. I only bought 2 for testing purposes. $85ea for complete unit(fixture, glass, chip, driver, cord all ready to go), and they are supposed tobe IP65.



HPS lights are top end and very popular. Very expensive andvery high maintenance.



Here's one with 27w LEDs. The size of these is nice, butthey don't put out enough light. Plus, the small 3w individual bulbs don't seemto be available in warm, only cool. Warm is preferable, seems to penetrate a littlebetter and it's easier on the eyes.



Most people that are going to LEDs are doing these 50w.



This is one of the vendors that specialize in the bowfishinglights. These are coming from Shenzhen China, this guy just stores a few in theUS and keeps some repair parts. Most of the orders ship directly from China. http://www.customfitzled.com/apps/webstore/products/category/771420?page=1

As far as configuration, I can't really say. The smallercluster LEDs seem to give 60deg angle which is too narrow, the 50w chips give a140deg angle, which is too wide. Perfect would be something that could mountvertically and shine only straight out and straight down. Probably impossible.



Need to be able to shine out about 30 feet with smoothcoverage ( no bright spots or shadows). Light penetration is the biggestconcern. The general way of thinking seems to be more lumen=more penetration.

The latest and greatest development in LEDs for bowfishingis the color change lights. They go from about 2200K to 6500K so you can playwith which color penetrates the different types of water you come across. I'm alittle skeptical of these. As you can see in this video, they seem to strobe alittle. I think this would give headaches after long usage.

http://www.youtube.com/watch?v=O9MX37chKYA

For sourcing, I do about 10 million dollars in purchases fromChina a year for the company I work for, in addition to 3 million from the US. Lotsof experience with US CNC shops and quality problems that come with Chineseproducts. What I do know is anything that has to do with casting or forgingneeds to come from China. US CNC shops aren't competitive until you get intothousands of pounds, where freight costs from China become hindering.

This is the idea I'm kicking around. Cast aluminum labyrinthsection, aluminum lid that will create the "tubes" for water flow. Sealed clearcover to keep water off the electronics.

 

[tjones96761]

forgot to mention that nearly every boat has a generator. So power options can be 12vdc, 24vdc, or 120vac.

 

[AnAppleSnail]

Why can't you mount these underwater? You'd probably want to raise them before driving the boat, but that would help glare. Right?

 

[SemiMan]

From an optics standpoint, having lights that close to the waterline is about the worst thing you could ever do. A creative solution would be to get the lights high up in the air on a boom. I realize they would move a lot, but you would need less power to penetrate the water and the effectiveness farther from the boat will be way better. You could stick to narrower beam angles so you are not throwing light in places it is not helping with.

I am not sure there is a lot of benefit of being able to change the spectrum, at least from cool white to warm white. Blue/Green penetrates water much better than Red/Orange/Yellow. Likely the only difference you are seeing is that Blue also creates a lot more perception of glare. Certain fish may be more visible with warmer light. Your glare issue is mostly the angle between the light and the water.

I would be tempted to play with polarization to see what effects/impact you can get.

Water cooling will of course help, but that is not really fixing the basic problem and you add in the cost/complexity of the water system without a lot of benefit. Total operational hours on something like this is pretty low, even for those super serious about the sport. So you lose 10-15% in light output, and your lights are good for 10,000 not 100000 hours ... really not a big deal.

 

[tjones96761]

Why can't you mount these underwater? You'd probably want to raise them before driving the boat, but that would help glare. Right?

the boat is always moving, it's the nature of the sport. underwater isn't a bad idea except most people don't want to cut holes in a $20k boat. That and the fear of dying when the light gets busted out on a rock or a stump and the boat starts taking on water...

 

[tjones96761]

From an optics standpoint, having lights that close to the waterline is about the worst thing you could ever do. A creative solution would be to get the lights high up in the air on a boom. I realize they would move a lot, but you would need less power to penetrate the water and the effectiveness farther from the boat will be way better. You could stick to narrower beam angles so you are not throwing light in places it is not helping with.

I am not sure there is a lot of benefit of being able to change the spectrum, at least from cool white to warm white. Blue/Green penetrates water much better than Red/Orange/Yellow. Likely the only difference you are seeing is that Blue also creates a lot more perception of glare. Certain fish may be more visible with warmer light. Your glare issue is mostly the angle between the light and the water.

I would be tempted to play with polarization to see what effects/impact you can get.

Water cooling will of course help, but that is not really fixing the basic problem and you add in the cost/complexity of the water system without a lot of benefit. Total operational hours on something like this is pretty low, even for those super serious about the sport. So you lose 10-15% in light output, and your lights are good for 10,000 not 100000 hours ... really not a big deal.

there are some boats that do a tower to get a better angle on the water. it's pretty regional, mostly offshore ocean guys do that. freshwater the fish are always in the shallows against the bank. overhanging trees along the shoreline pretty much rule out towers for 98% of bowfishing rigs.
water cooling isn't for increasing the life of the LED, it's for decreasing the size and weight of the housing. the 50w housings in the picture above weight 8 lbs each. because boats only float so much weight, everyone is looking for a lighter solution. you can decrease the available weight capacity for passengers by 10% pretty quickly with 8lb lights.