Thanks for the responses folks, while this thread has not gone in the direction I intended it too I think the information it will provide to people like myself is invaluable.
I'm really keen to learn and solve the problems with my lights but I need some info from you to help me do that.
When talking about heatsink mass what and where are we actually talking about. One of the concepts I think I'm clear on is providing a clear route for heat to travel from the led's to the outside shell of the light for disipation (wish I could get the ispell check to work in this window :thinking
into the surrounding air.
Using my cateye housing the problem I had was the housing bezel was a bit bigger (43mm dia) than the led mouting MCPCB (34mm dia). The fins inside the housing are 33mm diameter at their smallest. I mounted the MCPCB onto a thin aluminium plate with artic silver epoxy and coated the fins, edge and back of the plate with heat sink compound to maximise any available contact surface between the led mountings and the housing. When the light is put together and the bezel screws on it fits just about perfectly so that the bezel tightens against the optics pushing the led mounting tight up against the fins on the housing.
So the questions I'm now asking myself and would really like opinions on are
1. what are the problems with this that are causing experienced light builders to think it will boil like a kettle?
2. how 'wide' does the thermal path NEED to be? How much heat conductive material needs to form the path? What is the balance between having a nice wide path and also weight as more material = more weight. Ideally of course the led's would be backed onto a heatsink open to the flow of air. I've seen designs like this but I think there are other problems with this design route when the light is going to be used in a mountain bike enviroment and its not for me.
3. how can I improve the design using the materials and tools available to me? Unfortunately I don't know anyone with a tool shop so for me its all done on the kitchen table with very basic tools and materials.
4. how do I ascertain if my design is working effectively enough? when thinking about heat transfer how do I effectively measure the temperature at the led's and at the housing? I can't take the bezel off to get at the led's as that would reduce the contact of the led's against the housing so reducing the efficiency and also exposing the led's to the air so providing a cooling route that would not be there under normal conditions. At the moment all I can tell is the housing gets ******* hot quite quickly when running 1000mA on the kitchen table but stays pretty cool when out riding.
One of the reasons I started this thread was to help people like myself who would like to build lights but haven't got access to much in the way of tools. The kit from cutter has suddenly made it very simple for people like myself to start putting lights together and we're learning racks along the way too which I'm really excited about (my friends all roll their eyes and shake their heads about how excited I am about it). The problem us newbies without tools have is housings. By buying an off the shelf housing we can play the build your own lights game. Maybe this will lead us into more ambitious and creative projects, maybe this will be a one off that just leaves us with an effective light that we built ourselves.
So what are peoples opinions on what makes an effective housing for an led mountain bike light? Does it need to have a minimum amount of material or surface area? Take for instance the two housings I started this post with, the lumicycle one and the cateye one.
The lumi cycle housing is just a can. This makes it really light weight and simple but does this mean it won't have enough material in it to disapate the heat, or maybe it hasn't got enough surface area? People are using these though and I haven't heard of burnt out boards and leds as yet so where is the line drawn between "perfect" and "effective enough"? Perhaps they just havent had enough time to burn things out yet though and they will soon.
The cateye housing is a different design. Its a aluminium casting and has a lot more material in it than the lumicycle. I only have an inacurate set of scales but I think they were in the region of 230g with the halogen bulb in so probably about 200g of aluminium. Because its a casting they have also been able to put some fins on the outside to increase the surface area all be it a very little bit.
Bottom line is I cant see why my design prompted the kettle boiling comments as yet but I'm really interested in finding out why and solving the problem if there is one.
So I can feel myself coming to a long rambling post asking a lot of questions. I hope some of you can find the time to answer some of them for me, I'd really appriciate it. As I've said before I'm really keen to learn.
Thanks
Ian
Its great to have a thread like this..well done...
BUT a lack of understanding of just how much 'heatsink' mass is needed to cool 3 LEDs @ 750ma-1A, may mean that some housings presented in this thread, are
undesirable. thus the 'teapot whistling' post...
unlike most forums, which spout off misinformed rubbish, we here @ CPF like to present the true facts.
sure I don't know everything, but I can tell you that you need a lot more thermal mass than both pics I have seen so far...
Iggs, maybe you should read the info given by the LED company's, re thermal mass required, before posting anymore LED cooking designs, would not like to give the good 'just starting' people the wrong idea / understanding...
we are all working towards the same goal. Mate, I am not knocking you, but trying to give you a shove in the right direction...
keep up the great effort
Best
Ktronik
