Results 1 to 20 of 20
  1. #1
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    90

    Reducing weight of alu lighthead

    I would like to reduce the weight of this light towards 100gr. I will make the fins slightly deeper. That and making the back a little thinner could maybe save 10-15gr. I am considering to mill some slots straight trough the fins, perhaps two on either side of the switch (at 45, 135, 225 and 315dgr, 0 being up). It will shave off some weight, possibly improve air exchange and slightly reduce surface area.

    Does anyone know of existing lights with such deep slots? It would be similar to the ones found at Magicshine, but as deep as the fins, 9mm. I would like to see what it would look like, and I am not very good at using CAD programs for visualization. How wide should they be?
    Attached Thumbnails Attached Thumbnails Reducing weight of alu lighthead-r0011361.jpg  

    Reducing weight of alu lighthead-r0011362.jpg  

    Last edited by kmjelle; 01-29-2010 at 03:14 PM.

  2. #2
    mtbr member
    Reputation:
    Join Date
    May 2005
    Posts
    3,602
    The deeper fins will be fine. I have made most of my fins by cutting grooves that are 1.5 to 2mm wide and leaving 1.5mm for the fin thickness. That has worked well. One light I did had the grooves only 1 mm wide and the fins were 2mm thick. That light seems to not dissipate heat as well as my others, but still only requires very slight movement through the air to keep it under 50C. I usually make the wall thickness on my housings .5 to 2mm except on the surface the LEDs mount to. That I leave 4mm thick. I think I would be more inclined to mill flats on the sides instead of slots. There would be fewer cuts to make and no sharp edges. Don't remove too much area though in the quest for light weight. Allow 6.5 sq. cm per watt of LED for adequate heat dissipation.

  3. #3
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    90
    I have a program, Frigus primore, that calculates optimal distance between fins when only natural convection occurs. It seems to be 7-9mm between fins, 1-1,5mm wide fins at 10mm fin height. But I wanted to try less distance because I use the light in various moving activities or the wind is usually blowing. If the fins are too close they will end up transfering heat to each other, if there is no forced air movement.

    I didn`t quite understand what milling flats on the sides would look like. Do you have an example?

  4. #4
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    90
    I was thinking of slots like the ones here post # 82 by Yetibetti. Only deeper and all the way from front to back.

  5. #5
    mtbr member
    Reputation:
    Join Date
    May 2005
    Posts
    3,602
    The slots make the light have too many sharp edges IMO. They resemble teeth on a saw. I am not too good at drawing a jpeg, but this is what I was trying to mean. Imagine this is an end view of your light. Cut off at where the red lines are.
    Attached Thumbnails Attached Thumbnails Reducing weight of alu lighthead-light.jpg  


  6. #6
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    90
    That could be an option. I have about 400 cm2 of outer surface, counting in both sides of each fin. The minimum recommended has been said to be 6cm2 per watt: 6cm2 per watt x 24watt= 160cm2. Meaning that I could double the space between each rib, and still be on the safe side. Where does this 6 number come from?

  7. #7
    mtbr member
    Reputation:
    Join Date
    May 2005
    Posts
    3,602
    The area figure I stated came from something I read a couple of years ago. I have tried to stay in that area with the lights I have built. In still, room temp air they tend to heat to ~60C then level off. The ones with Taskled drivers I set to switch to low at 50C. That usually happens after 5-7 minutes in still, room temp air. I have checked the temps with a thermocouple and an IR device to look for even heat distribution. In use riding at 7-9C air they lights stay just luke warm.

  8. #8
    mtbr member
    Reputation: BrianMc's Avatar
    Join Date
    Jan 2010
    Posts
    4,272
    Heatsink to air area per watt of LED seems to be a set of rules of thumb for minimum ratios. Guidelines not laws.

    In interior lighting close in fixtures with some ventilation and no forced air, the figure I saw was 7-8 cm/watt. For the same in a light open to air flow in the room, 6 cm/watt. For many road bikes a figure of 2 cm/watt is considered to be decent, but they will get hot at traffic lights it not powered down. The MagicShine is pushing towards the 1 cm per watt especially when its poor thermal path from heat sink to body is factored in. Yet these live though at 2.4 not max of 2.8 amps.

    Assuming you are below max spec junction temp., a low resistance thermal path to the air interface is at least as important as the area of that interface in keeping the LED junction temps down and the efficiency of turning electrons into light, up. That doesn't get any better than in a light like yours, with a single milled heat sink-body when the star/led is mounted well.


    Finned area is often not as effective as a the same area of smooth air interface. When sitting still, fins increase the thickness of the air boundary layer and its insulating effects compared to the same area in a smooth body. Close fins of course radiate heat at each other but more importantly, the air between becomes more of a thermal path between close fins than to the air beyond the boundary layer of the light. The surface boundary layer is swept away by very low air speeds and the cooler air increases the heat flow to the air like a steeper hill increase speed. This happens first at the tips of your fins but the heat has to travel the height of the fins. The deeper and closer the fins, the higher the air speed needed to suck out the air between them. So densely packed copper fins on CPU heat sinks have High volume (air speed) fans. If the fins are deep enough and close enough, you won't be able to ride fast enough to move that air from between them. Yours aren't THAT deep or close but the effect will be there at low speeds.

    So it is better when using deep fins, that they are oriented in the direction of air flow. This is a PITA to mill so few do it this way in round lights. There is also a significant resistance in heat flow to the outer surface of long fins.

    So yes, you can reduce the height of your fins with little loss in cooling while riding and it may even be a bit cooler (counter-intuitive) running in a room test.

    I think avoiding the saw tooth effect, and rounding over the ends of the fins for a smoother transition would be a decent solution. But maybe it would be better to consider either round or oval lightening holes cut through front to back? Cool look and a cooling effect. That shouldn't be hard to do.

  9. #9
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    90
    I ended up just anodizing the light. I recently bought a mini lathe so I will experiment more with my next builds. The weight stopped at 135gr with switch, gland and mounting bracket. A battery status led will be mounted in front of the switch. It will be interesting to see how it handles heat.
    Attached Thumbnails Attached Thumbnails Reducing weight of alu lighthead-r0011369.jpg  

    Reducing weight of alu lighthead-r0011370.jpg  

    Last edited by kmjelle; 01-30-2010 at 06:11 PM.

  10. #10
    mtbr member
    Reputation: BrianMc's Avatar
    Join Date
    Jan 2010
    Posts
    4,272
    Gorgeous in blue!

    A few extra grams look pretty good to me!

  11. #11
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    90

    Cooling of light

    The highest graph shows heat dissipation from a piece of aluminium 180x75mm, with 10mm high cooling fins, 1mm wide, at different pitches. Temperature difference between aluminium and ambient is 60 dgr C. There is no active movement of air (no fan og bike riding). The data is made with this software. Should we belive that this gives a realistic picture?
    Attached Images Attached Images  

  12. #12
    mtbr member
    Reputation: BrianMc's Avatar
    Join Date
    Jan 2010
    Posts
    4,272
    Nice software. Looks like increasing pitch from 2 to 7.5 mm helps, but is no longer a factor at this temp differential, once fins are that far apart and reduced area of fins becomes more important than convection.

    How do 1 mm thick 1 cm high fins look at pitches 1 and 2 mm apart when the air temp difference is only 20C? A differential of 60 C is a high thermal driving force. Looks like I was wrong, but the 20C and narrow pitches will put the last nails in. Good job.

  13. #13
    mtbr member
    Reputation:
    Join Date
    May 2005
    Posts
    3,602
    Quote Originally Posted by kmjelle
    Should we belive that this gives a realistic picture?
    While it may give a realistic picture of how one's light might perform while resting or repairing a flat, it does not represent the condition that we want our lights to perform well at. Can the software accept an input for air movement at differing velocities and orientation to fin directions? That would be very interesting to see results of for deciding on light design.

  14. #14
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    90
    Heat dissipation at different pitches and wind speeds. I have set the temperature difference at 60C, as I live above the arctic circle. Low pitch is efficient at high speed.

    I also stripped of the blue anodizing, made some changes and gave it a red colour. The naked alu head weighs only 40gr, but to be honest the fins are too thin now. You can pretty easily flex them with your fingers. About 0.8mm would be a descent thickness.
    Attached Thumbnails Attached Thumbnails Reducing weight of alu lighthead-r0011386.jpg  

    Reducing weight of alu lighthead-varmeleding-ved-ulik-fart.gif  


  15. #15
    mtbr member
    Reputation: yetibetty's Avatar
    Join Date
    Dec 2007
    Posts
    1,781
    Looks familiar I like 1.4mm fins, thin & strong and just happen to be the thickness of the average parting tool.

  16. #16
    mtbr member
    Reputation: troutie-mtb's Avatar
    Join Date
    Sep 2007
    Posts
    3,062
    Just maybe you got a bit obsessed with weight on that one and as you are using it for a head light I see you reasoning .

    Nothing scientific here just good old hands on testing
    where you live I can see your light being kept cool no problem but in my experience a certain amount of mass is good also to act as a true sink in times when stationary
    and with the Flex`s making use of the thermal sensing

    My seven up light weighs in at 140 grammes built and working



    and the guy who bought the first one used like yours on his head in Finland for dog sled racing was more than happy with the weight and reported back that even on full whack it stayed cool all the time ( not that he needed full anytime except for showing off )
    He said the lupine betty head was 150 grammes

    yes mine could lose a bit of weight but it also needs to work in Saudi arabia with hi ambients

    your fins in my opinion are way too thin to carry heat to the tips for transfer to the air

  17. #17
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    90
    Yes these fins are too thin og and weak. Acording to the software referred to earlier, about 1mm fins and 2-3mm space between seems to be optimum with air actively moving around the light.

    Troutie
    Are your 7 up lights made in a mill or lathe, or both? The oval shape seem hard to achieve in a lathe.

  18. #18
    mtbr member
    Reputation: troutie-mtb's Avatar
    Join Date
    Sep 2007
    Posts
    3,062
    Quote Originally Posted by kmjelle
    Yes these fins are too thin og and weak. Acording to the software referred to earlier, about 1mm fins and 2-3mm space between seems to be optimum with air actively moving around the light.

    Troutie
    Are your 7 up lights made in a mill or lathe, or both? The oval shape seem hard to achieve in a lathe.

    Thet were done on a cnc mill by a very nice machinist who only charged and arm and half a leg for them

  19. #19
    mtbr member
    Reputation: Goldigger's Avatar
    Join Date
    Nov 2009
    Posts
    1,669
    Quote Originally Posted by troutie-mtb
    Thet were done on a cnc mill by a very nice machinist who only charged and arm and half a leg for them
    How much does an arm and half a leg cost Troutie?
    If it adds any value both my legs have titanium rods..

  20. #20
    mtbr member
    Reputation:
    Join Date
    Mar 2009
    Posts
    35
    What are the physical dimensions? (excluding the lip thing on the front)

Members who have read this thread: 0

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

THE SITE

ABOUT MTBR

VISIT US AT

© Copyright 2018 VerticalScope Inc. All rights reserved.