CGGorman's 99 Wide Glide build thread
#1241
#1243
#1244
You really only need a heat sink with thermal paste to the circuit board if the LEDs are on for an extended period of time and you are pushing a fair amount of current. How many LEDS and what current are you pushing for the driving lights? Those are probably the only ones o be concerned about except for maybe the brake light if you are pushing a lot of current there as well. I can help out specifically here as I am an electrical engineer that primarily does firmware design on LED lighting systems.
#1245
You really only need a heat sink with thermal paste to the circuit board if the LEDs are on for an extended period of time and you are pushing a fair amount of current. How many LEDS and what current are you pushing for the driving lights? Those are probably the only ones o be concerned about except for maybe the brake light if you are pushing a lot of current there as well. I can help out specifically here as I am an electrical engineer that primarily does firmware design on LED lighting systems.
#1246
The leads are good to dissipate heat with copper as long as the voltage and current are at proper levels for LEDs. For most applications where LEDs are on at a constant current though the lead temp can bleed to the resistor and cause it to short. As temp increases the junction voltage increases as well increasing current draw and making things heat up pretty quick. My work application is for pushing 2W through around 100 LEDs but the junction temp is transferable on a linear scale when downsizing the LED count. We deal with junction temps of up to 200F. It really depends on how bright you want them to be and how long they will on for a consideration on how much heat dissipation is needed.
#1247
Join Date: Jan 2014
Location: Depends on who wants to know.........and why.
Posts: 8,688
Received 1,901 Likes
on
1,109 Posts
#1248
I'm still extremely early in the planning and testing phase for the lights. I have a rough goal...that's about it.
2" round (or roughly equivalent non-round area) diode array.
Dual color, red/amber
Sequential/animated stop and turn
Brightness that's eye-catching but not blinding.
My current lights are 5W red LedEngin rated for around 500 lumen if memory serves. They're on custom copper heat sinks and I run them well below max current. They sit behind red Fresnel lenses. I'd have to go back and look. The details are earlier in this thread somewhere. I like the performance but really want the dual color and animation, plus the housings aren't the right style anymore.
The 5mm leds I got were rated at a supposed 34,000 mcd at about 35 degrees (10 lumen). 2.1 volts and 20mA max They're bright enough but the beam pattern is vey uneven and the array density won't be quite as high as I'd like. They'll work and haven't given up on them yet.
The PLCC leds are rated at a supposed 750 mcd at 120 degrees (2 lumen). 2.1V, 20mA max
I haven't decided on a substrate or fabrication method yet. Still want get hands on and see how they look and if I can even physically work with them. I did download ExpressPCB so I can play with laying out a custom board. We'll see.
I have resistors coming to run them series-parallel groups at both 15mA and 20mA (again, just testing). Just regular thru-hole metal film (1%) resistors that would be remotely wired.
I'm just a hack that likes to tinker so if any of this sounds stupid and you know a better way, please don't hesitate to correct me!
2" round (or roughly equivalent non-round area) diode array.
Dual color, red/amber
Sequential/animated stop and turn
Brightness that's eye-catching but not blinding.
My current lights are 5W red LedEngin rated for around 500 lumen if memory serves. They're on custom copper heat sinks and I run them well below max current. They sit behind red Fresnel lenses. I'd have to go back and look. The details are earlier in this thread somewhere. I like the performance but really want the dual color and animation, plus the housings aren't the right style anymore.
The 5mm leds I got were rated at a supposed 34,000 mcd at about 35 degrees (10 lumen). 2.1 volts and 20mA max They're bright enough but the beam pattern is vey uneven and the array density won't be quite as high as I'd like. They'll work and haven't given up on them yet.
The PLCC leds are rated at a supposed 750 mcd at 120 degrees (2 lumen). 2.1V, 20mA max
I haven't decided on a substrate or fabrication method yet. Still want get hands on and see how they look and if I can even physically work with them. I did download ExpressPCB so I can play with laying out a custom board. We'll see.
I have resistors coming to run them series-parallel groups at both 15mA and 20mA (again, just testing). Just regular thru-hole metal film (1%) resistors that would be remotely wired.
I'm just a hack that likes to tinker so if any of this sounds stupid and you know a better way, please don't hesitate to correct me!
Last edited by cggorman; 03-04-2019 at 03:53 PM.
#1249
The leads are good to dissipate heat with copper as long as the voltage and current are at proper levels for LEDs. For most applications where LEDs are on at a constant current though the lead temp can bleed to the resistor and cause it to short. As temp increases the junction voltage increases as well increasing current draw and making things heat up pretty quick. My work application is for pushing 2W through around 100 LEDs but the junction temp is transferable on a linear scale when downsizing the LED count. We deal with junction temps of up to 200F. It really depends on how bright you want them to be and how long they will on for a consideration on how much heat dissipation is needed.
What kind of resistor are you using to have it short out under temp. I usually see them open up under heat and current. For me they made good fuses.. Solder flowing under the resistor from the heat? Temps would have to be really high for that.
You using a switched mode power supply to drive a constant current? I assume that's why your the supply current goes up as the forward voltage drop increases.
2 Watts for 100 LEDS is only 20 mW for each LED (assuming they draw all the power) They must be very small and clustered all together.
#1250
Interesting stuff.
What kind of resistor are you using to have it short out under temp. I usually see them open up under heat and current. For me they made good fuses.. Solder flowing under the resistor from the heat? Temps would have to be really high for that.
You using a switched mode power supply to drive a constant current? I assume that's why your the supply current goes up as the forward voltage drop increases.
2 Watts for 100 LEDS is only 20 mW for each LED (assuming they draw all the power) They must be very small and clustered all together.
What kind of resistor are you using to have it short out under temp. I usually see them open up under heat and current. For me they made good fuses.. Solder flowing under the resistor from the heat? Temps would have to be really high for that.
You using a switched mode power supply to drive a constant current? I assume that's why your the supply current goes up as the forward voltage drop increases.
2 Watts for 100 LEDS is only 20 mW for each LED (assuming they draw all the power) They must be very small and clustered all together.