---------------------------------------------------------- AeroElectric-List Digest Archive --- Total Messages Posted Sun 11/01/20: 25 ---------------------------------------------------------- Today's Message Index: ---------------------- 1. 12:12 AM - Matronics List Fund Raiser Postponed During 2020... (Matt Dralle) 2. 12:59 AM - Re: DIY LED landing light -- Wing skin as heat sink? (Todd Bartrim) 3. 01:02 AM - Re: DIY LED landing light -- Wing skin as heat sink? (Todd Bartrim) 4. 05:47 AM - Re: Lead Acid Batteries (Robert L. Nuckolls, III) 5. 05:50 AM - Re: DIY LED landing light -- Wing skin as heat sink? (Finn Lassen) 6. 06:17 AM - Re: DIY LED landing light -- Wing skin as heat sink? (Robert L. Nuckolls, III) 7. 06:26 AM - Re: DIY LED landing light -- Wing skin as heat sink? (Finn Lassen) 8. 06:30 AM - Re: DIY LED landing light -- Wing skin as heat sink? (Robert L. Nuckolls, III) 9. 06:33 AM - Re: DIY LED landing light -- Wing skin as heat sink? (Robert L. Nuckolls, III) 10. 07:01 AM - Re: DIY LED landing light -- Wing skin as heat sink? (Finn Lassen) 11. 08:12 AM - Re: DIY LED landing light -- Wing skin as heat sink? (Ernest Christley) 12. 09:07 AM - Re: DIY LED landing light -- Wing skin as heat sink? (Robert L. Nuckolls, III) 13. 10:06 AM - Re: DIY LED landing light -- Wing skin as heat sink? (Robert L. Nuckolls, III) 14. 10:54 AM - Re: DIY LED landing light -- Wing skin as heat sink? (Finn Lassen) 15. 11:12 AM - Re: DIY LED landing light -- Wing skin as heat sink? (Finn Lassen) 16. 12:37 PM - Re: DIY LED landing light -- Wing skin as heat sink? (Charlie England) 17. 01:24 PM - Re: DIY LED landing light -- Wing skin as heat sink? (Ernest Christley) 18. 03:38 PM - Microair 760 transceivers for sale (Robert L. Nuckolls, III) 19. 03:40 PM - Re: Breaker then pump failure (Robert L. Nuckolls, III) 20. 03:44 PM - Re: DIY LED landing light -- Wing skin as heat sink? (Robert L. Nuckolls, III) 21. 04:00 PM - Re: DIY LED landing light -- Wing skin as heat sink? (Finn Lassen) 22. 04:41 PM - Re: DIY LED landing light -- Wing skin as heat sink? (Finn Lassen) 23. 04:53 PM - Re: ADS-B and Transponder antenna... (Eric Page) 24. 06:02 PM - Re: DIY LED landing light -- Wing skin as heat sink? (Charlie England) 25. 06:16 PM - Re: DIY LED landing light -- Wing skin as heat sink? (Ernest Christley) ________________________________ Message 1 _____________________________________ Time: 12:12:50 AM PST US From: Matt Dralle Subject: AeroElectric-List: Matronics List Fund Raiser Postponed During 2020... Dear Listers, Normally, in November I hold a List Fund Raiser where I ask Listers to make a Contribution to support the operation and upkeep of these List resources. In light of the happenings during this 2020 season, I have decided to postpone the Fund Raiser this year and encourage members to focus on their families and personal needs. I want to thank everyone that has supported the Lists in the past. Despite the uncertain times we're experiencing right now, I want to assure everyone that I fully intend to continue supporting the Lists as I always have; with no anticipated degradation in service levels. Happy Holidays to all my friends around the world! Matt Dralle Matronics List Manager ________________________________ Message 2 _____________________________________ Time: 12:59:47 AM PST US From: Todd Bartrim Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? Hi Finn; This may not be using the skin as a heat sink, but it works great and is lightweight. I bought some of these lights from Amazon and took them apart to make my landing lights. https://www.amazon.ca/Lightfox-Driving-Offroad-Waterproof-Warranty/dp/B01M9 8GWR2/ref=sr_1_2_sspa?crid=31MNRKKNOSJAA&dchild=1&keywords=led+spot +lights+offroad&qid=1604214923&sprefix=led+spot+%2Caps%2C274&sr=8-2-s pons&psc=1&spLa=ZW5jcnlwdGVkUXVhbGlmaWVyPUEySTdLUVhZSDFQQTEmZW5jcnlwdGV kSWQ9QTA2NDM1NzcyUVdZV1QyR0tJS1dRJmVuY3J5cHRlZEFkSWQ9QTAwNzU1NTUzSzVRRTYzNT ZSUjZHJndpZGdldE5hbWU9c3BfYXRmJmFjdGlvbj1jbGlja1JlZGlyZWN0JmRvTm90TG9nQ2xpY 2s9dHJ1ZQ= this is from Amazon's Canada site, but I'm sure you can find them even cheaper on the US site. Then I took them apart as you can see in these pictures posted here on Google photos https://photos.app.goo.gl/S68h3zpVy7RVJDTT8 I can't remember how much weight I discarded from the parts I removed, but it was most of the weight of the light. The mounting plate was originally part of the Duckworths landing light that I wasted my money on. That is a simple part to make. I have 2 of those mounted mid wing and are pointed down for use as taxi lights, then I mounted 2 into the wingtips (required some fibreglass work) and have them pointed straight ahead for landing lights. These lights are very durable and bright. I must have bought more than a dozen so far as I've put some on my truck, my son's Jeep, my skidsteer, 4 on the plane and probably the harshest duty is the pair I put on my ATV, which are thoroughly abused. They are extremely bright in the bush with very little current draw so my ATV has no problem running them continuously. I think those ones are at least 4 years old and are still working great despite the abuse. It looks like there are even better versions of these lights now, that appear to be the same construction so you can still dismantle them to get the good parts only. Todd Bartrim RV9 13Bturbo On Sat, Oct 31, 2020 at 8:20 PM Finn Lassen wrote: > > Might say the most critical thing for LED life and high lumens output is > conducting heat away from the LED. > > That usually means attaching heavy heat sinks. > > Could we save weight by using the wing skin as a heat sink? > > I understand it would present problems in adjusting (pointing) the light. > > But, could we bond/attach the LED to a (alum?) plate that is then > bonded/riveted/screwed to the leading edge wing skin? > > Let's say we use a 100W LED (~13,000 lumens). We'd have to remove 70 to > 80W of heat from the LED. LED doesn't like more than 150=C2=B0C and it wo uld > be better to keep it cooler. LED area 34x34mm. > > I don't have a good grasp of thermal conductivity (conductance?). How > thick would the mount plate have to be? How much area would be needed > where the mount plate contacts the wing skin (using some kind of heat > sink compound/adhesive between the plate and skin)? > > Am I right in assuming the plate has to be thickest were the LED is > mounted and then taper down in thickness as it expands out in width and > height? > > Perhaps tapered layers of carbon fiber (with thermal epoxy) would be > better as the LED mount plate? (After a diamond, graphene or graphite > apparently has the best thermal conductivity.) > > Any (thermal) engineers here that could enlighten me? > > Finn > > > --- > This email has been checked for viruses by Avast antivirus software. > https://www.avast.com/antivirus > =========== =========== =========== =========== =========== > > ________________________________ Message 3 _____________________________________ Time: 01:02:55 AM PST US From: Todd Bartrim Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? Ooops, that was the link for my own collection of reference photos, This is the one you need to see how I did the lights https://photos.app.goo.gl/wQvyosNK1QBFjz7c8 Todd Bartrim On Sun, Nov 1, 2020 at 12:54 AM Todd Bartrim wrote: > Hi Finn; > This may not be using the skin as a heat sink, but it works great and i s > lightweight. > I bought some of these lights from Amazon and took them apart to make my > landing lights. > > https://www.amazon.ca/Lightfox-Driving-Offroad-Waterproof-Warranty/dp/B01 M98GWR2/ref=sr_1_2_sspa?crid=31MNRKKNOSJAA&dchild=1&keywords=led+sp ot+lights+offroad&qid=1604214923&sprefix=led+spot+%2Caps%2C274&sr=8-2 -spons&psc=1&spLa=ZW5jcnlwdGVkUXVhbGlmaWVyPUEySTdLUVhZSDFQQTEmZW5jcnlwd GVkSWQ9QTA2NDM1NzcyUVdZV1QyR0tJS1dRJmVuY3J5cHRlZEFkSWQ9QTAwNzU1NTUzSzVRRTYz NTZSUjZHJndpZGdldE5hbWU9c3BfYXRmJmFjdGlvbj1jbGlja1JlZGlyZWN0JmRvTm90TG9nQ2x pY2s9dHJ1ZQ= > this is from Amazon's Canada site, but I'm sure you can find them even > cheaper on the US site. > Then I took them apart as you can see in these pictures posted here on > Google photos > https://photos.app.goo.gl/S68h3zpVy7RVJDTT8 > I can't remember how much weight I discarded from the parts I removed, bu t > it was most of the weight of the light. > The mounting plate was originally part of the Duckworths landing light > that I wasted my money on. That is a simple part to make. I have 2 of tho se > mounted mid wing and are pointed down for use as taxi lights, then I > mounted 2 into the wingtips (required some fibreglass work) and have them > pointed straight ahead for landing lights. > These lights are very durable and bright. I must have bought more than a > dozen so far as I've put some on my truck, my son's Jeep, my skidsteer, 4 > on the plane and probably the harshest duty is the pair I put on my ATV, > which are thoroughly abused. They are extremely bright in the bush with > very little current draw so my ATV has no problem running them > continuously. I think those ones are at least 4 years old and are still > working great despite the abuse. > It looks like there are even better versions of these lights now, that > appear to be the same construction so you can still dismantle them to get > the good parts only. > > Todd Bartrim > RV9 > 13Bturbo > > On Sat, Oct 31, 2020 at 8:20 PM Finn Lassen wrote: > > >> >> Might say the most critical thing for LED life and high lumens output is >> conducting heat away from the LED. >> >> That usually means attaching heavy heat sinks. >> >> Could we save weight by using the wing skin as a heat sink? >> >> I understand it would present problems in adjusting (pointing) the light .. >> >> But, could we bond/attach the LED to a (alum?) plate that is then >> bonded/riveted/screwed to the leading edge wing skin? >> >> Let's say we use a 100W LED (~13,000 lumens). We'd have to remove 70 to >> 80W of heat from the LED. LED doesn't like more than 150=C2=B0C and it w ould >> be better to keep it cooler. LED area 34x34mm. >> >> I don't have a good grasp of thermal conductivity (conductance?). How >> thick would the mount plate have to be? How much area would be needed >> where the mount plate contacts the wing skin (using some kind of heat >> sink compound/adhesive between the plate and skin)? >> >> Am I right in assuming the plate has to be thickest were the LED is >> mounted and then taper down in thickness as it expands out in width and >> height? >> >> Perhaps tapered layers of carbon fiber (with thermal epoxy) would be >> better as the LED mount plate? (After a diamond, graphene or graphite >> apparently has the best thermal conductivity.) >> >> Any (thermal) engineers here that could enlighten me? >> >> Finn >> >> >> --- >> This email has been checked for viruses by Avast antivirus software. >> https://www.avast.com/antivirus >> =========== =========== =========== =========== =========== >> >> >> >> ________________________________ Message 4 _____________________________________ Time: 05:47:47 AM PST US From: "Robert L. Nuckolls, III" Subject: Re: AeroElectric-List: Lead Acid Batteries > > >Second question is about allowing a lead acid >battery to discharge all the way down to zero=85or >if it had gotten severely discharged a few >timmes=85it will no longer be serviceable.=C2 Is >this accurate?=C2 If so why would a good battery >become junk just because you =9Cleft the lights >on=9D and drained it down to zero? It's not a 'sudden death' phenomenon. In fact, to achieve TSO qualification on aircraft batteries, one of the tests is to completely discharge the battery, place a dead short on it for a period of time (don't recall the interval off top of my head) and the conduct a 'recovery charging protocol'. But leaving a lead-acid battery in a discharged state accelerates corrosion of the positive plate which ultimately trashes the battery. >Third question is what is the difference between >a =9Ccar battery=9D and a =9Cmarine battery/ deep cycle battery=9D? Here's an excellent article that speaks to that topic . . . among others. https://tinyurl.com/y69ltgh6 By the way, there's a constellation of excellent articles on the characteristics and maintenance of various battery chemistries on this website: https://tinyurl.com/yalty6y7 Bob . . . ________________________________ Message 5 _____________________________________ Time: 05:50:34 AM PST US Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? From: Finn Lassen That would introduce drag. Name of the game here is no drag and lowest possible weight for max light output (lumens) directed in a narrow beam that can reach 3,000 feet or more down a deer infested runway. If it wasn't for needing the reflector to concentrate the light in a beam, we could simply mount the LED chip flush with the wing skin, using an alum backing plate (about the thickness of the one the LED chips comes bonded to, I guess) and the wing skin would be a perfect heat sink. My thermodynamic question comes into play as we move the LED chip further back from the leading edge wing skin. Surely someone has experience in designing heat sinks and what's required in terms of conducting the heat from the point where a chip is mounted and out into the fins exposed to air. Hopefully the thousands of heat sink design you see are not all the result of trial and error. There must be some physics (thermodynamics?) behind it. Finn On 11/1/2020 1:38 AM, Charles Davis wrote: > > > How about ducting some of the slip-stream directly onto the heat-sink, > you could then use a MUCH smaller/lighter sink than relying on ambient > air inside the fitting > > > On 01/11/20 05:08 am, Finn Lassen wrote: >> >> >> Might say the most critical thing for LED life and high lumens output >> is conducting heat away from the LED. >> >> That usually means attaching heavy heat sinks. >> >> Could we save weight by using the wing skin as a heat sink? >> >> > > --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ________________________________ Message 6 _____________________________________ Time: 06:17:14 AM PST US From: "Robert L. Nuckolls, III" Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? > >Let's say we use a 100W LED (~13,000 lumens). >We'd have to remove 70 to 80W of heat from the LED. >LED doesn't like more than 150=C2=B0C and it would >be better to keep it cooler. LED area 34x34mm. Don't think it's quite that bad. LED efficiency is on the order of 35 to 40% which would dump 60 to 65 watts of heat. >I don't have a good grasp of thermal >conductivity (conductance?). How thick would the mount >plate have to be? How much area would be needed >where the mount plate contacts the wing >skin (using some kind of heat sink >compound/adhesive between the plate and skin)? The flow of heat energy is analogous to the flow of electrical energy. Temperature rise is 'Volts'. Thermal resistance is 'Ohms' and Watts is still watts. Configuring thermal management for an LED is identical to that of say a power transistor. The working part of the LED has a known thermal resistance between the semiconductor junction and the heat sinking surface or case. It can be stated in degrees C per watt. So if you have a junction dissipating 100 watts of heat and it's packaged thermal resistance is say 1C/W then the junction is going to operate at 100C HIGHER than the heat sinking surface. If the thing is limited to 150C then you have to maintain the case at 50C. There will be a thermal resistance associated with the mounting of the device to the heat sink . . . add say 0.25C/W. Adding up the resistances we're at 1.25C/W . . . 100 Watts will drop 125C so now your heatsink has to operate at 25C. These resistance numbers are pulled out of my hat . . . but you get the picture. Now we get to the proposed 'heat sink'. Just how good is it. VERY difficult to calculate if complex like structure of an airplane. Best way to test it is bolt a power resistor to the LED mounting surface and cause it to dissipate X watts. Thermocouple the area under the resistor to get degrees C/Watt of thermal resistance. Now to throw a really big monkey wrench into the study. AIR FLOW has a PROFOUND influence on thermal resistance. Those tiny chips in your computer dissipate a bucket load of heat in terms of their mass. The heat sink on your CPU has lots of SURFACE area and a FAN. The only time your LED nav lights are at risk for thermal abuse is while sitting on the ground or taxi. If I were designing an LED fixture, I'd probably include over temperature sensing to dim the lighting to prevent thermal damage. Once airborne, you'll have plenty of air flow. I suspect the 'big guys' in aviation lighting already do this. After all, there are TSO requirements for thermal robustness that have to be met and by now, they're very good at it. If you're looking at a DIY LED illumination project, start with a study of how the big guys do it. When assigned a task for development of a new product in years past, one of the first things I would do it search out the patents on legacy technology. HISTORY can give you a leg up on avoiding exercises in re-invention. >Am I right in assuming the plate has to be >thickest were the LED is mounted and then taper >down in thickness as it expands out in width and height? No . . . thermal resistance of base material, surface area and AIRFLOW . . . but without a doubt, include ELECTRONIC overheat protection. Bob . . . ________________________________ Message 7 _____________________________________ Time: 06:26:31 AM PST US Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? From: Finn Lassen Thanks Todd. There are a multitude of LED solutions out there. I have a $20 4oz Trustfire 3T6 flashlight head that puts out 1,300 lumens (actually capable of 3,300 if I can get to and modify the internal driver in it and add proper cooling). However , the 10 watt Cree XM-L2 or similar LED chips can now be had for $1.64 each. That's about 1,160 lumens if fed 10 watts. The newer XHP35, XHP72, etc. may or may not be slightly more efficient, but puts out light from a more dense location but are more expensive. The point here is that the LED solutions all come with (heavy) heat sinks. Even the drivers will need some cooling (think mounting the driver ICs/diodes on a wing rib). 70 to 80% of the power being fed to the LED chips needs to be dissipated as heat. That's right, LEDs are not more than 20 to 30% efficient! All commercial and DIY LED landing light solutions I've seen so far come stand-alone. But why not design it into the aircraft? I guess not everyone is a weight freak like me. Reason for using flashlight (torch) heads is they come with reflectors -- trick is to find ones with a narrow beam and low loss reflector and not having to pay for battery holder tube, heat sink, batteries etc. etc. But they all come with integrated drivers and heat sinks. Some torches come with adjustable beam, but from what I've read they have rather lossy reflectors. Now, a reflector like this https://www.stratusleds.com/module should probably do it, although I don't know the beam width. That's what started me thinking about using the LE wing skin as the heat sink. Needs a mount plate that can transfer the 70 or 80 watts of heat from the 34x34mm LED chip base to a sufficient area of wing skin. What's the heat "resistance" of 0.025 thick alum? Finn On 11/1/2020 2:58 AM, Todd Bartrim wrote: > Ooops, that was the link for my own collection of reference photos, > This is the one you need to see how I did the lights > https://photos.app.goo.gl/wQvyosNK1QBFjz7c8 > Todd Bartrim > > > On Sun, Nov 1, 2020 at 12:54 AM Todd Bartrim > wrote: > > Hi Finn; > This may not be using the skin as a heat sink, but it works > great and is lightweight. > I bought some of these lights from Amazon and took them apart to > make my landing lights. > https://www.amazon.ca/Lightfox-Driving-Offroad-Waterproof-Warranty/dp/B01M98GWR2/ref=sr_1_2_sspa?crid=31MNRKKNOSJAA&dchild=1&keywords=led+spot+lights+offroad&qid=1604214923&sprefix=led+spot+%2Caps%2C274&sr=8-2-spons&psc=1&spLa=ZW5jcnlwdGVkUXVhbGlmaWVyPUEySTdLUVhZSDFQQTEmZW5jcnlwdGVkSWQ9QTA2NDM1NzcyUVdZV1QyR0tJS1dRJmVuY3J5cHRlZEFkSWQ9QTAwNzU1NTUzSzVRRTYzNTZSUjZHJndpZGdldE5hbWU9c3BfYXRmJmFjdGlvbj1jbGlja1JlZGlyZWN0JmRvTm90TG9nQ2xpY2s9dHJ1ZQ= > this is from Amazon's Canada site, but I'm sure you can find them > even cheaper on the US site. > Then I took them apart as you can see in thesepictures posted > here on Google photos > https://photos.app.goo.gl/S68h3zpVy7RVJDTT8 > I can't rememberhow much weight I discarded from the parts I > removed, but it was most of the weight of the light. > The mounting plate was originally part of the Duckworths landing > light that I wasted my money on. That is a simple part to make. I > have 2 of those mounted mid wing and are pointed down for use as > taxi lights, then I mounted 2 into the wingtips (required some > fibreglass work) and have them pointed straight ahead for landing > lights. > These lights are very durable and bright. I must have bought more > than a dozen so far as I've put some on my truck, my son's Jeep, > my skidsteer, 4 on the plane and probably the harshest duty is the > pair I put on my ATV, which are thoroughly abused. They are > extremely bright in the bush with very little current draw so my > ATV has no problem running them continuously. I think those ones > are at least 4 years old and are still working great despite the > abuse. > It looks like there are even better versions of these lights > now, that appear to be the same construction so you can still > dismantle them to get the good parts only. > > Todd Bartrim > RV9 > 13Bturbo > > On Sat, Oct 31, 2020 at 8:20 PM Finn Lassen > wrote: > > > > > Might say the most critical thing for LED life and high lumens > output is > conducting heat away from the LED. > > That usually means attaching heavy heat sinks. > > Could we save weight by using the wing skin as a heat sink? > > I understand it would present problems in adjusting (pointing) > the light. > > But, could we bond/attach the LED to a (alum?) plate that is then > bonded/riveted/screwed to the leading edge wing skin? > > Let's say we use a 100W LED (~13,000 lumens). We'd have to > remove 70 to > 80W of heat from the LED. LED doesn't like more than 150C and > it would > be better to keep it cooler. LED area 34x34mm. > > I don't have a good grasp of thermal conductivity > (conductance?). How > thick would the mount plate have to be? How much area would be > needed > where the mount plate contacts the wing skin (using some kind > of heat > sink compound/adhesive between the plate and skin)? > > Am I right in assuming the plate has to be thickest were the > LED is > mounted and then taper down in thickness as it expands out in > width and > height? > > Perhaps tapered layers of carbon fiber (with thermal epoxy) > would be > better as the LED mount plate? (After a diamond, graphene or > graphite > apparently has the best thermal conductivity.) > > Any (thermal) engineers here that could enlighten me? > > Finn > > > --- > This email has been checked for viruses by Avast antivirus > software. > https://www.avast.com/antivirus > > ========== > - > Electric-List" rel="noreferrer" > target="_blank">http://www.matronics.com/Navigator?AeroElectric-List > ========== > FORUMS - > eferrer" target="_blank">http://forums.matronics.com > ========== > WIKI - > errer" target="_blank">http://wiki.matronics.com > ========== > b Site - > -Matt Dralle, List Admin. > rel="noreferrer" > target="_blank">http://www.matronics.com/contribution > ========== > > --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ________________________________ Message 8 _____________________________________ Time: 06:30:07 AM PST US From: "Robert L. Nuckolls, III" Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? >Let's say we use a 100W LED (~13,000 lumens). We'd have to remove 70 >to 80W of heat from the LED. AND the associated electronics. See attached: Bob . . . ________________________________ Message 9 _____________________________________ Time: 06:33:06 AM PST US From: "Robert L. Nuckolls, III" Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? At 07:46 AM 11/1/2020, you wrote: > >That would introduce drag. > >Name of the game here is no drag and lowest possible weight for max >light output (lumens) directed in a narrow beam that can reach 3,000 >feet or more down a deer infested runway. Save yourself a LOT of fuss and bother. Consider adapting a COTs LED assembly to your task. There's a boat-load of really capable emitters with built in power supplies and cooling. I got 'em on all my vehicles. Dirt is more expensive . . . https://tinyurl.com/y5kky5f6 Bob . . . ________________________________ Message 10 ____________________________________ Time: 07:01:16 AM PST US Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? From: Finn Lassen Sorry, I was looking at the Cree XM-L2. 116 l/w. 116/350 = 33%. At least 5% loss in the driver. So about 31% or 69% heat dissipation. (Not sure about optical efficiency. I guess that's heat in the lens and reflector.) So thickness of mount plate should not be a factor? A 10W LED I have has a 1.6mm thick metal base, but I guess that's needed to screw it down to heat sink with proper contact (minimal heat sink compound thickness). Alum: 205 W/mK or 0.205W/C. But, but ... is cross area not a factor? Trying to compare it to electrical resistance in a wire. The higher the cross area -- the less resistance. Trying to wrap my head around the transition from flat contact area under the LED chip to the heat sink plate as it expands out in area. Seems mounting it to an infinite piece of alum foil would not be the same as mounting it to 10x10" 0.063" thick alum plate ... Finn On 11/1/2020 9:25 AM, Robert L. Nuckolls, III wrote: >> Let's say we use a 100W LED (~13,000 lumens). We'd have to remove 70 >> to 80W of heat from the LED. > > AND the associated electronics. > > See attached: > > > Bob . . . > --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ________________________________ Message 11 ____________________________________ Time: 08:12:32 AM PST US From: Ernest Christley Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? I got might lights from https://www.mpja.com/20-30-50-Watt-LEDs/products/5 80/ The drivers are riveted to the wing skin and the LED riveted to a plate of .080" aluminum.=C2- They were then riveted to Zenith's mount plate (I hav e a 601XL).=C2- I didn't bother to use any reflectors, as the LEDs tend t o concentrate light in one area.=C2- One 30W driver wiill handle three 10 W LEDs.=C2- That will spread the heat out.=C2- Cooling air is provided by a 1/8" hole drilled to stop a crack in the plexi glass cover=C2- :-)=C2- If anyone can measure the drag of that 1/8" hol e, you're smart enough that I want to be your friend. On Sunday, November 1, 2020, 10:02:03 AM EST, Finn Lassen wrote: Sorry, I was looking at the Cree XM-L2.=C2- 116 l/w. 116/350 = 33%. At least 5% loss in the driver. So about 31% or 69% heat dissipation. (Not sur e about optical efficiency. I guess that's heat in the lens and reflector.) So thickness of mount plate should not be a factor? A 10W LED I have has a 1.6mm thick metal base, but I guess that's needed to screw it down to heat sink with proper contact (minimal heat sink compound thickness). Alum: 205 W/mK or 0.205W/C. But, but ... is cross area not a factor? Trying to compare it to electrical resistance in a wire.=C2- The higher the cro ss area -- the less resistance. Trying to wrap my head around the transition from flat contact area under t he LED chip to the heat sink plate as it expands out in area. Seems mountin g it to an infinite piece of alum foil would not be the same as mounting it to 10x10" 0.063" thick alum plate ... Finn On 11/1/2020 9:25 AM, Robert L. Nuckolls, III wrote: Let's say we use a 100W LED (~13,000 lumens). We'd have to remove 70 to 80W of heat from the LED. =C2- AND the associated electronics. =C2- =C2- See attached: =C2- Bob . . . | | Virus-free. www.avast.com | ________________________________ Message 12 ____________________________________ Time: 09:07:53 AM PST US From: "Robert L. Nuckolls, III" Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? > > >Trying to wrap my head around the transition from flat contact area >under the LED chip to the heat sink plate as it expands out in area. >Seems mounting it to an infinite piece of alum foil would not be the >same as mounting it to 10x10" 0.063" thick alum plate ... But the sheet has thermal resistance too. And rate of dissipation is a function of temperature differential. As you move out onto the sheet, the fastest transfer is right around the base of the LED; temperatures fall as you move out hence energy dissipated per unit area declines. Intuitively, it makes sense that 'sink' temperature at the base of the LED will be much higher than at a remote location. Are we talking about landing lights? Bob . . . ________________________________ Message 13 ____________________________________ Time: 10:06:39 AM PST US From: "Robert L. Nuckolls, III" Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? At 11:03 AM 11/1/2020, you wrote: >>Trying to wrap my head around the transition from flat contact area >>under the LED chip to the heat sink plate as it expands out in >>area. Seems mounting it to an infinite piece of alum foil would not >>be the same as mounting it to 10x10" 0.063" thick alum plate ... > > But the sheet has thermal resistance too. And > rate of dissipation is a function of temperature > differential. As you move out onto the sheet, > the fastest transfer is right around the base > of the LED; temperatures fall as you move > out hence energy dissipated per unit area > declines. > > Intuitively, it makes sense that 'sink' > temperature at the base of the LED will > be much higher than at a remote location. > > Are we talking about landing lights? Duuh . . . need another cup of coffee . . . Is this the kind of critter we're talking about? https://tinyurl.com/y4tespt7 Do you have a link to share for the exact device and its spec sheet? Bob . . . ________________________________ Message 14 ____________________________________ Time: 10:54:02 AM PST US Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? From: Finn Lassen On 11/1/2020 12:03 PM, Robert L. Nuckolls, III wrote: >> >> >> Trying to wrap my head around the transition from flat contact area >> under the LED chip to the heat sink plate as it expands out in area. >> Seems mounting it to an infinite piece of alum foil would not be the >> same as mounting it to 10x10" 0.063" thick alum plate ... > > But the sheet has thermal resistance too. And > rate of dissipation is a function of temperature > differential. As you move out onto the sheet, > the fastest transfer is right around the base > of the LED; temperatures fall as you move > out hence energy dissipated per unit area > declines. > > Intuitively, it makes sense that 'sink' > temperature at the base of the LED will > be much higher than at a remote location. > > Are we talking about landing lights? > > > Bob . . . > Yes landing lights. So my understanding is right. Thicker plate near LED base, then tapering out. You'll also see that in old big heat sinks: thick base then fins tapering out. I guess optimal design is total surface area versus enough material to conduct the heat to the entire surface (when designing for least weight). Some newer heat sinks I've seen have numerous rods sticking out from the base. Perhaps (for least weight) the rods should actually be tapered. This my original question: "Am I right in assuming the plate has to be thickest were the LED is mounted and then taper down in thickness as it expands out in width and height?" Another interesting item from Googling "best thermal conductivity": "Along with its carbon cousins graphite and graphene, diamond is the best thermal conductor around room temperature, having thermal conductivity of more than 2,000 watts per meter per Kelvin, which is five times higher than the best metals such as *copper*." leading to my: "Perhaps tapered layers of carbon fiber (with thermal epoxy) would be better as the LED mount plate? (After a diamond, graphene or graphite apparently has the best thermal conductivity.)" One could shape the carbon fiber layup to closely fit the leading edge skin. I guess I'm thinking way, way, way too far out of the box :) Finn --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ________________________________ Message 15 ____________________________________ Time: 11:12:06 AM PST US Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? From: Finn Lassen Yes, that's where I got the position and strobe LEDs I'm working with at the moment, thanks to your posting here a long time ago. Got 5 of the 10W super white LEDs. One for tail position/strobe. Two for each wingtip strobe.I also got the 10W green and red ones for wing position lights. The tail LED will be driven with 300mA (via series diode) as a position light and then high-current pulses for strobing it. I've read of over-driving LEDs for strobes. Max specs for the 10W LEDs says 1 amp, 1.5A peak. I would like to give them a 100ms (or less) 10A pulse. My big question is if that would violate the 1.5A peak spec. Fiddling with a LM324 quad single-supply op-amp. Looking at application examples I found that it's actually possible to make just one op-amp generate 100ms pulses every 1 sec with 5 resistors, one capacitor and one diode. Also can run directly on ship's power. (max 32V). Now I'm looking for a P-channel MOSFET to drive the LED to max voltage. I guess I could use a PNP power transistor I've got laying around ... Finn On 11/1/2020 11:06 AM, Ernest Christley wrote: > I got might lights from > https://www.mpja.com/20-30-50-Watt-LEDs/products/580/ > > The drivers are riveted to the wing skin and the LED riveted to a > plate of .080" aluminum. They were then riveted to Zenith's mount > plate (I have a 601XL). I didn't bother to use any reflectors, as the > LEDs tend to concentrate light in one area. One 30W driver wiill > handle three 10W LEDs. That will spread the heat out. > > Cooling air is provided by a 1/8" hole drilled to stop a crack in the > plexiglass cover :-) If anyone can measure the drag of that 1/8" > hole, you're smart enough that I want to be your friend. > > On Sunday, November 1, 2020, 10:02:03 AM EST, Finn Lassen > wrote: > > > Sorry, I was looking at the Cree XM-L2. 116 l/w. 116/350 = 33%. At > least 5% loss in the driver. So about 31% or 69% heat dissipation. > (Not sure about optical efficiency. I guess that's heat in the lens > and reflector.) > > So thickness of mount plate should not be a factor? A 10W LED I have > has a 1.6mm thick metal base, but I guess that's needed to screw it > down to heat sink with proper contact (minimal heat sink compound > thickness). > > Alum: 205 W/mK or 0.205W/C. But, but ... is cross area not a factor? > Trying to compare it to electrical resistance in a wire. The higher > the cross area -- the less resistance. > > Trying to wrap my head around the transition from flat contact area > under the LED chip to the heat sink plate as it expands out in area. > Seems mounting it to an infinite piece of alum foil would not be the > same as mounting it to 10x10" 0.063" thick alum plate ... > > Finn > > > On 11/1/2020 9:25 AM, Robert L. Nuckolls, III wrote: >>> Let's say we use a 100W LED (~13,000 lumens). We'd have to remove 70 >>> to 80W of heat from the LED. >> >> AND the associated electronics. >> >> See attached: >> >> >> >> Bob . . . >> > > > > Virus-free. www.avast.com > > > > <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2> --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ________________________________ Message 16 ____________________________________ Time: 12:37:07 PM PST US From: Charlie England Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? On Sun, Nov 1, 2020 at 12:59 PM Finn Lassen wrote: > > > On 11/1/2020 12:03 PM, Robert L. Nuckolls, III wrote: > > > Trying to wrap my head around the transition from flat contact area under > the LED chip to the heat sink plate as it expands out in area. Seems > mounting it to an infinite piece of alum foil would not be the same as > mounting it to 10x10" 0.063" thick alum plate ... > > > But the sheet has thermal resistance too. And > rate of dissipation is a function of temperature > differential. As you move out onto the sheet, > the fastest transfer is right around the base > of the LED; temperatures fall as you move > out hence energy dissipated per unit area > declines. > > Intuitively, it makes sense that 'sink' > temperature at the base of the LED will > be much higher than at a remote location. > > Are we talking about landing lights? > > > Bob . . . > > Yes landing lights. > > So my understanding is right. Thicker plate near LED base, then tapering > out. You'll also see that in old big heat sinks: thick base then fins > tapering out. I guess optimal design is total surface area versus enough > material to conduct the heat to the entire surface (when designing for > least weight). Some newer heat sinks I've seen have numerous rods sticking > out from the base. Perhaps (for least weight) the rods should actually be > tapered. > > This my original question: "Am I right in assuming the plate has to be > thickest were the LED is mounted and then taper down in thickness as it > expands out in width and height?" > > Another interesting item from Googling "best thermal conductivity": > "Along with its carbon cousins graphite and graphene, diamond is the best > thermal conductor around room temperature, having thermal conductivity of > more than 2,000 watts per meter per Kelvin, which is five times higher than > the best metals such as *copper*." > > leading to my: "Perhaps tapered layers of carbon fiber (with thermal > epoxy) would be better as the LED mount plate? (After a diamond, graphene > or graphite apparently has the best thermal conductivity.)" > > One could shape the carbon fiber layup to closely fit the leading edge > skin. > > I guess I'm thinking way, way, way too far out of the box :) > > Finn > Jim Weir has a couple of articles in Kitplanes magazine that give a 'cookbook' method of figuring heatsink size. I don't know, but strongly suspect that the epoxy used as a binder in a typical carbon layup will more than kill any thermal advantage of the carbon. How about some aluminum flashing sheet? Bend up multiple ' [ ' channels with increasing center widths, stack up up 'concentric' drill for the LED base, then reassemble with heat sink compound. Not perfect, but a lot faster/cheaper than carbon, and you get the extra thickness next to the LED. Of course, by the time you do all that, you'll have spent more in parts/labor than the FlyLEDs light we talked about, off-line. :-) I doubt it will be much lighter, either. Charlie ________________________________ Message 17 ____________________________________ Time: 01:24:27 PM PST US From: Ernest Christley Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? Another option is to mount the LED in the center of a disk.=C2- Drill a ring of 1/8" holes right up next to the LED.=C2- Then a ring of 3/16" hol es a little further out.=C2- Then a ring of 1/4", etc.The plate provides a way to mount to the wing. On Sunday, November 1, 2020, 3:38:03 PM EST, Charlie England wrote: On Sun, Nov 1, 2020 at 12:59 PM Finn Lassen wrote: On 11/1/2020 12:03 PM, Robert L. Nuckolls, III wrote: Trying to wrap my head around the transition from flat contact area under the LED chip to the heat sink plate as it expands out in area. Seems mounti ng it to an infinite piece of alum foil would not be the same as mounting i t to 10x10" 0.063" thick alum plate ... =C2- But the sheet has thermal resistance too. And =C2- rate of dissipation is a function of temperature =C2- differential. As you move out onto the sheet, =C2- the fastest transfer is right around the base =C2- of the LED; temperatures fall as you move =C2- out hence energy dissipated per unit area =C2- declines. =C2- Intuitively, it makes sense that 'sink' =C2- temperature at the base of the LED will =C2- be much higher than at a remote location. =C2- Are we talking about landing lights? =C2- Bob . . . Yes landing lights. So my understanding is right. Thicker plate near LED base, then tapering o ut. You'll also see that in old big heat sinks: thick base then fins taperi ng out.=C2- I guess optimal design is=C2- total surface area versus eno ugh material to conduct the heat to the entire surface (when designing for least weight). Some newer heat sinks I've seen have numerous rods sticking out from the base. Perhaps (for least weight) the rods should actually be t apered. This my original question: "Am I right in assuming the plate has to be thi ckest were the LED is mounted and then taper down in thickness as it expand s out in width and height?" Another interesting item from Googling "best thermal conductivity": "Along with its carbon cousins graphite and graphene, diamond is the best thermal conductor around room temperature, having thermal conductivity of m ore than 2,000 watts per meter per Kelvin, which is five times higher than the best metals such as copper." leading to my: "Perhaps tapered layers of carbon fiber (with thermal epoxy ) would be better as the LED mount plate? (After a diamond, graphene or gra phite apparently has the best thermal conductivity.)" One could shape the carbon fiber layup to closely fit the leading edge ski n. I guess I'm thinking way, way, way too far out of the box :) Finn Jim Weir has a couple of articles in Kitplanes magazine that give a 'cookbo ok' method of figuring heatsink size.=C2- I don't know, but strongly susp ect that the epoxy used as a binder in a typical carbon layup will more tha n kill any thermal advantage of the carbon. How about some aluminum flashin g sheet? Bend up multiple ' [ ' channels with increasing center widths, sta ck up up 'concentric' drill for the LED base, then reassemble with heat sin k compound. Not perfect, but a lot faster/cheaper than carbon, and you get the extra thickness next to the LED.=C2- Of course, by the time you do all that, you'll have spent more in parts/lab or than the FlyLEDs light we talked about, off-line. :-) I doubt it will be much lighter, either. Charlie ________________________________ Message 18 ____________________________________ Time: 03:38:32 PM PST US From: "Robert L. Nuckolls, III" Subject: AeroElectric-List: Microair 760 transceivers for sale A subscriber to this List has determined that a couple of Microair 760 transceivers are surplus to his needs. He asked my assistance with getting them into new homes. I've listed them on eBay at auction. All proceeds will go to Matt Dralle to support his closet full of byte-thrashers that make all this good stuff run! The first is a 7 day listing up now at https://tinyurl.com/y2lobwew The second listing is a 10 day auction that goes 'live' at 1900 PST. Will post the link later. Bob . . . ________________________________ Message 19 ____________________________________ Time: 03:40:20 PM PST US From: "Robert L. Nuckolls, III" Subject: Re: AeroElectric-List: Breaker then pump failure >Bob, if I get the old pump back, I'll send it to you along with the breaker. Thanks >BTW, I just became aware of an AD from=C2 2008 to >replace this type of CB switch in Beechcraft products. Bingo . . . I worked the failure analysis team on that fiasco. As I recall, the risk of failure came to light when a pilot reported, "smoke squirting from around the handle on the prop de-ice switch breaker." Here's link to relevant pictures and documents: https://tinyurl.com/y48y93rk That was a real can of worms. There were tens of thousands of switch breakers in the field covering decades of production in Bonanzas and Barons. Being a commercial off the shelf product, it was unclear which fielded aircraft might be vulnerable to this condition. I.e. smoke AFTER failure of the braid. As it turns out, only the prop de-ice switch breaker had a risk for generating some smoke. Switches carrying lower currents would continue to function although loss of the braided jumper on alternator control switches increased field supply path resistance which MIGHT lead to regulator instability. It was finally deduced that this failure did not represent any greater hazard than failure of the protected device . . . hence removal/replacement of a breaker could be left 'on condition' as opposed to a fleet wide swap out. Of course, ongoing production and spares breakers have the magic insulator so the risk of smoke has been mitigated for the future. Bob . . . ________________________________ Message 20 ____________________________________ Time: 03:44:58 PM PST US From: "Robert L. Nuckolls, III" Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? >> Are we talking about landing lights? >> >> >> Bob . . . >Yes landing lights. Okay . . . in the right pew now. Is there an engineering data spec sheet available for the device(s) you're considering? Do you know the thermal resistance from mounting surface to LED junction? Bob . . . ________________________________ Message 21 ____________________________________ Time: 04:00:03 PM PST US Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? From: Finn Lassen Cool. But how thick should the disk be? I don't know why I'm so thermodynamics challenged ... A Google search finds that alum has a thermal resistance of 0.5W/C. But what does that mean? There's got to be a formula that includes material thickness and area. Electrical resistance in a wire includes cross section (area) of wire and length. Surely something similar must be the case for thermal resistance or conductivity. Charlie, can you point me to the Kitplanes article? Finn On 11/1/2020 4:18 PM, Ernest Christley wrote: > Another option is to mount the LED in the center of a disk. Drill a > ring of 1/8" holes right up next to the LED. Then a ring of 3/16" > holes a little further out. Then a ring of 1/4", etc.The plate > provides a way to mount to the wing. > > On Sunday, November 1, 2020, 3:38:03 PM EST, Charlie England > wrote: > > > On Sun, Nov 1, 2020 at 12:59 PM Finn Lassen > wrote: > > > On 11/1/2020 12:03 PM, Robert L. Nuckolls, III wrote: >>> >>> >>> Trying to wrap my head around the transition from flat contact >>> area under the LED chip to the heat sink plate as it expands out >>> in area. Seems mounting it to an infinite piece of alum foil >>> would not be the same as mounting it to 10x10" 0.063" thick alum >>> plate ... >> >> But the sheet has thermal resistance too. And >> rate of dissipation is a function of temperature >> differential. As you move out onto the sheet, >> the fastest transfer is right around the base >> of the LED; temperatures fall as you move >> out hence energy dissipated per unit area >> declines. >> >> Intuitively, it makes sense that 'sink' >> temperature at the base of the LED will >> be much higher than at a remote location. >> >> Are we talking about landing lights? >> >> >> Bob . . . >> > Yes landing lights. > > So my understanding is right. Thicker plate near LED base, then > tapering out. You'll also see that in old big heat sinks: thick > base then fins tapering out. I guess optimal design is total > surface area versus enough material to conduct the heat to the > entire surface (when designing for least weight). Some newer heat > sinks I've seen have numerous rods sticking out from the base. > Perhaps (for least weight) the rods should actually be tapered. > > This my original question: "Am I right in assuming the plate has > to be thickest were the LED is mounted and then taper down in > thickness as it expands out in width and height?" > > Another interesting item from Googling "best thermal conductivity": > "Along with its carbon cousins graphite and graphene, diamond is > the best thermal conductor around room temperature, having thermal > conductivity of more than 2,000 watts per meter per Kelvin, which > is five times higher than the best metals such as *copper*." > > leading to my: "Perhaps tapered layers of carbon fiber (with > thermal epoxy) would be better as the LED mount plate? (After a > diamond, graphene or graphite apparently has the best thermal > conductivity.)" > > One could shape the carbon fiber layup to closely fit the leading > edge skin. > > I guess I'm thinking way, way, way too far out of the box :) > > Finn > > Jim Weir has a couple of articles in Kitplanes magazine that give a > 'cookbook' method of figuring heatsink size. I don't know, but > strongly suspect that the epoxy used as a binder in a typical carbon > layup will more than kill any thermal advantage of the carbon. How > about some aluminum flashing sheet? Bend up multiple ' [ ' channels > with increasing center widths, stack up up 'concentric' drill for the > LED base, then reassemble with heat sink compound. Not perfect, but a > lot faster/cheaper than carbon, and you get the extra thickness next > to the LED. > > Of course, by the time you do all that, you'll have spent more in > parts/labor than the FlyLEDs light we talked about, off-line. :-) I > doubt it will be much lighter, either. > > Charlie > --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ________________________________ Message 22 ____________________________________ Time: 04:41:38 PM PST US Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? From: Finn Lassen On 11/1/2020 6:37 PM, Robert L. Nuckolls, III wrote: >>> Are we talking about landing lights? >>> >>> >>> Bob . . . > > >> Yes landing lights. > > Okay . . . in the right pew now. > > Is there an engineering data spec sheet available > for the device(s) you're considering? > > Do you know the thermal resistance from mounting > surface to LED junction? > > Bob . . . > Looking at the Cree XM-L2 with a thermal resistance of 2.5W/C from junction to mount surface. Add to that a conservative 0.5W/C for heat sink compound and we get 3W/C. Max die temp is 150C. But unlimited life is 35C. It appears that when you go above 35C you start reducing life of the LED. The LEDs are "binned" (color and efficiency determined) at 85C. So lets set die temp of 85C as the design goal. With built-in temp limiting, I think we can probably consider wing skin to air thermal resistance as 0W/C (air flowing over the skin in flight). Consider 32C ambient we get 85 -3 -32 = 50 allowed temp difference from mount plate to skin. Each LED has a mount surface of about 20x20mm. If we have eight 10W LEDs at 30% efficiency, we need to get rid of 56 watts. So max thermal resistance of mount plate and mount late to wing skin should be 56/50 = 1.1W/C. If we use heat sink compound between plate and wing skin, we're down to 0.6W/C max thermal resistance of the mount plate. Wow! Can that be right? If alum has thermal resistance of 0.2 W/C we need 3 of something. What is the something? Thickness vs area? This is where I bog down. Finn --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ________________________________ Message 23 ____________________________________ Time: 04:53:14 PM PST US Subject: AeroElectric-List: Re: ADS-B and Transponder antenna... From: "Eric Page" Here's another useful resource for learning the nanoVNA. It's a demo board that you connect to the device in order to learn what the test response should look like for various kinds of circuits. https://preview.tinyurl.com/yxz2hast I bought one and it seems to be well made. My only complaint is the tiny U.FL connectors they used on the board. They're a bit of a pain to mate, and they're only designed for a few dozen mating cycles, so it's not a lifetime tool. But hey, whaddya want for $13.00?! Read this topic online here: http://forums.matronics.com/viewtopic.php?p=499175#499175 ________________________________ Message 24 ____________________________________ Time: 06:02:48 PM PST US Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? From: Charlie England On 11/1/2020 5:53 PM, Finn Lassen wrote: > > Cool. But how thick should the disk be? > > I don't know why I'm so thermodynamics challenged ... > > A Google search finds that alum has a thermal resistance of 0.5W/C. > But what does that mean? There's got to be a formula that includes > material thickness and area. > > Electrical resistance in a wire includes cross section (area) of wire > and length. Surely something similar must be the case for thermal > resistance or conductivity. > > Charlie, can you point me to the Kitplanes article? > > Finn > > > On 11/1/2020 4:18 PM, Ernest Christley wrote: >> Another option is to mount the LED in the center of a disk. Drill a >> ring of 1/8" holes right up next to the LED. Then a ring of 3/16" >> holes a little further out. Then a ring of 1/4", etc.The plate >> provides a way to mount to the wing. >> >> On Sunday, November 1, 2020, 3:38:03 PM EST, Charlie England >> wrote: >> >> >> >> >> On Sun, Nov 1, 2020 at 12:59 PM Finn Lassen > > wrote: >> >> >> >> On 11/1/2020 12:03 PM, Robert L. Nuckolls, III wrote: >>>> >>>> >>>> Trying to wrap my head around the transition from flat contact >>>> area under the LED chip to the heat sink plate as it expands >>>> out in area. Seems mounting it to an infinite piece of alum >>>> foil would not be the same as mounting it to 10x10" 0.063" >>>> thick alum plate ... >>> >>> But the sheet has thermal resistance too. And >>> rate of dissipation is a function of temperature >>> differential. As you move out onto the sheet, >>> the fastest transfer is right around the base >>> of the LED; temperatures fall as you move >>> out hence energy dissipated per unit area >>> declines. >>> >>> Intuitively, it makes sense that 'sink' >>> temperature at the base of the LED will >>> be much higher than at a remote location. >>> >>> Are we talking about landing lights? >>> >>> >>> Bob . . . >>> >> Yes landing lights. >> >> So my understanding is right. Thicker plate near LED base, then >> tapering out. You'll also see that in old big heat sinks: thick >> base then fins tapering out. I guess optimal design is total >> surface area versus enough material to conduct the heat to the >> entire surface (when designing for least weight). Some newer heat >> sinks I've seen have numerous rods sticking out from the base. >> Perhaps (for least weight) the rods should actually be tapered. >> >> This my original question: "Am I right in assuming the plate has >> to be thickest were the LED is mounted and then taper down in >> thickness as it expands out in width and height?" >> >> Another interesting item from Googling "best thermal conductivity": >> "Along with its carbon cousins graphite and graphene, diamond is >> the best thermal conductor around room temperature, having >> thermal conductivity of more than 2,000 watts per meter per >> Kelvin, which is five times higher than the best metals such as >> *copper*." >> >> leading to my: "Perhaps tapered layers of carbon fiber (with >> thermal epoxy) would be better as the LED mount plate? (After a >> diamond, graphene or graphite apparently has the best thermal >> conductivity.)" >> >> One could shape the carbon fiber layup to closely fit the leading >> edge skin. >> >> I guess I'm thinking way, way, way too far out of the box :) >> >> Finn >> >> Jim Weir has a couple of articles in Kitplanes magazine that give a >> 'cookbook' method of figuring heatsink size. I don't know, but >> strongly suspect that the epoxy used as a binder in a typical carbon >> layup will more than kill any thermal advantage of the carbon. How >> about some aluminum flashing sheet? Bend up multiple ' [ ' channels >> with increasing center widths, stack up up 'concentric' drill for the >> LED base, then reassemble with heat sink compound. Not perfect, but a >> lot faster/cheaper than carbon, and you get the extra thickness next >> to the LED. >> >> Of course, by the time you do all that, you'll have spent more in >> parts/labor than the FlyLEDs light we talked about, off-line. :-) I >> doubt it will be much lighter, either. >> >> Charlie >> > Boy, am I glad that this thread isn't a novel. Finn, Try this link (you probably need a subscription to access it): https://www.kitplanes.com/aero-lectrics-141/ There are a lot of other articles by Weir that deal with it, too. I just logged in and searched for 'heat sink', and watched for Weir's name as author. Charlie -- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ________________________________ Message 25 ____________________________________ Time: 06:16:02 PM PST US From: Ernest Christley Subject: Re: AeroElectric-List: DIY LED landing light -- Wing skin as heat sink? https://electronics.stackexchange.com/questions/294583/does-heatsink-trans fer-plate-thickness-matter I think the 4th answer down give the answer.=C2- Draw out the path of whe re you are inputting the heat (back of the module), to where air starts flo wing over it. If it were me, I'd mount it on an 0.80 or 0.90 plate and tape a thermocoupl e on the back.=C2- Maybe set a fan in front to simulate the prop turning. =C2- If not cool enough, use a thicker plate, or drill more holes. On Sunday, November 1, 2020, 7:42:33 PM EST, Finn Lassen wrote: On 11/1/2020 6:37 PM, Robert L. Nuckolls, III wrote: =C2- Are we talking about landing lights? =C2- Bob . . . Yes landing lights. =C2- Okay . . . in the right pew now. =C2- Is there an engineering data spec sheet available =C2- for the device(s) you're considering? =C2- Do you know the thermal resistance from mounting =C2- surface to LED junction? =C2- Bob . . . Looking at the Cree XM-L2 with a thermal resistance of 2.5W/C from junctio n to mount surface. Add to that a conservative 0.5W/C for heat sink compoun d and we get 3W/C. Max die temp is 150C. But unlimited life is 35C. It appears that when you go above 35C you start reducing life of the LED. The LEDs are "binned" (color and efficiency determined)=C2- at 85C. So lets set die temp of 85C as the design goal. With built-in temp limiting, I think we can probably consider wing skin to air thermal resistance as 0W/C (air flowing over the skin in flight). Consider 32C ambient we get 85 -3 -32 = 50 allowed temp difference from mount plate to skin. Each LED has a mount surface of about 20x20mm. If we have eight 10W LEDs at 30% efficiency, we need to get rid of 56 watt s. So max thermal resistance of mount plate and mount late to wing skin shoul d be 56/50 = 1.1W/C. If we use heat sink compound between plate and wing skin, we're down to 0.6W/C max thermal resistance of the mount plate. Wow! Can that be right? If alum has thermal resistance of 0.2 W/C we need 3 of something. What is the something? Thickness vs area? This is where I bog down. Finn | | Virus-free. www.avast.com | ------------------------------------------------------------------------------------- Other Matronics Email List Services ------------------------------------------------------------------------------------- Post A New Message aeroelectric-list@matronics.com UN/SUBSCRIBE http://www.matronics.com/subscription List FAQ http://www.matronics.com/FAQ/AeroElectric-List.htm Web Forum Interface To Lists http://forums.matronics.com Matronics List Wiki http://wiki.matronics.com Full Archive Search Engine http://www.matronics.com/search 7-Day List Browse http://www.matronics.com/browse/aeroelectric-list Browse Digests http://www.matronics.com/digest/aeroelectric-list Browse Other Lists http://www.matronics.com/browse Live Online Chat! http://www.matronics.com/chat Archive Downloading http://www.matronics.com/archives Photo Share http://www.matronics.com/photoshare Other Email Lists http://www.matronics.com/emaillists Contributions http://www.matronics.com/contribution ------------------------------------------------------------------------------------- These Email List Services are sponsored solely by Matronics and through the generous Contributions of its members.