What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Another data point for those rolling their own LED Nav Lights

Brantel

Brantel

Well Known Member
****EDIT**** If you are new to the idea of rolling your own LED nav lights, several have reported issues using the constant current LED drivers talked about in this thread. It seems these things are terrible RFI noise makers and will render your aircraft radio next to useless. All attempts at sheilding/shunting this noise has failed that I am aware of. Most have decided to just use the tried and true method of a voltage dropping resistor instead. If someone had the time to work this out, I am sure that a solution could be found but most of us have bigger fish to fry in other areas of the aircraft build. Use the constant current LED power supplies at your own risk.


UPDATE: Bob Knuckles makes a filter now that works great!!! http://www.vansairforce.com/community/showthread.php?t=34987&highlight=filter

The mail dropped off my K2 Stars yesterday. Bought them, the drivers and some thermal grease from LEDSUPPLY.

Hooked them up temporarily and did some test.

Man these things are bright, you cannot look at them too long....

I am planning 6 K2's per side. Running at 500mA by using a two parrallel runs of three in series. I am using a 1000mA wired Buck Puck driver. The drivers are smaller than a quarter and make no heat.

This is 250mA less than the recommended max for these K2's so they run slightly cooler and a little less bright than at a full 750mA.

I tested this combo with a variable power supply and from 20v all the way down to 10v on the red and 10.5v on the green, there was no change whatsoever in the output of the emmitters. Once the voltage drops below that, the driver can no longer provide the constant current and the emmitters start to dim. Should be no problem for a typical airplane.

Tonight I will start working on the base plate/heat sink and mount the K2's. The mirrored plexi will be here Wednesday from a vendor on Ebay.

Other than a minor difference here and there, mine will be built similar to the other examples that others have done already.
 
Last edited:
Last night I cutout the aluminum sheet liner that will cover the entire area of the wingtip cuttout. I stole this idea from Chad Jensen. I used the left over scrap .025 that was what remained of the wing walk doubler skin on the wing kit.

Proceeded to bend the liner to match the angle of the wingtip cuttout.

I then cut the aluminum base plate that the Stars will mount to. I decided to use 4-40 screws instead of epoxy to mount the stars because that gives the best heat transfer to the base plate. I made the base plate about 3/8 smaller all the way around than the wintip cutout area.

Then the layout and drilling/tapping for the screws. I mounted up the green Stars and then commenced to soldering the wiring onto them.

Soldering to the stars is not the easiest thing to do. Without the Star mounted to the base plate, the unit gets extremely hot really quick and the fear is that it will damage the unit. The manufacturer recommends that the soldering take place while the unit is mounted to the heat sink. This robs the heat from the solder pad making a great solder joint tough to do. Be carefull here.....

I will post some pics later.

Others have asked how much this stuff cost:

$131.85 for 12 K2 Stars, 2 wired Buck Puck Drivers, thermal grease, shipping
$20.00 for enough mirrored plexi to do two planes from Ebay
$5.00 for capacitors for the input of the driver
$0.00 for leftover and scrap sheet alum.
$10.00 for heatsinks to mount behind the base plate
$166.85 total

I expect it to take at least 6hrs to fab up the lights. This does not count preparing the wingtip, or doing the wingtip lens which you have to do no mater whether you roll your own or buy someones completed units. Jeff offers excellent value in his sets for $300.00 when you factor the time it takes to do them into the equation.
 
Preheat

If you have a heat gun you can prehead most devices to 120-150 degrees. This will make the 'thermal shock' of soldering easier on the part and easier for you to flow. This is standard for surfacemount parts and tends to relieve some of the stress caused by the thermal expansion difference between the part and the mounting surface.
It's also a great way to save the copper traces on a PCB when doing rework ;)
In most cases I can run a cooler soldering iron and still get a good solder flow!
 
Thanks for the tip. I come from the old school prior to the onslaught of surface mount technology.

I just do not want to burn up the emitter before it ever gets a chance to run.
 
Here is a better pic of how I mounted the K2 Stars and a shot showing them lit up. This is less the mirror at this point.

2003281013650872940_rs.jpg


2003217562044508984_rs.jpg


Here is the alum cover plate that I stole from Chad's design. I like this in conjunction with making the LED package some smaller than the cuttout area. I like the look of that better than the mirror comming all the way out to the lens. This plate gets polished and gives it a finished look. I will also be putting mirror on the strobe side as well.

2003297112821762849_rs.jpg


2006065753771086952_rs.jpg
 
Keep going Brian! I'm also planning making navigation lights as well strobos from LEDs. It's hard to say from those pics (or pretty much from any picture) how bright those lights really are.

I'm planning to put my LEDs to MCPCB but wish to know how you success with yours -- temperatures. When you keep those on, how much over the ambient temperature that aluminum will be finally? Just that is it enough if they are put on ~3mm thick MCPCB or would it still need additional sinks (which I guess it does). Plan was to have (per tip) 6 LEDs with color and 12 white for stobo. Hopefully that isn't overkill. The tail light is much more complicated -- there you don't have much space and you should also keep those LEDs on all the time and then even push them harder to make the flash. As my workshop has no insulation it gets cold during winter therefore I might have some time to work with wingtips during next months.

I haven't decided which LEDs to go currently Rebels seems to beat K2s and other I've found so targeting to them now.

My worst fear is that I end up paying more than System 6 costs especially when dollar per euro rate is getting better all the time. And that with similar cost my performance is worse or equal... we'll see. Just asked some price info for MCPCB -- that stuff seems to be expensive. Planting ready-made MCPCBs to the aluminum as you've done would be much cheaper but I wish to maximize heat transfer from junction to heatsink.

Keep us updated when you progress!
 
Pirkka,

I know that it will depend on how many and how hard you drive your LED's but my thought is that it will definately require a heat sink on the back of the mounting plate. I don't think you will be able to get a large enough MCPCB in there.

With just the base plate alone and not mounted to the thinner and larger sub plate, the temperature rise is significant with six K2's running @ 500mA each.

I am going to either make my own heat sink or use a socket A CPU heat sink. I am trying to find some out of old junk computers but I am having problems finding two of the same. This things are long outdated now a days.

I have noticed that the integral K2 base transfers heat really well to my thicker homemade base. As long as I keep the heat transfer good between my base and the heat sink, I should be golden!
 
Brantel said:
I don't think you will be able to get a large enough MCPCB in there.
Click to expand...

That is my assumption as well... it will work for a while but not transfer enough heat to the air.

I am going to either make my own heat sink or use a socket A CPU heat sink.
Click to expand...
That was also my idea as finding suitable heat sinks for a reasonable price may be hard task. CPU heat sinks, compared to bigger heat sink has often a lot more surface area. Also they might be nearly freely available from some re-cycling place of similar.


Do you have IR thermometer or similar you could measure actual temperatures of your Al-plates? Touching is not very good especially because you feel different materials differently even they would have same temperature. Metals are "hot" way before plastic for example.

Edit: For the strobos the heat sink requirement is not such an issue as they have light on only a fraction of a time. Although they are mounted to same heat sinks so they share the same heat eventually.
 
One way of getting rid of the surplus heat would be to cut out the end of the wing tips where you are mounting the AL and LED's. Just leave a perimeter with nut plates on it to hold the assembly down. This would give air circulation around both sides of the plate. Also if that is not enough surface area to get rid of the heat rivet an other piece of AL at 90degrees to it and about the same surface area. A few planes ago I used LED's. I used super bright 5mm LED. I used 9 on each surface and they worked great, lots of light. ATC in the tower 5 miles out on final said they were quite bright. The pattern of the LED's was laid out in a square of 3X3 with about 1/8 inch spacing.
 
Brian,

It looks like you ordered the Green K2 Stars rather than Cyan. I don't know if you realize it or not, but the SAE color requirement for green position lights is met by the Lumileds Cyan LEDs, but is not met by the Green LEDs.

SAE green is really a bluish-green rather than a true green.

I did a power/heat estimate based on an what I guess the area of your aluminum plate has, and figure that if you allow the lights to run for about an hour, the aluminum will get up around 65C in a 25C ambient (40C rise). Is that about right? Have you measured the plate temperature with a thermometer?

If you can attached that aluminum plate to the aluminum structure of the wing, that would help dissipate the heat by spreading it out to a much larger area.

Dean Wilkinson
AeroLEDs LLC
www.aeroleds.com
 
JD: How much light does one strobe will create light? In theory: I guess we might have about 1 square inch for a tail light, where we have space for 30 Luxeon Rebels (LXML-PWN1-0080, available in shops and 4100K rather than some cool white) and they will generate +160 lm per die within specification (driven with 1A). That would be 4800 lm total - which is actually less as it will heat up and luminosity will drop. That should beat 100 W bulb at least even the number of LEDs would be reduced.

There are also something like this coming up some day http://www.led-professional.com/content/view/464/56/ so if you have time to wait these can be even easier to use.

And in practice our (in Finland) specification for lights is like from 70's and I believe that nobody will ever really measure my lights so I just go as I feel good. ;) And I think those can be pushed even harder when the pulses are short. Total energy driven to LED is easily 1/10 an less compared to the situation where LED is driven all the time. Which in case of tail light is true -- so I guess I end up having pretty low current for navi white and then push strobe per spec. Don't know how easy it is to have there heat sink -- at least the place is worst possible (in weight wise) but over the heat sink cooler would improve situation significantly.

I already found some (high current) LEDs which had also considered this short pulse and allowed to drive them about twice (if I recall correctly) current compared to constant situation. Needless to say that in constant situation they didn't were at the level of Rebels.

And what comes to comparing LEDs for example in case of tail light where space is limited (compared to wing tips) I would look for size, luminosity with angle, color temperature (cool white is better in term of numbers so don't be fooled). Rebels seem to beat other brands as well in junction temperature.

Rebel datasheet http://www.lumileds.com/pdfs/DS56.pdf

PS. Hopefully I didn't mix units as kindly tried to use something you would understand as well as 1 square inch. ;)


Norman: Nice to hear such info as these new high power LEDs available will be light years lighter than any of those 5 mm bright LEDs from couple years back.


Dean: That was interesting info about those colors -- need to consider and read our specification as well once more.

What comes to leading heat to the wing I feel it nearly hopeless in wing tips. The distance is just too big from tip to rib. However this might somehow work in tail light case so need to consider. Although rudder gets pretty narrow at that part so considering that may be waste of time as well. :rolleyes:
 
Dean_aeroleds said:
Brian,

It looks like you ordered the Green K2 Stars rather than Cyan. I don't know if you realize it or not, but the SAE color requirement for green position lights is met by the Lumileds Cyan LEDs, but is not met by the Green LEDs.

SAE green is really a bluish-green rather than a true green.

I did a power/heat estimate based on an what I guess the area of your aluminum plate has, and figure that if you allow the lights to run for about an hour, the aluminum will get up around 65C in a 25C ambient (40C rise). Is that about right? Have you measured the plate temperature with a thermometer?

If you can attached that aluminum plate to the aluminum structure of the wing, that would help dissipate the heat by spreading it out to a much larger area.

Dean Wilkinson
AeroLEDs LLC
www.aeroleds.com
Click to expand...

Dean,

I really did not know that they made cyan K2 Stars but I now see them on the data sheet. The only problem is that I cannot find anywhere to actually purchase them. I was just following the long line of other people that have made their own lights with the green ones. Most if not all the examples I see have the green ones. Even the Whelen example on their website looks like the green ones.

I have been trying to decipher that reg on the color and it looks like there is a range to play in. I don't know for sure but it looks like the green ones possibly would squeak by on the green side of the range. I figure that if it is that close, you would have to be a color expert to know the difference.

I have a socket A cpu heat sink mounted to the back of my base plate and it has about 30 square inches of surface area. The base plate has about 30 (front and back) not counting the overall trim plate that all of this is mounted on.

I ran them this morning in the garage and the ambient temp in the room was 15°C and after 45 minutes with a thermometer in direct contact with the base plate directly behind the group of two stars and right above the heat sink, the temp rose to 30°C. I plan to do some test with a IR thermometer as soon as I can get the one from work on loan.

The final assembly with the artic silver paste and the cpu heat sink drastically reduced the running temps of the base plate.
 
Last evening I cuttout the mirrored plexi, drilled and countersunk it for the emitters and screws. Affixed the stars to the base plate using a thin coat of artic silver and red locktite on the threads of the screws. Cuttout the hole for the heat sink in trim plate that all of this is mounted to. I still have to drill a hole for the wires to come thru the base plate. Total time invested to date is about 4 hours.

2005887041263919408_rs.jpg


I also riveted this socket A cpu heat sink to the back of the base plate using artic silver in the joint. The nuts here are temporary as there will be nutplates in the tip.

2005885501486277013_rs.jpg


Here it all is mounted less the strobe and strobe side mirror. This will get mounted to the wingtip after I cuttout the hole for the heat sink area. I will most likely just paint the trim plate the same as the wing someday.

2005837249860019013_rs.jpg
 
Last edited:
Regulation references....

Dean_aeroleds said:
Brian,

It looks like you ordered the Green K2 Stars rather than Cyan. I don't know if you realize it or not, but the SAE color requirement for green position lights is met by the Lumileds Cyan LEDs, but is not met by the Green LEDs.

SAE green is really a bluish-green rather than a true green.
......
Dean Wilkinson
AeroLEDs LLC
www.aeroleds.com
Click to expand...

Dean... why quote the SAE specifications?

The FAA uses a different color reference, defined in the FARs....

Sec. 23.1397

Color specifications.

Each position light color must have the applicable International Commission on Illumination chromaticity coordinates as follows:
(a) Aviation red--
"y" is not greater than 0.335; and
"z" is not greater than 0.002.
(b) Aviation green--
"x" is not greater than 0.440-0.320 y;
"x" is not greater than y -0.170; and
"y" is not less than 0.390-0.170 x.
[(c) Aviation white--
"x" is not less than 0.300 and not greater than 0.540;
"y" is not less than "x-0.040" or "y0-0.010", whichever is the smaller; and
"y" is not greater than "x+0.020" nor "0.636-0.400 x";
Where "y0" is the "y" coordinate of the Planckian radiator for the value of "x" considered.]

Is the FAA green the same as SAE green?

gil A
 
Great Job Brian

You are really doing a great job with this. That heat sink looks to be massive enough and similar to what I see on commercial products for lighting.
You have four hours into this? including research, procurement, writing these posts! Ha - just giving you a bad time. Everything always takes longer than it seems. Again, nice job.
 
JonJay said:
You are really doing a great job with this. That heat sink looks to be massive enough and similar to what I see on commercial products for lighting.
You have four hours into this? including research, procurement, writing these posts! Ha - just giving you a bad time. Everything always takes longer than it seems. Again, nice job.
Click to expand...

Hahaha,

Total time invested is at least 2 months, just kidding. I am only counting the actual time working on these. Like the plane itself all the other thousands of hours of research and talking about it does not count....
 
az_gila said:
Dean... why quote the SAE specifications?

The FAA uses a different color reference, defined in the FARs....
gil A
Click to expand...

Same question here...

I do see that the FAA (same as the mil spec listed below for green) is significantly different from the SAE spec.

If this is true, I suspect that the 530nm wavelength of the K2 stars is within the range for the FAA but out of the range for the SAE.

What is the SAE spec and why would we care about it?

FAA
2000368155682235176_rs.jpg


SAE
2000363877116810819_rs.jpg
 
az_gila said:
Dean... why quote the SAE specifications?

The FAA uses a different color reference, defined in the FARs....

Sec. 23.1397

Color specifications.

Each position light color must have the applicable International Commission on Illumination chromaticity coordinates as follows:
(a) Aviation red--
"y" is not greater than 0.335; and
"z" is not greater than 0.002.
(b) Aviation green--
"x" is not greater than 0.440-0.320 y;
"x" is not greater than y -0.170; and
"y" is not less than 0.390-0.170 x.
[(c) Aviation white--
"x" is not less than 0.300 and not greater than 0.540;
"y" is not less than "x-0.040" or "y0-0.010", whichever is the smaller; and
"y" is not greater than "x+0.020" nor "0.636-0.400 x";
Where "y0" is the "y" coordinate of the Planckian radiator for the value of "x" considered.]

Is the FAA green the same as SAE green?

gil A
Click to expand...

Gil,

The SAE color requirements are a little tighter than the FAR calls for, but it is still centered on the same bluish-green color. The aviation green envelope called out in the FARs is larger than the SAE green envelope on all sides, but is still bluish-green.

The green LEDs will fall right on the upper edge of the aviation green envelope, and it appears that they will often be outside of it (there is a chromaticity distribution for the LEDs).

The Cyan LEDs will fall right near the center of the envelope.

I guess the bottom line is whether or not a builder cares if he complies with FAR 23.1397. If he does, he will use Cyan and Red LEDs. If he doesn't, he will use Green and Red.

The reason for this requirement is color discrimination. It is easier for people with some degree of red/green color blindness to discriminate between Cyan and Red than between Green and Red (using the LED color names from Lumileds).

AeroLEDs will be coming out with an LED position light product, and it will use Cyan LEDs to ensure FAR compliance.

Dean Wilkinson
AeroLEDs LLC
www.aeroleds.com
 
Dean_aeroleds said:
Gil,
I guess the bottom line is whether or not a builder cares if he complies with FAR 23.1397. If he does, he will use Cyan and Red LEDs. If he doesn't, he will use Green and Red.

Dean Wilkinson
AeroLEDs LLC
www.aeroleds.com
Click to expand...

Dean, Thanks for the education on this factor. You have helped us understand this better. If I had know what you have shown us, and if there is a 750mA Cyan K2 star readily available, I would have purchased those instead because after all, if you can be centered on the tolerance range then why not?

It does seem though that the greens will meet the FAR but not the SAE as long as you know you are pushing the upper limit of the tolerance range. So that being said, it looks like all of us that have used green will be ok...

If you want to be absolutely 100% no questions asked, positively sure you are meeting the FAR's, go with the Cyan.
 
Cyan

I found that the cyan Luxeon III's are much closer to aviation green when I built my lights (after Jeff suggested it to me).

Looking at my hangar mates RV-6 side by side with my -9A, cyan is a very close match to the certified green nav lights.
 
petehowell said:
I found that the cyan Luxeon III's are much closer to aviation green when I built my lights (after Jeff suggested it to me).

Looking at my hangar mates RV-6 side by side with my -9A, cyan is a very close match to the certified green nav lights.
Click to expand...

It seems that at some point Jeff changed to cyan. By looking at the pics of examples of his lights some seem to be green and some cyan.

Jeff if you are watching can you shed some light on this?

I would say at this point, if you are buying, buy cyan if you can. Still not sure where to find a K2 Star in cyan that can run at 750mA.
 
Dean_aeroleds said:
Gil,

The green LEDs will fall right on the upper edge of the aviation green envelope, and it appears that they will often be outside of it (there is a chromaticity distribution for the LEDs).

The Cyan LEDs will fall right near the center of the envelope.
Click to expand...

Great discussion all.

Certainly some of the Green LEDs will fall outside of the FAA envelope, but it also appears some of the Cyan ones may as well, based on my understanding of the CIE diagrams.

The lower edge of the envelope is defined by a line from a wavelength of 494 nm on the edge of the curve to the white point (x=y=1/3). The upper edge of the envelope is similarly defined by a line from 534 nm to the white point. Because of this, as long as the dominant wavelength is greater than 494 nm and less than 534 nm you should comply with the spec.

If you look at the Luxeon binning and labeling guide (AB21) there are tables showing the bin codes for the dominant wavelength of each LED, which I believe should be printed on the LED.

For the Green there are six different color bin codes, with only codes 1 & 2 that are guaranteed to stay within the spec (< 530 nm). For the Cyan there are also six possible, with all but bin code 1 (< 495 nm) that meet the spec.

I don't know if it's possible to order specific color bin codes, or it's the luck of the draw. Either way, it looks like you're more likely to comply using the cyan than the green.
 
Illumination pattern?

Its not clear to me how the 6 LEDs cover the illumination "volume" included in the FARs 23.1385 tthrough 23.1397. The wide version of the Star 2s illuminate a solid angle of 50 degrees according to the spec sheet I read and the FARs talk about 110 degree horizontal and something like 150 vertical (memory only, don't quote me please) Am I missing some "aiming" of the LEDs that is not apparent in the photos? Or does the lexan do some diffusion and redirection of the light pattern?? There is certainly enough power being emitted to do the job if they shine the right direction.
 
Last edited:
rzbill said:
Its not clear to me how the 6 LEDs cover the illumination "volume" included in the FARs 23.1385 tthrough 23.1397. The wide version of the Star 2s illuminate a solid angle of 50 degrees according to the spec sheet I read and the FARs talk about 110 degree horizontal and something like 150 vertical (memory only, don't quote me please) Am I missing some "aiming" of the LEDs that is not apparent in the photos? Or does the lexan do some diffusion and redirection of the light pattern?? There is certainly enough power being emitted to do the job if they shine the right direction.
Click to expand...

The mirrors and the edge of the acrylic definately play a part in this but the K2's also have a pretty wide angle range. The FAR also allows for a reduction in output the further away you get from the primary axis. The edge of the acrylic is very bright as well as it has a tendacy to pickup and transfer the light output. Take a look a Jeff's pic, http://www.jeffsrv-7a.com/JimsRV-9ALEDNavLights.htm notice the very bright color on the ground and the top of the hangar directly above and below @ 90° to the emitters.

2004459595564623780_rs.jpg


2004473224995266564_rs.jpg
 
Last edited:
rzbill said:
Its not clear to me how the 6 LEDs cover the illumination "volume" included in the FARs 23.1385 tthrough 23.1397. The wide version of the Star 2s illuminate a solid angle of 50 degrees according to the spec sheet I read and the FARs talk about 110 degree horizontal and something like 150 vertical (memory only, don't quote me please) Am I missing some "aiming" of the LEDs that is not apparent in the photos? Or does the lexan do some diffusion and redirection of the light pattern?? There is certainly enough power being emitted to do the job if they shine the right direction.
Click to expand...

I designed a spreadsheet model that allows you to determine the intensity of multiple LEDs. It allows you to vary the angle of each. It also converts the luminous flux (lumens) quoted in the product guides to intensity (candelas). An example of the output is below. Be glad to share my spreadsheet if you're interested - send me a PM. I'm currently revising it for the K2's.

luxeonnavintensityfc7.gif
 
Last edited:
Here is another graphic that is about the best thing I have seen that shows the required output at a particular angle per the FAR's. As you can see, off the most important axis, the requirement is quite low.

2001735284027410392_rs.jpg


Here is a visual on the vertical planes at each horizontal requirement. As you can see the requirements go way down as you move off the horizontal plane.

2000006702346429387_rs.jpg
 
Last edited:
rzbill said:
Its not clear to me how the 6 LEDs cover the illumination "volume" included in the FARs 23.1385 tthrough 23.1397. The wide version of the Star 2s illuminate a solid angle of 50 degrees according to the spec sheet I read and the FARs talk about 110 degree horizontal and something like 150 vertical (memory only, don't quote me please) Am I missing some "aiming" of the LEDs that is not apparent in the photos? Or does the lexan do some diffusion and redirection of the light pattern?? There is certainly enough power being emitted to do the job if they shine the right direction.
Click to expand...

There are two ways to really know; Very sophisticated Photometric Calculation Software that simulates performance based on a boat load of criteria, or, submission of an actual product to a Photometric Testing Laboratory, which is the only true way.
Both of these would require an investment of thousands of dollars.

If you are "rolling your own" you can only assume you are meeting the FAR's. You can not know for sure. In my industry, the only acceptable method is an actual test performed by an independant test lab.

Not trying to dampen any spirits as this is "experimental" aviation after all. That's part of the fun.
 
The physics of light

This is one of the oldest observational sciences, since man looked up to the night sky and say stars or shadows the sun cast. It is also one of the most convoluted and esoteric. Just knowing a little bit about light people discover new planets. Lets face it time, space and "light speed" are all wrapped up in the theory of "relativity". Light is so taken for granted and really not well understood. Now for NAV lights.

Brian I don't know if your set up is "perfect" or meets the letter of the law, but keep up the good work. The coverage is what I'm a little dubious about, but you could be fine?

LED's get their luminosity rating by focusing in a small pattern. LED's are not great for power and omni directional abilities.

Since I think from what I read you are using reflection, you may be fine. That is what Whelen does I am sure. There is no doubt their new 12v LED NAV lights are in a reflective face for a reason. If you don't use reflection you get dozens and dozens of leds aiming in all directions. The reflection method is more elegant but does need a good reflective surface.

If you are relying on polished aluminum oxidization will be a bane. May be consider some corrosion resistant plating or finish? At least some kind of clear coat to keep the aluminum for getting dull?

I have no doubt people will see you at night. I suspect you will be so bright in some directions people will think your a UFO. :D
 
Last edited:
alpinelakespilot2000 said:
Can someone point me toward this mirrored plexiglass that is available on ebay? I'm not finding it. Thanks.
Click to expand...


Search for acrylic mirror sheet.

FYI, I did not know how much to purchase and so I bought 4 12x12 sheets and this is way too much!
 
Last edited:
I did my homework during my R&D of my version of LED Navs. Its not as complicated as everyone tries to make it.
As a matter of fact its quiet simple. So put away all your manuals and documents and rule books.
Shown here is a perfect example of how Luxeon Lambertians supply ample coverage.
Note: 7 LEDs were used which IMO driven at maximum levels are serious overkill so therefore Im sure they are underdriven.

Build, design, experiment, have fun and be safe!!!!

-Jeff

WhelenLEDc1.jpg
 
wing tip lights

Hi;
Great work on the lights!!!
You mentioned that someone manufactures the lenses. I have the old style wing tip on My 6A and have to do the CUT I would rather Buy the Lenses and not screw them up. and then cut the holes.
Any information would be a great help.

Happy holidays

Bob Wieczorek
Closing wings up.:confused:
 
Here is a visual on the vertical planes at each horizontal requirement. As you can see the requirements go way down as you move off the horizontal plane.

2000006702346429387_rs.jpg
 
Last edited:
where at?

Brantel said:
$131.85 for 12 K2 Stars, 2 wired Buck Puck Drivers, thermal grease, shipping
$20.00 for enough mirrored plexi to do two planes from Ebay
$5.00 for capacitors for the input of the driver
$0.00 for leftover and scrap sheet alum.
$10.00 for heatsinks to mount behind the base plate
$166.85 total

I expect it to take at least 6hrs to fab up the lights. This does not count preparing the wingtip, or doing the wingtip lens which you have to do no mater whether you roll your own or buy someones completed units. Jeff offers excellent value in his sets for $300.00 when you factor the time it takes to do them into the equation.
Click to expand...

Got 2 questions.
1 - where did you order your lights and drivers from?
2 - Do you have a link to where Jeff sells his wingtip sets? any pics?

Thanks!
 
Last edited:
Overlaps...

Brantel said:
Here is another graphic that is about the best thing I have seen that shows the required output at a particular angle per the FAR's. As you can see, off the most important axis, the requirement is quite low.

2001735284027410392_rs.jpg

.....
Click to expand...

Brian.... it looks that the overlap requirements may be harder to meet, since they are most likely dependent on the actual housing and size/shape of any adjacent reflective surfaces... They are also maximum numbers...

Sec. 23.1395

Maximum intensities in overlapping beams of [position lights.]

No position light intensity may exceed the applicable values in the following table, except as provided in Sec. 23.1389(b)(3):

Overlaps Maximum intensity --- Area A (candles) ----Area B(candles)
Green in dihedral angle L------------- 10 -------------- 1
Red in dihedral angle R---------------10 -------------- 1
Green in dihedral angle A-------------- 5 --------------1
Red in dihedral angle A---------------- 5 --------------1
Rear white in dihedral angle L---------- 5 --------------1
Rear white in dihedral angle R---------- 5 --------------1

where--
(a) Area A includes all directions in the adjacent dihedral angle that pass through the light source and intersect the common boundary plane at more than 10° but less than 20°; and
(b) Area B includes all directions in the adjacent dihedral angle that pass through the light source and intersect the common boundary plane at more than 20°.

gil A
 
Last edited:
Testing and FARs

RV_7A said:
Yep Cyan is the correct color range IMO.
But.... you are rolling your own. ;)
-Jeff
Click to expand...

Yes... we are rolling our own....

However, our typical operating restrictions require "approved" navigation lights for night flight (check them out...)

Approved in this case means meeting the FAR intensity/coverage/color requirements....

I would disagree with the other posting claiming an independent lab is required.

If you rent a calibrated light measuring meter, then a full test to the FAA requirements can probably be performed in a dark hangar or on a dark ramp with just a ruler and some careful measurements. Document the test results and then you can show you meet the FARs - if your design is good...:)

Any ambient can simply be subtracted by turning the nav. lights off and subtracting the background reading.

Color should be able to be defined from the manufacturer's specifications, as mentioned earlier.

Has anyone run these actual tests, or is it all being done on guesswork? the FARs are quite specific in this area, and we don't get a pass for being Experimental....:)

gil A .....Used to make those calibrated light measuring instruments a long time ago...:)
 
I plan mirrored plexi on the strobe side of the cuttout as well effectively reflecting any light in that hits that plane back. The tip is cuttout parallel to the longitudinal axis which helps to shadow and prevent the overlap. This problem would exist for any wingtip mounted lights LED or conventional.

This is also why too bright is bad. That is one reason why I am underdriving mine a little.

Notice there is no requirement for overlap compliance in the 0-10° range and this is where the maximum output is required.
 
Last edited:
az_gila said:
I would disagree with the other posting claiming an independent lab is required. .....Used to make those calibrated light measuring instruments a long time ago...:)
Click to expand...

Ok - make sure your hand held meter is calibrated, it is recommended to have done at least once a year. Make sure that it has appropriate technology, as most are very spectrum specific and red and green may not even show up, or may show up very inaccurately, make sure you dont contribute to the light level with your reflection (no white shirts), make sure you dont have any shadows, all walls must be black, no reflection, make sure you document each data point specifically, then retest.....

Obvioulsy, none of this is necessary....however, the only way you will be able to truly know if it meets the FAR's, and document to anybody who actually cares, is to put it in a lab. I think you could aproximate it and feel comfortable with a hand held meter, but you still would not truly know....

All of this from a guy who has exactly "0" night flying hours in the 80 hours on my RV6. So, obviously, I dont really care....

Still luv ya Gil, not trying to be argumentive.
 
Not quite...

JonJay said:
Ok - make sure your hand held meter is calibrated, it is recommended to have done at least once a year. Make sure that it has appropriate technology, as most are very spectrum specific and red and green may not even show up, or may show up very inaccurately, make sure you dont contribute to the light level with your reflection (no white shirts), make sure you dont have any shadows, all walls must be black, no reflection, make sure you document each data point specifically, then retest.....

Obvioulsy, none of this is necessary....however, the only way you will be able to truly know if it meets the FAR's, and document to anybody who actually cares, is to put it in a lab. I think you could aproximate it and feel comfortable with a hand held meter, but you still would not truly know....

All of this from a guy who has exactly "0" night flying hours in the 80 hours on my RV6. So, obviously, I dont really care....

Still luv ya Gil, not trying to be argumentive.
Click to expand...

OK.. still not wanting to be argumentative.... but...

I said rent a calibrated light meter....

If the color is OK by the LED manufactures specs., then the spectrum of the light meter should be the "standard" one that mimics the eye's response.. You are talking about a simple instrument like this one....

http://www.udtinstruments.com/products/S370.shtml

..with this sensor...

http://www.udtinstruments.com/products/photometric/model211.shtml

The technology hasn't changed (neither has the eye or candle's definition) much since I was Chief Engineer there 30 years ago...:D ...just the electronics bit has got smaller and better...

Shadows, shirts, ambient does not matter if you get each reading with the LED on and off, and work on a dark ramp. The LED is the extra luminance when it's switched on. If it's nice and dark, and you stand behind the detector, the brightness of the lights should easily overwhelm the ambient....:) An accurate ruler and knowledge of angles to the flight path direction are all that is needed for the test.

Actually, it sounds like the designs should have lots of margin, so the thing I think I would be most concerned with is the coverage and overlap specifications... This is highly dependent on the mounting situation, built-in reflectors and wing tip cut outs, so just one careful test could define the system as meeting the FAR requirements on RV tip installations for all users.

Of course, if you don't fly at night (like you said...:)...) then you are correct, the lights are for show only, and don't have to meet any requirements....:D

gil A
 
wing tip light kit.

Brantel said:
Your in luck! Jeff is running a special!

http://www.vansairforce.com/community/showthread.php?t=23326&highlight=led+special

additional pics and pricing info:

http://www.jeffsrv-7a.com/LEDPROJECT1.htm

Complete with pics on his site!

I purchased mine from ledsupply.com
Click to expand...

Thanx Brian,

For some reason I thought you meant that he was selling a different kit that did the whole thing (including the fiberglass part that you get from vans). So that's where I was confused. I still need to purchase the wing tip light kit here http://www.vansaircraft.com/cgi-bin...91269-508-342&browse=lighting&product=airtech (notice this is the older style cause this is for a RV-6 with the older style wing tips)

Another thing I'm wondering about.. putting the mirror in the wing tip, it looks like the mirror part is actually above where the leds are so it looks like the only thing that would get reflected is when you have the strobes on.

Christopher.
 
loopfuzz said:
Another thing I'm wondering about.. putting the mirror in the wing tip, it looks like the mirror part is actually above where the leds are so it looks like the only thing that would get reflected is when you have the strobes on.
Christopher.
Click to expand...

The lens of the led actually extends thru the mirrored plexi and also he edges of the plexi transmit light very well. Kind of like a large flat piece of fiber optic cable.

You are correct that the mirror on the led side is also there to help reflect the strobe light. There will also be a mirror on the strobe side to help reflect the strobe and led light.
 
Here is an updated picture of the cyan vs the green LED's. Finally got around to changing them out and connecting them up.

They definately have more of a blue green tint. Hard to see unless you compare them side by side.

2001541259934742169_rs.jpg



2001520675129528832_rs.jpg


Original Green:

2004252564676309010_rs.jpg



New Cyan:

2003680383039971649_rs.jpg


2003608776539621193_rs.jpg


These things will blind you....

2003613791653080454_rs.jpg
 
Nice to see some progress here.


I have been designing driver stage as not planning to use any commercial versions. And I had also strobo capability in my mind so I'll have to use much greater number of LEDs than these commercial solutions are currently able to offer. I wish also control brightness if needed to adjust green (I don't like Cyan :p) and red to look same.

My plans are currently based for LM5022 and driving 24 Luxeon Rebel LEDs in series (in strobo case). I originally had a great problems with the application sheet ('demo board') they had for the IC because there were great number of errors -- I was able to trace them and now there is new version online which significantly helps to get results you desire. Now I'm about to finish my specification, I should do final calculations for the circuitry and choose proper components. Rest is pure work.

During last weeks I wrote the SW part of the "LCU" (light control unit) which shall take care of driving my LM5022s as I'm using PWM dimming in case needed like for navigation lights. It also takes care of timing of wig-wag etc... LCU will be based on ATMega32 micro controller.

Yet, no money involved -- that's why making specification or SW is such a fun. :D
 
Warning about the constant current LED drivers

If you are new to the idea of rolling your own LED nav lights, several have reported issues using the constant current LED drivers talked about in this thread. It seems these things are terrible RFI noise makers and will render your aircraft radio next to useless. All attempts at sheilding/shunting this noise has failed that I am aware of. Most have decided to just use the tried and true method of a voltage dropping resistor instead. If someone had the time to work this out, I am sure that a solution could be found but most of us have bigger fish to fry in other areas of the aircraft build. Use the constant current LED power supplies at your own risk.
 
Brantel said:
several have reported issues using the constant current LED drivers talked about in this thread. It seems these things are terrible RFI noise makers and will render your aircraft radio next to useless. All attempts at sheilding/shunting this noise has failed that I am aware of.
Click to expand...

Does anyone have really measured these constant current drivers? Which frequencies they are working at and so? I've just plans to build my own so adjusting frequency might help a bit here. I can't consider using commercial drivers as you can read my specification from above postings.

I've some progress but also some drawbacks as well: the LEDs I look for my system should be available in next year, not before so I'll hold with my design for a while. I've got demo board for micro controller where I can test my code for controlling lights of the plane - haven't tested yet whether it does as it does in simulator. I've most parts to build driver for LEDs but it's hard to test without LEDs (or proper load) so it's hold for now as well... I'm not in a hurry so that's why I'm not pushing too hard this forward and you never know when new and better LEDs come out. ;)
 
Very good information. The Devil is in the details!

Pirkka said:
Philips has relased some tools to design things around their LEDs:

http://www.philipslumileds.com/newsandevents/releases/PR97.pdf

The tools are available free to all designers at both www.philipslumileds.com/resources/design and www.futurelightingsolutions.com/services/tools.asp.
Click to expand...

In the industry of Commercial Lighting, standards are just now being recognized that allow products to be properly evaluated and compared. These "tools" bring up some of these important factors. Keep in mind that an LED is not a system or product in itself. You have to build something around it to make it work. It's "real" efficiency, performance, and life will be totally dependant on the design and construction of the "fixture" and system.
Right now, Technology in the LED itself is quadrupling every two years or so. The DOE expected technology to double every four years but recently republished their expectations.
We are telling our clients that are interested in commercial LED products that we are happy to sell them, but be aware, in two years you will have a product that is twice as good at half the price. Almost as bad as Avionics!
 
You must log in or register to reply here.
Back
Top