Homemade Canopy flasher

[Mark Mocho]

I cobbled together a $40 canopy flasher and I'm starting to test it for visibility and power consumption.

https://www.facebook.com/groups/331051487600511

[Eric Greenwell]

On 11/19/2022 6:42 AM, Mark Mocho wrote:
> I cobbled together a $40 canopy flasher and I'm starting to test it for visibility and power consumption.
>
> https://www.facebook.com/groups/331051487600511

I can't see it on Facebook (no FB account, and not a member of the Pegase group). Could
you post it on RAS_Prime, which does allow pictures?
--
Eric Greenwell - USA
- "A Guide to Self-launching Sailplane Operation"
https://sites.google.com/site/motorgliders/publications

[Moshe Braner]

On 11/19/2022 9:42 AM, Mark Mocho wrote:
> I cobbled together a $40 canopy flasher and I'm starting to test it for visibility and power consumption.
>
> https://www.facebook.com/groups/331051487600511

Great news Mark. Can you post some details here, for those of us who
are not on Facebook generally or that group specifically?

[Mark Mocho]

Here are Dropbox links to a couple of photos and a 10 second video.

https://www.dropbox.com/s/ehm7ycspge281cx/Flasher3.mp4?dl=0
https://www.dropbox.com/s/4yypemsjfz27vne/Flasher2.jpg?dl=0
https://www.dropbox.com/s/ytd31o6pbz5h60q/Flasher1.jpg?dl=0

[Eric Greenwell]

It looks good, but I wish there was way gliders slightly below you could see it.

[Jim Lee]

I like it Mark! What length extension cord is required?

[Moshe Braner]

On 11/19/2022 12:57 PM, Jim Lee wrote:
> I like it Mark! What length extension cord is required?

Depends how high you fly! :-)

I'm also thinking of a DIY strobe, possibly controlled by a tiny
computer board so it could be programmed in various ways. What are
people's thoughts here:
* Is a red strobe better than white?
* Are two flash units, pointing somewhat to the left and right, better
than a single unit points straight forward?
* Is one flash every 2 seconds frequent enough? (Got to save the battery.)
* Is a double-flash more visible than a single flash?
* Is flashing more often when FLARM gives a collision warning worth the
extra bother (e.g., more cabling)?

[Tony]

Moshe, I will say yes for all 5 questions, though hopefully it would work without FLARM connected

[Mark Mocho]

I've been running a power consumption test with a 12-volt, 5 Amp hour Sealed Lead Acid battery. The LED flasher has been running for 21 hours continuously and the battery voltage dropped from 12.37 volts to 11.97 volts. I think the consumption is acceptable for a glider system, but I will probably run it on a separate battery anyway. My glider has two 15 Ah LiFePo batteries, but I also have a provision for a separate 12-volt 10 Ah cell.

[Mark Mocho]

It looks good, but I wish there was way gliders slightly below you could see it.

Sotecc also has streamlined fuselage flashers for top, bottom or both, but they are also expensive. I am looking for something that might work as a fuselage flasher. These look interesting, but at 50 Watts, the power draw may be prohibitive.

https://www.amazon.com/Yorkim-Bright-Lights-Reverse-Chipsets/dp/B01A4Z3FOS

[Eric Greenwell]

These are not 50W bulbs, despite the spec on the listing. Take a look at the listing for
their newer version of the bulb, where it's listed as 4W. Even the tungsten filament 3157
bulbs are only 6 watts or so.

https://www.amazon.com/Yorkim-lumens-Reverse-Running-Replacement-dp-B09HZ6F3HD/dp/B09HZ6F3HD/ref=dp_ob_title_auto

[Moshe Braner]

On 11/19/2022 2:57 PM, Mark Mocho wrote:
> I've been running a power consumption test with a 12-volt, 5 Amp hour Sealed Lead Acid battery. The LED flasher has been running for 21 hours continuously and the battery voltage dropped from 12.37 volts to 11.97 volts. I think the consumption is acceptable for a glider system, but I will probably run it on a separate battery anyway. My glider has two 15 Ah LiFePo batteries, but I also have a provision for a separate 12-volt 10 Ah cell.
>

But is it bright enough to be seen from half a mile away in full
sunlight? In a head-on situation with a relative speed of 2 miles a
minute, half a mile is 15 seconds away.

[Moshe Braner]

On 11/19/2022 4:41 PM, Eric Greenwell wrote:
> On 11/19/2022 12:05 PM, Mark Mocho wrote:
>> It looks good, but I wish there was way gliders slightly below you
>> could see it.
>>
>> Sotecc also has streamlined fuselage flashers for top, bottom or both,
>> but they are also expensive. I am looking for something that might
>> work as a fuselage flasher. These look interesting, but at 50 Watts,
>> the power draw may be prohibitive.
>>
>> https://www.amazon.com/Yorkim-Bright-Lights-Reverse-Chipsets/dp/B01A4Z3FOS
>>
>>
> These are not 50W bulbs, despite the spec on the listing. Take a look at
> the listing for their newer version of the bulb, where it's listed as
> 4W. Even the tungsten filament 3157 bulbs are only 6 watts or so.
>
> https://www.amazon.com/Yorkim-lumens-Reverse-Running-Replacement-dp-B09HZ6F3HD/dp/B09HZ6F3HD/ref=dp_ob_title_auto
>

Maybe they meant 60W-incandescent-equivalent. Like they label the
household LED bulbs.

You can probably drive these harder for short pulses, but you'll need
more than 12 volts, since these are set up with internal resistors to
limit the power to what they can handle continuously.

All those incandescent-replacement LED bulbs for automobile tail lights
are built to send light in all directions: multiple LEDs are mounted on
it, all around. For that to work on, e.g., the belly of a glider, you'd
need a transparent faired bubble deep enough to expose the whole
LED-mounted portion of such a bulb.

And then again, will it be visible from a relevant distance? I doubt it.

[Mark Mocho]

> And then again, will it be visible from a relevant distance? I doubt it.

I haven't done any daylight range tests yet, but I can say with some confidence that it will be brighter and more noticeable than the one you DON'T have.

[Moshe Braner]

On 11/19/2022 7:47 PM, Mark Mocho wrote:
>> And then again, will it be visible from a relevant distance? I doubt it.
>
> I haven't done any daylight range tests yet, but I can say with some confidence that it will be brighter and more noticeable than the one you DON'T have.
>

Right :-)

One way to make it "brighter" is to concentrate the beam to a narrow
angle. A friend who has the Sotecc unit says it is noticeably dimmer
more than 10 degrees off-axis, and not visible at all 30 degrees off.
(This fits the diagram in the product page on W&W.) They claim more
than 2 miles (!) visibility.

But, that means no visual warning to aircraft that are not approaching
head-on. Seems to me that most collisions are not head-on?

[Mark Mocho]

True enough. Off-axis bearings are much less visible with the current configuration, but once again, anything is better than nothing. It is like relying solely on Flarm, knowing that intermittent contacts are common due to poor antenna placement, blockage of the signal by carbon fiber fuselages, inability to interpret warnings quickly, and other factors. None of these reasons are valid excuses to NOT have Flarm; they are merely caveats to remind us that nothing is perfect.

I am also going to experiment with side mounted flashers inside the canopy. Of course, power requirements will be higher, but nothing like what is needed for standard xenon anti-collision lights. Bright LED technology is quite amazing, considering the low cost, low energy demand and light weight. True 360-degree coverage in both horizontal and vertical planes is not feasible, but as I said before, anything is better than the current "stealth" mode.

I have also learned that LED lights can be made substantially brighter by increasing the power, at the penalty of reduced life. I don't really have any method of testing this, as the flashers are already so bright that they are potentially damaging to the eye if viewed for more than a few seconds at close range. And I am not sure whether boosting the input voltage would do anything with commercially available, inexpensive flashers which may have over-voltage protection circuitry.

Having observed the Sotecc flashers a few times this last season, I can attest that, under certain conditions (like under a cloud shadow), two miles is entirely possible. I have seen the red flash before a glider is discernable a couple of times.

[Dan Marotta]

A couple of possibilities:

A spherical reflector whose aberration would send some light off axis.

Or how about two units - one on top of and one on the bottom of the
fuselage? Yes, they'd stick up into the air flow a tiny bit but, for
most of us, the drag increase would not be noticeable. AND you'd get
360 degree visibility except from directly above and behind.

Dan
5J

[Moshe Braner]

On 11/20/2022 11:11 AM, Dan Marotta wrote:
> A couple of possibilities:
>
> A spherical reflector whose aberration would send some light off axis.
>
> Or how about two units - one on top of and one on the bottom of the
> fuselage?  Yes, they'd stick up into the air flow a tiny bit but, for
> most of us, the drag increase would not be noticeable.  AND you'd get
> 360 degree visibility except from directly above and behind.
>
> Dan
> 5J
>
> On 11/19/22 20:55, Mark Mocho wrote:
>> True enough. Off-axis bearings are much less visible with the current

>> configuration, but once again, anything is better than nothing. ...

>>
>> I am also going to experiment with side mounted flashers inside the
>> canopy. Of course, power requirements will be higher, but nothing like
>> what is needed for standard xenon anti-collision lights. Bright LED
>> technology is quite amazing, considering the low cost, low energy
>> demand and light weight. True 360-degree coverage in both horizontal
>> and vertical planes is not feasible, but as I said before, anything is
>> better than the current "stealth" mode.

>> ...


But neither "some light off axis", nor omnidirectional top and bottom
strobes, would be bright enough to see from much distance. At least not
with a power consumption that seems acceptable, e.g., under 0.2 amps.

I think that in order to make the best decisions on which directions to
send the light to, we should determine which directions are the most
likely collision directions. And ignore directions that are not visible
anyway. E.g., at least from a glider cockpit, you can't look down, nor
above-and-behind. Thus if one glider is following above and behind
another one, no strobe location will help either see the other.

[Mark Mocho]

> But neither "some light off axis", nor omnidirectional top and bottom
> strobes, would be bright enough to see from much distance. At least not
> with a power consumption that seems acceptable, e.g., under 0.2 amps.
>
> I think that in order to make the best decisions on which directions to
> send the light to, we should determine which directions are the most
> likely collision directions. And ignore directions that are not visible
> anyway. E.g., at least from a glider cockpit, you can't look down, nor
> above-and-behind. Thus if one glider is following above and behind
> another one, no strobe location will help either see the other.

OK, since the infallible Moshe Braner has decreed that there is no effective solution, I shall immediately suspend all efforts to improve glider-to-glider visibility. Thanks, Moshe!

And I have also discovered that there is a group with equal infallibility called the "Man Will Never Fly Society." Therefore, I urge you all to immediately sell your aircraft to some unsuspecting neophyte who believes otherwise.

[Martin Gregorie]

On Sun, 20 Nov 2022 10:42:25 -0800 (PST), Mark Mocho wrote:

> And I have also discovered that there is a group with equal
> infallibility called the "Man Will Never Fly Society." Therefore, I urge
> you all to immediately sell your aircraft to some unsuspecting neophyte
> who believes otherwise.
>

... their motto is: "Birds Fly-Men Drink."

This gives a strong clue about the purpose of that society. Oddly enough,
it seems to count quite a few model fliers and aircraft spotters among
their membership.






--

Martin | martin at
Gregorie | gregorie dot org

[Eric Greenwell]

RF transmissions, like Flarm, seem to have smaller dead spots than lights, and much lower
battery consumption. Maybe getting the "unFlarmed" gliders to install an OGN Soft-Rf
tracker would be the easiest, cheapest, and most effective way to alert other pilots of
their presence. No collision warning, but the proximity alert and cockpit display would
do, say, 80-90% percent of what we need.

Some people will equip with Soft-RF anyway, just to join OGN, so it might not be so hard
to equip all the "unFlarmed", even if the SSA gave them away (or sold cheaply).

[Hank Nixon]

Next easy test is take it to the runway and see how far away you see it effectively.
Comparing to store bought would be an even better measure
It looks like a pretty good start to me.
Good job
UH

[Mark Mocho]

> Next easy test is take it to the runway and see how far away you see it effectively.
> Comparing to store bought would be an even better measure

That is pretty much the plan, but I don't know of any installed Sotecc units at the field yet. One is scheduled for installation sometime this winter, but I don't know when. I am going to get my installation completed and try it outside this week. Will let you know the results.

[Moshe Braner]

I didn't say there is no solution, and I support your efforts Mark! And
I'm dabbling in the same approach (plus software) myself. Just trying
to discuss what would be the most effective approach, given the limited
power supply. Sotecc clearly think that a forward-only strobe is the
best compromise. That may or may not be true. E.g., I was proposing we
consider the pros and cons of having two strobes, aimed somewhat right
and left of forward.

[Hank Nixon]

My 29 has a fin mounted strobe and it6 is reported to be very visible in crummy conditions FWIW. Probably a somewhat wider viewing angle side to side and vertical. Not practical to try to replicate after market.
The effort is a good one.
UH

[Mark Mocho]

Moshe- Check the Wings & Wheels website. Sotecc also has fuselage flashers for either the top or bottom of the fuselage.

https://wingsandwheels.com/sotecc-strobe.html

Also rather expensive, but a viable solution for all around visibility.

[Mark Mocho]

Don't know if this would work as an effective external omnidirectional strobe, but it might be worth checking out.

https://www.oracletx.com/product/oracle-9-led-omni-directional-emergency-strobe-white/

I don't particularly like the 1-inch hole mounting, but it might work in the inspection cover on my glider.

[andy l]

Perhaps car headlights would be brighter than reversing lights, or there are even aircraft landing lights

Or maybe something like these light bars

https://www.amazon.co.uk/dp/B07SY7LLYX/
https://www.amazon.co.uk/HELLA-1FB-358-176-211-Headlight/dp/B09K998WFK/

I just looked at some Aerosun landing lights or strobes. $320 or $350, so I can see why you're hoping for a bit cheaper

[Moshe Braner]

Bicyclists have even less battery capacity on hand, and some of them now
use tail flashers that seem visible from about a quarter mile away even
in sunshine. The trick seems to be the directionality of the beam: only
straight rearwards? For example: https://www.amazon.com/dp/B07W62CCL2
or a slimmer but somewhat less bright model:
https://www.amazon.com/dp/B07YFDFXY4 - I wonder if any bike light would
be worth putting into a glider as-is? Would need to add a fairing
between it and the canopy.

[Stuart Venters]

"Not practical to try to replicate after market."

On the leading edge of the tail seems a really neat place.
Would it become practical if one repurposed the rear battery wires?

[Mark Mocho]

https://www.amazon.com/dp/B07W62CCL2
or a slimmer but somewhat less bright model:
https://www.amazon.com/dp/B07YFDFXY4 - I wonder if any bike light would
be worth putting into a glider as-is?

The two examples you list have 90 and 350 lumens. The Sotecc advertises 5,000 lumens. I am trying flashers with about 2,500 lumens, and I am not sure even they are bright enough.

[Eric Greenwell]

Or this focused, high intensity unit from Garmin that can be seen a mile in daylight:

"Garmin Varia UT 800 Smart Headlight Urban Edition"

[andy l]

I think it would also involve reworking the leading edge structure

It looks like the LED packaging goes to several mm back from the l.e. apex, so I assume there are temporary inserts on both mould halves during manufacture, to produce a blunt shape as a base for it.

Doing it afterwards would be like cutting an inch or two off the front then going from there, with maybe something like a flared U channel joined to both sides

And that scheme might be sufficiently different from the original manufacture as to need separate approval.

[Hank Nixon]

Wires are trivial.
The issue is structural.
On the gliders we fly the leading edge is either a butt joint , or most commonly, a lap joint where both skins are glued together, and usually having a quantity of epoxy/filler to bulk up the joint. This joint is one of the major structural elements that resist torsion. If one were to remove this area over the amount on my glider, about 12 inches, one would then have to fabricate a channel shaped part that would transfer the loads. Also producing a new face to match the leading edge profile and have the right transmission properties is another task requiring non common expertise. It can be done, but this is way beyond the kind of a thing a hobby type person should be doing.
UH

[Mark Mocho]

I seem to recall a glider (not sure what brand/model) at one SSA convention that had a fin mounted LED flasher that appeared to be "scarfed" on the leading edge. Looked like a molded housing with a long LED strip down the middle. You'd still have the hassle of running the wires, but at least you wouldn't have the challenge (and legal issues) of modifying the structure. Anybody else remember this?

[John Foster]

I believe forward-facing LED flashers would be the best compromise, due to what has already been brought up regarding closing speeds/times for head-on encounters. When approaching from the side or behind, closing speeds are typically much lower than head-on encounters, giving more time to react. As such, I believe head-on encounters to be more critical with regard to visibility.

[Sarah Anderson]

Thanks for experimenting with this, Mark! I'm glad to hear innovation is not dead yet.

I think the nose (or canopy forward) installation is the most important, due to the closing rate on head on conflicts.
More bang for the buck... well, least "bang".

Second most important would be 90-degree closure. A top, belly or dual-wingtip option would work there but it's difficult to see
how to install it.

[youngbl...@gmail.com]

On Monday, November 21, 2022 at 11:13:04 AM UTC-5, Mark Mocho wrote:
> I seem to recall a glider (not sure what brand/model) at one SSA convention that had a fin mounted LED flasher that appeared to be "scarfed" on the leading edge. Looked like a molded housing with a long LED strip down the middle. You'd still have the hassle of running the wires, but at least you wouldn't have the challenge (and legal issues) of modifying the structure. Anybody else remember this?

That would be a Alisport Silent 11. Yet it is a Motorglider . This is an excellent light, could be incorporated into current models of sailplanes and is extremely bright. Even Old Bob, The Purist knows a bit about motorgliders

[Charlie M. (UH & 002 owner/pilot)]

On Monday, November 21, 2022 at 11:13:04 AM UTC-5, Mark Mocho wrote:
> I seem to recall a glider (not sure what brand/model) at one SSA convention that had a fin mounted LED flasher that appeared to be "scarfed" on the leading edge. Looked like a molded housing with a long LED strip down the middle. You'd still have the hassle of running the wires, but at least you wouldn't have the challenge (and legal issues) of modifying the structure. Anybody else remember this?

ASG-29 has a LE fin mounted LED as an option, "UH" has one and he has had good comments on it. But again, this was a factory option when the glider was built, not an aftermarket unit.

[Moshe Braner]

For those interested in building a canopy strobe / flasher: software now
available for controlling such a device. Including increasing the
flashing frequency when an attached FLARM gives a collision warning.
And other features.
https://github.com/moshe-braner/SoftRF/tree/master/software/firmware/binaries/ESP32/SkyStrobe

[Mark Mocho]

Just found these:
https://luxlightingsystems.com/products/max-rgb-color-changing-led-rock-lights
A bit more expensive, but they appear to be nearly ideal for what I am trying to do.

[Moshe Braner]

Interesting! What in the world is a "rock light" - people mount then
UNDER jeeps? Does the polymer body diffuse the light into all
directions? Are you planning to use these mounted on the outside of a
glider? They're supposed to "mount" on steel objects via the built-in
magnets - will you glue them instead? Will you use only the red LEDs
within?

[Charlie M. (UH & 002 owner/pilot)]

Yes, under an off-road vehicle for night trail running or rock crawling.
The directions state "mount on metal with the magnets", we understand it has to be a ferrous metal like steel in a car/truck. The magnets are also part of the heatsink, thus that has to be considered as the steel mounting panel is more surface area for heat dissipation for longer LED life.

[Moshe Braner]

The heat dissipation is needed if they are continuously lit. In use as
a glider strobe, they would be driven (even over-driven) in short
pulses. Thus much less heat is generated. E.g., the LED module I'm
first trying out is rated 1 watt per LED, I drive them at 2 watts for
short pulses. The default in my "SkyStrobe" software is one burst every
2.4 seconds, each burst 3 flashes of 40 milliseconds each. Thus the
LEDs are lit only 4% of the time - 25 times less heat. That increases
to 20% of the time when flashing more frequently due to FLARM reporting
a collision warning. (These are preliminary timings, will tweak as
experience is gained in actual use. And there's a way to change the
defaults.) The software (which can also drive buzzer warnings) runs on
a $10 microcontroller.

[Moshe Braner]

Correction: that's 5%. The 4% came from my previous default, two 50 ms
flashes every 2500 ms. I find the triple-flash, which is spread over
more time, to be more eye-catching than the double-flash. At 8 watts
momentary total (with 4 LEDs X 2 watts each), the 5% duty cycle is 0.4
watts average, or about a 0.03 amp drain on your battery. The
microcontroller alone will use a similar additional amount! The total
power consumption is similar to a simple audio vario. Could easily
accommodate one or two additional LED modules.

[Dan Marotta]

Not wanting to place my GPS antennae too close together, I used magnetic
mount antennae attached to steel washers that I glued to the canopy
rails in my glider. During canopy jettison, they'll simply pull off.

Dan
5J