Glider canopy flasher

Other Types of Lights LED Spotlights
1

I am looking to build a canopy flasher to install in a sailplane and am after some guidance on components.
Commercial solutions (eg aeroflash ) are out of my budget!

Ideally it needs to be clearly visible in full sunlight from 3nm / 3.5 mile / 5.5km. Two gliders approaching head on can have a closing speed of 300-400 kmh.
There are good arguments for using red LED’s, but I think I’ll start with white for simplicity.
Gliders have 12V power, typically a 7.2Ah AGM lead acid or two. Power consumption is a concern.

I simply want a unit that flashes (40-60 fpm, 0.7-1 hz) when turned on.
My reading suggests the Luminus SST-40 is a good choice for the LEDs. Probably with something like a Carclo elliptical ripple TIR lens.

When it comes to drivers I get a bit lost.
It seems to me that a 5A driver should be about right but when I look at some of the budget suppliers I can’t figure if I can drive multiple LED’s from the one driver.
I also struggle to find something that I can be reasonably confident that I can organise so it just starts up in Strobe mode - it’s very likely that you really don’t want the strobe turned on while sitting at the launch point with people helping hooking you up to winch/tug, and once airborne the last thing you want is to have to do more than flick a switch.

Any pointers gratefully received.

2

Please enjoy your time here, Bumbled!

01 Scarlett Johansson
01 Scarlett Johansson2000×1181 193 KB

3

Sorry, I’m not much of a fly-boy, though I hope to earn my “wings” eventually :innocent:
But I think of going the way of the “common” LED’s won’t bring good results.
I suggest you look in the direction of XENON strobe lights that are used by divers and on sea.
Of course these are always battery-driven, but I would not be surprised if separate parts can be bought.

Xenec Strobe And Flashlight
Xenec Strobe And Flashlight845×1024 126 KB

4

Thanks for the suggestion and taking the time.
From my reading xenon strobes are considerably less efficient than LED strobes andI suspect that the commercial vendors would have gone down the xenon path if it was compelling, but they haven’t, they all use LED’s.
Not ruling it out, but I’m inclined to stick with LED.
I believe powered aircraft use xenon strobes, but they’ve got power to burn compared to a glider.

5

That may be so, but the manual of mine says it flashes 10 hours on two AA batteries and is visible from 2 km.

1 Thank
6

Hmm, I suspect a certain amount of marketing optimism there.

That’s 2km, at night, with little or no ambient light and with that large dose of optimism.
I’m looking for something that is visible at least twice that distance, in full sunlight, at 300kmh closing speed.
I suspect it would take a vastly larger xenon tube with some serious high voltage driving it.

I’m looking at xenon’s after your suggestion but initial reading suggests it’s likely to get all a bit out of hand compared to LED.

7

The aeroflash doesn’t specify intensity so it’s impossible to validate their visibility claims.

You can calculate what you need (hypothetically) using calculators here.

8

Please consider this to be a thought, and not a reaction, or an answer, to one of your statements above.

My XENON flasher sits in a hemisphere and therefore is visible for all who are situated above the flasher.
That’s why commercial airplanes have one on their back and one on their “belly”.

Unless you are thinking of a COB or a mule, LED’s are only visible at the front (lens, TIR).
It is easy to think of avoiding a collision with a glider that is coming from the opposite direction.
But what if the other glider comes from the left, or the right, or above, or below, or from behind you?

9

Aeroflash say “4000-4500 Lumen” and appear to use elliptical PIR lenses aligned horizontal on 4 LED’s and vertical on two. So the majority of those lumens should be in a fairly narrow band in the plane of the aircraft’s movement.

I’ll check out the calculators a bit more, but without figures on what the human eye responds to it’s a bit hard to know the relevance of any figures produced. I’ve no idea of those figures so am just going with “I want something really, blindingly, bright at closer range”.

1 Thank
10

There are a few restrictions for gliders:

  • Power limitations - that 7.2Ah battery might need to last a solid 6 hours + of flying and still have sufficient capacity to make strong radio transmissions at one of the most hazardous times of the flight - airport approach & landing.
  • It’s illegal to cut anything larger than a 6mm hole in the skin of the glider, and sticking something protruding out of the fuselage is generally anathema to glider pilots seeking the best glide performance.
  • Canopy flashers aren’t ideal, no question. I’m sure in years to come gliders will come with flashers aerodynamically integrated into the tailplane and belly so they are visible from multiple directions. But for now the forward direction tends to be highest risk and it’s possible to relatively easily install a flasher inside the canopy out of the airstream, so that’s where my focus is.

Bottom line: something is considerably better than nothing.

Sotecc make a streamlined external strobe - nice looking, but making something like that is a job for later, if ever.

I’m thinking 2 or 3 SST-40’s with elliptical TIR’s at 5amp should be pretty darn bright and hopefully at 1Hz or less the average power draw will be suitably low. Without knowing the flash duty cycle I’ve no way of even estimating what that consumption is so it’s “try it and see”.

I’ve now seen some stuff about different firmwares (guppydrv etc) on drivers that seem to allow setting them up so they stay in a single flash mode. I just haven’t figured how you discover what driver has what firmware.
But I’m still a bit lost on drivers.

11

Just some thoughts, I’ve separated this into the main components I see as being the challenges, I’m not an expert by any means but maybe these will provide you with inspiration:

Power:
As I’m sure you’re aware, Pb-acid “real capacity” is generally half nameplate, though 7.2Ah units aren’t too expensive to replace if you kill it by over-discharging.

So, that gives you c. 3.6Ah x 12V = 43.2Wh to play with.

Divided by (worst case) 8 hr flight, is 5.4W per hour of energy available.

I’m assuming your desired drive current of 5A is for each LED, so (5A x 3.2V) X 4 LEDs = 64W.

You can afford a duty cycle of c.9%. Apologies- I don’t know how to calculate the radiator requirements but this 9% will help reduce your required heatsink size at any rate.

Heat:
It’s unclear from the commercial product pictures where the heatsink actually is; I’m assuming their numbers are “at LED” numbers rather than output numbers which include losses in optics.

The commercial product claims 4000 lumens from 6 LEDs, at an efficiency of 175lm/LED Watt, this is maybe a 1.2ish amp current per LED for c.3.8W per LED, total of 23W, duty cycle will reduce the heat load on the heatsink.

You are looking at building something more powerful so will need better heat sinking.

Driver:
This is likely going to be one of the most tricky obstacles, I’m not a “driver guy” so take my advice with a pinch of salt- I once looked up how to make an LED flasher with a 555 circuit but decided it was a tad beyond my electrical engineering capability and the power output would have been insufficient for my intended use.

You likely could use a flashlight driver with your own firmware, there are lots of topics about this on this forum.

I’m guessing the commercial product is using a boost driver with the LEDs in series, or 3S3P arrangement. While it is possible to run LEDs in parallel, it’s generally better to run them in series.

Housing:
Unless you have something in mind already, I suspect easiest solution will be to design and 3D print a housing to allow you to mount LEDs+Optics+driver+heatsink to your canopy.

12

Awesome, thanks for helping bring some clarity to the process I need.
You say:

I don’t quite follow this.
4000 / 6 = 666.7 not 175.
What am I missing?

EDIT: Got it I think. 175lm/WATT, not LED.
4000 / 175 = 22.86 W
22.86 / 6 = 3.8 W/LED

Driver
I’m really struggling here at present. I don’t know how many drivers I’ve looked at but it seems really hard to find a driver that:

  • Accepts say 9-13V input
  • ~5A max output. 3A would be fine to get going with.
  • Strobe mode that can be locked to default, or can clearly be re-flashed, although I really don’t want to get into that if I can avoid it.
  • Can drive 3 x 3V LED’s (eg SST40).

From what I think I’ve learnt, it would be best to find a driver that puts out ~9V and wire the LEDs in series. They seem rare!

I’ve found a few 5-15V ‘single mode 100%’ drivers. I wonder if it might work to fire these on/off with an external timer? Maybe something as simple as a 555 driving a FET? Would that work or is there some reason that would be a bad idea? I could put each LED on it’s own driver and fire them simultaneously from the one timing circuit. Seems a bit optimistic but maybe it would work?

Also found some drivers with Memory mode, but I’ve no idea if that memory applies to Strobe.

Any ideas what happens if you press & hold but never release the button on the drivers that use P&H to go to Strobe? Maybe I could use a toggle switch to turn them on - the toggle would hold forever (until turned off).

Finally the units with side switches - could I use the toggle switch idea on that?

Single Mode 2.5A
Single mode 2.8A
Long press strobe Could I connect a toggle switch that stays on, hence providing a long (very, very long!) press?
XHP 1-mode 4.5A Could I use this with 2xSST40 in series? 2 x SST40’s at 4.5A might be enough light.
5A side switch Again, could I use a toggle switch to “long press” the side switch? Does memory mode work to turn on default in Strobe?

If anyone can shed light on any of these questions I’d appreciate it.
Failing that I guess it’s start buying things and see what works.

1 Thank
13

Yes! Apologies

1 Thank
14

Using a 12v lifepo4 battery can give you way more power for less weight.
And will last through many many more charge/discharge cycles than a lead battery. 50Ah for 12lbs.

I wonder if something could be cobbled together from a bunch of camera strobes.

3 miles in daylight against a bright sky?
Maybe also consider a flasher from a cop car or other emergency vehicle?

Let us know if you find/build something.
All the Best, Jeff

71NidiY6B7L.AC_SL1500
71NidiY6B7L.AC_SL15001500×1500 145 KB

15

I’ve also wondered about camera flashes but haven’t followed up on it. Maybe I should.

Some people use LFP batteries, 12Ah in the same form as 7.2Ah pb. But they cost 4-5x pb & it’s illegal to charge in flight (solar panels), only approved models allowed. Also can require new weight & balance for the aircraft. Definite advantage to the discharge curve though.

My current thinking is something like this LED Brake Light Flasher

Driving the input of 2.8A single mode driver

I’m not sure how the driver will respond to power on/off cycling, nor if multiple units will all flash precisely simultaneously when driven like this.
I’m mildly optimistic that it could work though.

Still trying to find a straight flasher driver.

16

Here is video from a few years ago.
Really is not pertinent to your quest.
Shows how much horsepower is needed to be seen a long way off.
Just question, can you really see those high $$ lights 3 miles away in daylight?
All the Best, Jeff

17

Interesting.

I really don’t know how effective the commercial flashers are, but the general view seems to be that they can be noticed from several km away.
The powers-who-be seem to think they work because they are starting to mandate canopy flashers on aircraft in competitions.

I had a bit of a look at camera flash units. Not so cheap even from Aliexpress surprisingly.
I think I have an old unit sitting around, I might drag it out and see what distance it is visible from then decide if it’s worth exploring further. Would need to build provide a voltage regulator to supply it, and a trigger. From days of yore with SLR cameras you had to wait several seconds for the caps to recharge unless you had a seriously expensive flash unit, and even then the expensive ones tended to be capable of providing a short series of flashes, then have to wait for a while to charge up again.

I also looked at aviation strobes. Some of the strobes themselves are quite reasonably priced but often the drivers are not. The only ones I found that came in at a reasonable total price were omni directional - while potentially a good safety feature, it just doesn’t lend itself well to installing inside the canopy. There are even some plug in globes that contain the driver/flasher circuits, but again omni directional.

18

Unno, but if it were me, I’d skip the drivers and wire up my own board with a capacitor dump. You ain’t dumping amps but joules (½CV²). Result is a brief but intense flash. Look up the appropriate maximums for LEDs to make sure they can handle it, despite the extremely low duty-cycle.

The sound of it (charge up to its trigger voltage, dump to the LEDs, rinse, repeat) sounds like a 555 can in fact be used for its internal comparator, flipflop, voltage divider, etc., albeit to control an external FET. Could still be pretty compact.

Go crazy…

19

It’s probably trivial for someone with the skills & knowledge to do that, but it would take me days or weeks to work out how to design that & build it.

I understand the concept, but the implementation is another matter.
Sad but true.