The problem is that a 50F supercap is considerably more expensive than the LiPo battery, and a typical 50F supercap is only rated at 2.5v so in practice you'll need 2 x 100F in series, 3x the price of a small LiPo.

I haven't any experience with super caps. I was just postulating a possibility for a solution. It probably isn't the best solution, but it offers food for thought. And from what I've read since my post it appears some people are starting to think along the lines of a super cap. I know I don't have all the answers. And sometimes my answers are just plain wrong. I'm not special in any way. I just like to sometimes throw out an idea and see where it goes. Light weight and cheap - I get that. Makes perfect sense to me. And I never thought of a 50F cap - geez, that'd be pretty dang big and the Q-copter couldn't fly under that weight. Also, the reason why I went with the SC (super cap) was to avoid having to change batteries (as with AAA's) or placing a lithium battery that will need to be charged before the flight, otherwise one could forget to charge it and crash, then have a very hard time finding the QC.

Just wanted to present an alternative thought to carrying an on-board battery which would be heavier than a SC.

 

I haven't any experience with super caps. I was just postulating a possibility for a solution. It probably isn't the best solution, but it offers food for thought. And from what I've read since my post it appears some people are starting to think along the lines of a super cap. I know I don't have all the answers. And sometimes my answers are just plain wrong. I'm not special in any way. I just like to sometimes throw out an idea and see where it goes. Light weight and cheap - I get that. Makes perfect sense to me. And I never thought of a 50F cap - geez, that'd be pretty dang big and the Q-copter couldn't fly under that weight. Also, the reason why I went with the SC (super cap) was to avoid having to change batteries (as with AAA's) or placing a lithium battery that will need to be charged before the flight, otherwise one could forget to charge it and crash, then have a very hard time finding the QC.

Just wanted to present an alternative thought to carrying an on-board battery which would be heavier than a SC.

Supercap works fine, actually in this application it is the best solution when you only need energy for "short" period of time. 50F supercap just is absolutely bonkers, total overkill, weights and cost a lot, and takes space, just what TS tried to avoid. Just as having AAA batteries for buzzer on a 200g racing quad...

 

Price, weight and size is all based on a proper designing. With 3 components you achieve none of those. Energy storage is the largest, most expensive and heaviest of all these parts, figure out first how to reduce the size of the energy storage, then the rest is easy.

The key is to lower your current consumption. This is easily done by pulsing the output instead of keeping the buzzer on all the time. Pulsing the output say every one second you get 0.1second beep, that brings down the current consumption by 90%, so comes down the energy storage requirement by 90%, as comes down the weight and price. This way 5V 3.3F supercap can hold energy close to 2 hours, and weighs 3grams. You have perfect circuit for this on my post #15, just convert it tiny bit for this purpose, like this:

View attachment 246823

Its not just about weight but volume as well. Yes, energy source is important but I've not found a source for a 3.3F 5v cap from a reputable manufacturer & supplier (Mouser, Digikey, RS, Farnell, etc) anywhere close to your 3g spec, link please.

The best I've found is a Cornell-Dublier/Illinois Capacitor DGH355Q5R5 from Digikey typically $3.5 each, 20 up and weighs 6g, volume 6.5cc. By comparison a CR2 camera battery, for example, is $4 - $5, weighs 11g and is 5.7cc (30hr continuous) and even a CR14250 (1/2AA) is $3.5 - $.5, weighs 11g and 5.1cc (20hr continuous). Given most flights don't end in a crash, these primary cells will last a year at least. I'm sure there's a primary or rechargable solution that will meet or better the supercap one. I'm not anti-supercap, used them many times, just saying they have their own issues..

Charging the supercap is an issue. it needs over 45s to charge from the main battery through a 3ohm resistor and diode to just over 4.6v, during that time it draws over 300mW from the main battery... @bigjoncoop How big is the main battery? If this is unacceptable you'll need to find a way to externally charge the supercap at least once per session and check before each flight.

 

Its not just about weight but volume as well. Yes, energy source is important but I've not found a source for a 3.3F 5v cap from a reputable manufacturer & supplier (Mouser, Digikey, RS, Farnell, etc) anywhere close to your 3g spec, link please.

The best I've found is a Cornell-Dublier/Illinois Capacitor DGH355Q5R5 from Digikey typically $3.5 each, 20 up and weighs 6g, volume 6.5cc. By comparison a CR2 camera battery, for example, is $4 - $5, weighs 11g and is 5.7cc (30hr continuous) and even a CR14250 (1/2AA) is $3.5 - $.5, weighs 11g and 5.1cc (20hr continuous). Given most flights don't end in a crash, these primary cells will last a year at least. I'm sure there's a primary or rechargable solution that will meet or better the supercap one. I'm not anti-supercap, used them many times, just saying they have their own issues..

Charging the supercap is an issue. it needs over 45s to charge from the main battery through a 3ohm resistor and diode to just over 4.6v, during that time it draws over 300mW from the main battery... @bigjoncoop How big is the main battery? If this is unacceptable you'll need to find a way to externally charge the supercap at least once per session and check before each flight.

You can reach one hour working time (5V to 3V drop or even further, depending how low the circuit can go in voltage) for given pulsed output with 1,5F cap and 3grams, that was the TS specification for maximum time. Cornerr Dubilier DSF155Q6R0HAE for example weighs 3g (including wires).

Charging is no problem, the charge remains in the cap when it is once charged and it can be pre charged at first. Only during crash it is drained. Sooner you find the crashed Quad less you loose the charge. My miniQuad is with 4S 1300mAh LiPo.
It's is just numbers, build the prototype, test the timing and specially what is the shortest pulse from buzzer to satisfy your needs will eventually help further reduce the current consumption, maybe only 50mS is enough every two seconds?, this will again double the working time... It is all about iteration and optimization.

 

It's is just numbers, build the prototype, test the timing and specially what is the shortest pulse from buzzer to satisfy your needs will eventually help further reduce the current consumption, maybe only 50mS is enough every two seconds?, this will again double the working time... It is all about iteration and optimization.

Fair enough, though given the relative complexity of your circuit , compared to the attiny85 one, I'd probably revert to that. The power consumption isn't so dissimilar and the timing options are more flexible (and PWM sensing could be a software option too).

I think a cmos 555 option could be done too, with less parts/joints. .

 

Fair enough, though given the relative complexity of your circuit , compared to the attiny85 one, I'd probably revert to that. The power consumption isn't so dissimilar and the timing options are more flexible (and PWM sensing could be a software option too).

I think a cmos 555 option could be done too, with less parts/joints. .

What I like in electronics, is that its like a book, you put the letters in correct order and you get different books, some bestsellers, some trash novels. There are multiple ways to build the circuit, mine was one of the many and sure you can optimize it the way you think it is necessary for ones own preferences. I still would focus on current consumption in this application since the design parameters are light, small, inexpensive, and would continue to iterate the design (what ever possible) for best possible result.

 

Here's the LinMOS 555 timer version, approx 0.1s tone burst per second. For 1F5 supercap at C5, change R3 to 2.4M for approx 0.1sec tone burst per 2 sec. However, I'd still go for the Attiny85 solution running off a supercap, using the watchdog timer to wake the MCU from power-down every 1 or 2 secs and turn on the buzzer.

 

Some ATtiny85 options..