Hello fellow flyers

Well I finally have a plane that can be flown with a wide throttle range, level flight at around 1/3 stick and almost 1:1 thrust to weight at full, and I’ve found a new problem, how do I know when the battery is empty? I have my flight timer set at 4 min and after landing my packs (10 cell CP2400), will take from 1200 to 2100mAH depending on how aggressively I’ve flown. I’ve only flown the plane a dozen times so I’m still getting a feel for how the power feels as it approaches LVC, I have yet to hit LVC in flight, but I have once taxing back after landing and I had no idea I was so close. I’ve never had a plane with this kind of power to weight ratio (75W/lb), and the power is still very strong very late in the charge and the cells dump quick at the end (as good NiCd's should). From my calculations I should get 7 min of lazy circuits or 4 min of near full throttle aerobatics, it’s when I fly a mixture that it gets tricky. I suppose in time I’ll get a better feel for the plane and it will be less of a problem. But this situation gave me an idea...

What if you measured mAH consumed and set off an alarm at a set limit, leaving enough power left for an approach or two. You could have a device on the plane that connected between the battery and ESC (like a Wattmeter), with a programmable alarm point, this method would be fairly accurate but the alarm could be hard to hear at a noisy field, and you'd have one more thing in the plane to fail. A better way might be to have a feature in the TX that would let you calibrate the throttle stick position to a given rate of power consumption, maybe a three point curve (0, mid, full), a program in the radio would compute time and stick position to determine mAH consumed. This system would be less accurate than the direct measurement but would be easy to hear and could be configured for use with fuel models too (makes it more likely that manufacturers would consider making it). I guess you could even do it with a aftermarket device that read the pulse width of the throttle channel through the trainer port.

Just a thought… somebody get Mike for Hitec on the phone.

Gary

Anyone… anyone… Bueller… Bueller… Bueller?

Well I thought is was kinda a neat idea.

I've been thinking about doing something similar for a while.

The main problem I think is the measuring of current. The normal method of doing this is bu using a current shunt (very accurate low value resistor) but the only ones I can find are reasonably heavy for a 10 Amp shunt. Also the rapid on/off cycle of the ESC means that you would have to do some clever smoothing of the readings you get in order to get a good reading.

It's certainly possible, and there are some solid state current measuring components out there, but I'm not sure any would work for an application like this (most work on AC)

Barny

Originally posted by ybanrab
Also the rapid on/off cycle of the ESC means that you would have to do some clever smoothing of the readings you get in order to get a good reading.

Just put it in line between the battery and ESC, like a whattmeter..
It wouldn't be hard to do. You could combine it with a circuit that goes between your receiver and ESC, to 'pulse' the power or something when the capacity got low. Also to make a low voltage monitor for LiPo cells..
..a

One obvious question ... how would you program in the pack capacity (and make it easily changeable)?

The low voltage check for Li cells is being addressed by some people already:

http://www.rcgroups.com/forums/showt...threadid=88852

Something similar (an ultra-bright LED that comes on at a set voltage) would work fine for early warning on nicd & nimh packs too, you'd just need to be able to adjust the 'trigger' voltage to suit the no. of cells you're using.

I like this idea!

Bill, how about a series of jumpers for a custom pack mah setting?

For anyone starting to design this, please come out with one that weighs 1g so as not to affect the flight performance of my scale park flyers

Martin

The MGM Compro controllers already taper the power. When the power drops it is telling you to land. I don't know how the controller knows when the battery is spent, but I think it's probably based on the the voltage of the pack.

Bobby

5 gram voltage monitors are available....

http://www.gws.com.tw/english/produc...ircuit/obi.htm

http://www.aeromicro.com/Catalog/gws...rs_1625121.htm

Here's a method that I use that's simple (although, I too would pay the extra bucks to have real time mah consumption on the transmitter):

For each plane that I fly, I'll do a bench test with an amp meter.
Peak the battery and prep the plane as if ready to fly
move throttle to 1/3 throttle and measure amp consumption
repeat for 1/2, 3/4, and 4/4 throttle positions and put these values on a card or if you are a geek like me, put it in your Palm Pilot.

Beside each value, do a little math to determine running time at a given consumption, such as:
(60/(current/(capacity in mah/1000)))

i.e. if you are using CP1700s, (assume 1600 mah capacity) and note that full throttle pulls 25 amps, than running time at full throttle is (60/(25 amps/(1600 mah/1000))) = 3.84 or 3:51 minutes. This is the shortest run you will have. If you note that you rarely fly at this level, but typically have the throttle ~ 2/3 and note that 2/3 equates to 16 amps or 6 minutes, than look to have the plane on the ground ~ 5 minutes. There are factors such as prop-unloading and such that will extend running time, so tweak as you see fit.

Note that if any variable changes, such as temperature, prop, battery, etc. you will have to retest.

Well if anyone's up to it I'm sure you could design something around a Basic Stamp, you can enter in the battery capacity you want for it to start kicking in an alarm/etc, or even telemetry the data back. Along the same topic ... sooner or later we should be seeing ESC (bush/brushless) with integrated u-controllers. eg. you can have a watt meter onboard along with a A/D converter and PIC, optional tach lead/airspeed pilot tube, some flash RAM for data recording. (fwiw there's already some tach/speed telemetry system availabel for IC).

You don't need to measure how much capacity you have used. Your flightpack will do that for you. Just monitor the voltage, and beep when it gets to about .9V per cell.

I can tell cos my models stop climbing about then anyway

The problem I see with the voltage sense method is that the voltage drop is relative to current draw… If I hit the voltage alarm in a full power climb (about 35 amps), I can throttle back and will most likely have enough power to position for final. But, if I’m strolling along at slow level flight (around 15 amps), and hit the warning the pack has dumped and I’m most likely in big trouble if I’m not in position to drop right into the approach. The real advantage to knowing the mAh consumed is (assuming a complete charge), is you know what you have left.

See if my logic follows, If I can get from any place in the flying to final approach in one min (should be much less) and my current rate for that min is 25 amps (assuming I might need to climb some), then the math says I need about 420mA left, so figure 500mAh (or more if you might miss the first pass). Now my CP2400's consistently take 2200mAh on charge, figure 2100 to be safe, so I set the alarm to go off at 1600mAh and no matter what mix of flying I do I’m safe. Make sense?

Having the system in (or adapted to) the TX would be the best way and I’m sure it could be done, just need some smart guy to figure it out, build it, and sell it for say… $19.95?

Gary

Originally posted by All__talk
Hello fellow flyers
...

A better way might be to have a feature in the TX that would let you calibrate the throttle stick position to a given rate of power consumption, maybe a three point curve (0, mid, full), a program in the radio would compute time and stick position to determine mAH consumed.
...
Gary

This is amazing. I thought of it last sunday, while seting up my timer to 5 minutes, due to medium to strong winds blowing where I was going to fly.

In fact, it would be a very good idea a tx programable throtle alarm with a pre-selectable multipoint positioning of the throtle, each of them corresponding to a directly mearured amp draw, with thte total and cumulative amp draw being calculated through linear functions between the throtle positions and the fligth time measured by the timer. Then one could set up the timer for a perecentage of the cumulative amp draw / pack capacity.

A better idea, AS YOU SAID, would be to produce an external device that could perform this function, and one could intsall by ourselves in our transmiter by either linking it to a port (yes, the trainer port as you said) or somewhere inside, in the circuit (I do have a prism 7 without a trainer port). And the alarm sound should be provided by this external device, with a sound volume control (again and as you said, some IC engines are really noisy, too noise for the low volume of the TX sound).

Well ... hopes and thoughts, like the dreams, do not pay taxes ... yet ...

PT