Taranis Continues to Read the Last Vfas When Battery Disconnected

FrSky Current sensors

FrSky sells two current sensors, 40A and 150A. I hope they introduce one in between. 40A is a bit low for anything other than small and medium models. 150A is over the top for all but large and EDF models, and it has to be threaded on to the red ESC wire so requiring unsoldering the connector. It's probably best to make up a special lead with XT90 connectors on each end. The 40A one has male and female XT60 connectors as you see from the picture.

OpenTx has a very clever telemetry function called Consumpt. This takes the current (Curr) from a 40A FrSky current sensor, and presumably time from its clock, and calculates how many mAh of energy have been used. It removes the guesswork involved in relying on the varying voltages from a lipo sensor. However only the 40A version reads the battery voltage as well.  If you want a voltage reading, for the 150A you would need also to have a lipo sensor to give the data item Cels.

You can calibrate both Curr and VFas (battery voltage). This involves using a multimeter to read the true current and voltage and comparing them with the values shown in telemetry. If either is different, you open the Edit screen for the datum and enter a Ratio to correct the difference. For example the telemetry shows 22.5V and the meter shows 23.6V. The ratio is 23.6/22.5 or 1.05.

Testing the accuracy of Consumpt

I did a bench test with a tethered model. Unlike the nitro heads I was able to do the tests indoors in the warm, not at 3°C in my workshop. I used a brand new, newly charged, 2200 mAh Zippy Compact lipo and a similar, used, newly charged, Graphene Panther. Each had similar measured internal resistances of about 3mΩ. I ran the motor mostly at about half throttle until the Consumpt reading showed that 75%, 50% and 25% of the energy was left. I then stopped the motor and noted the battery % readings by connecting a battery capacity meter to the balance lead, though of course its accuracy is not certain. The meter was disconnected whilst running as its consumption would not be metered by Consumpt. I also noted the battery voltages when stopped. The results were:

Zippy                                                                               Panther

% left	Consumpt	Meter %	Voltage	Meter %	Voltage	'Standard' data: lipo volt v %
100	0	100	12.60	100	12.62	12.60
75	550	74	11.82	74	11.85	12.07
50	1100	44	11.45	46	11.47	11.55
25	1650	14	11.15	21	11.20	11.03

Volts on chargers after test          11.23                              11.20

For the record the full-throttle current was 37A on the Zippy and 39A on the Panther. That was nearing the limit for the device and it was only a Tundra motor running on 3S with a 12x6 propellor.  However the person who wrote the OpenTx code seems to imply that the sensor will take higher currents.

The results are pretty consistent but show a lower %age than that calculated by Consumpt  especially towards the end of the 'flight'. I checked my battery meter against another to see if it might be reading incorrectly. It does give a lower % reading than my iSDT charger when I connect the batteries after a flight, though slighly higher that the second meter. We must remember that none of these devices are calibrated test instruments. They are much too cheap for that. They are probably 10% tolerance What the tests show is that this method of monitoring batteries is reasonably accurate, and will help prevent battery damage and dead-stick landings.

150A sensor

It is a little larger and heavier than the 40A but not much. It threads on to the red lead from the battery to the ESC. The arrow must point from the battery. Remember current is conventional current from + to -, rather than electron flow.

I'd love to know how the sensor works. It looks like a mains current clamp meter, but they only work on alternating currents as they measure induced current. From battery to ESC it is direct current. Perhaps it is a Hall Effect transducer which works with steady magnetic fields. If so why the coil? Or does the coil create a steady field that the varying direct current changes? If anyone knows please let me know on peter@peterscott.website.

Here are pictures of the sensor and an XT90 12 awg lead I made up for it.

                                              Sensor                                                                                                                 Lead

Reading out the state of the battery

There is more than one way, apart from the display on your telemetry screen. The simplest would be to read out the mAh used when energy reaches a low level. Set a Logical switch to switch on when mAh reaches 1500 in the above test case. That will switch on a Special function that reads out the value every 10 seconds. You would need to enter a suitable value into the Logical switch code line for each model's battery size.  Or you could simply read out the mAh every twenty seconds throughout the flight..

Peter Scott © 2019

Last edit 5 February 2019

For more information and full manual see www.peterscott.website under Flying.

rileyshimption.blogspot.com

Source: https://www.rclife.co.uk/News/FrSky-Current-sensors