Fly Electric!

2.4GHz Lockout Logger

This page describes a tool to assess reception of 2.4GHz signals with airborne measurements. Pretty much all 2.4 radios work. However, some will cope with difficult situations better than others and this tool can be used to assess these with a bit more data.

The tool is programmed when to expect signals and then records when they are not received. These situations are commonly called '2.4 lockouts' because the receiver goes into 'hold' or failsafe and the pilot has no control. Altitude, the number and accumulated duration of lockouts are recorded every second.

Reception of 2 receivers is monitored in all tests. Assan receivers have been used (X8R6 and X8R7 with short aerials). These only have one real aerial so are good for testing diversity. They can be put in 'difficult' models and situations to assess performance. The results will generally be relevant for all makes of 2.4 radio.

Conclusions: Both Transmitter and Receiver aerial orientation is important. So too is multi-path/diversity on the receiver(s).


FULL POWER TESTS (100mW)


Flight 5
Tx aerial 'side on'
Lets start with some good flights and simple results. Altitude is shown on the left axis and lockout duration (in ms) on the right axis. Two 2.4GHz receivers were used with vertical and horizontal orientation. Both had standard aerials in parallel as they come. The angle of climb is usually similar to the shape of the graph. 'Flight 5' (above) shows excellent reception. The Tx was stationary on a plastic chair with its aerial 'side on' to the flight path.


Flight A4
Tx aerial 'side on'

Flight A3
Tx aerial 'end on'
In the next two tests, a friend held the Tx and kept it pointing at the model. In Flight A4 the Tx aerial was always 'side-on' to the model (best possible conditions). In Flight A3 the Tx aerial was 'end-on' to the model (weakest signal for the model). The difference between these flights is very obvious. The Assan radio demonstrates good ability to maintain a high quality signal when the Tx aerial is aligned well. The last part of flights A3 and A4 were aerobatics at short-range.


Flight A12
Total lockout time

Flight A12
Average lockout time
These graphs show one flight to 1440 feet (440m) (which is very difficult with a 4' wingspan model). The Tx aerial was horizontal and the Tx was on a chair but not pointing directly at the flight so there is greater signal loss. The Vertical receiver experienced 364 lockouts averaging 20ms each (3% of flight time). The Horizontal receiver had 638 lockouts averaging 21ms (6%).

The left graph shows total lockout time in each second. This is quite high during the climb on the Horizontal receiver because the climb was steep. The Vertical receiver is weaker on the descent due to it being a fairly flat decending circle. The right graph shows an average for each lockout. This shows that lockouts are typically very brief so control will be maintained. OK, but diversity would be nice for when signals become weaker.


REDUCED POWER TESTS (1mW)

To be more 'extreme' with the testing, my friend James has lent me a -20dB attenuator. This reduces transmitted power by a factor of 100 and theoretical range by a factor of 10. These tests hopefully exagerate the simulation of congested airspace and the infrequent but 'very difficult' conditions we may experience.


-20dB attenuator
.

Flight 4
.

Flight 4
Orientation overlay
The graph on the left shows a descent by way of a mild spin followed by continuous rolling at low altitude. These are clearly difficult for individual receivers. The part of the whole flight with the worst reception extends from the climb from 700' to the descent to 200'. The Vertical Rx had no control for 37% of the time and the Horizontal 32% (with reduced power, remember). However, sorting and overlaying this part of the flight (right graph) reveals how the two antennae orientated at 90' to each other compensates for the other's deficiencies. So, diversity is important for robust control.


ASSAN - 2 vs 1 AERIAL (at 1mW)

The Assan 'mini series' receivers have two aerials. One is active and the other grounded. These tests attempt to assess whether the grounded aerial serves any purpose (this is an Assan-specific set of tests). Results vary so my view it makes no difference.

Both Rx Horizontal
Both Rx Vertical
Vertical/Horizontal
Standard
(2 aerials on each Rx)

Flight A7
Standard
7/8% loss
24/26ms ave

Flight A8
Standard
16/24% loss
36/43ms ave

Flight A17
Standard
31/27% loss
37/32ms ave
No Ground
(1 aerial on each Rx)

Flight A10
No Ground
16/19% loss
36/36ms ave

Flight A9
No Ground
7/11% loss
19/28ms ave

Flight A18
No Ground
30/32% loss
33/47ms ave

ASSAN - PERPENDICULAR VS 'STRAIGHT' AERIALS (at 1mW)

In this next set of tests, the active aerial was soldered perpendicular to the PCB. Reception improved significantly in the first pair of tests (A5/A6 done together) but was worse under different conditions the next day (A17/A18/A20/A21 done together). Not conclusive yet.


Assan Rx
Pependicular
.

Flight A6
Standard
25/35% loss

Flight A5
Pependicular
17/17% loss

Flight A20
Pependicular
42/35% loss

Flight A21
Pependicular
33/32% loss

THE LOGGER EXPLAINED

The logger comprises a PIC 16F88, MPXA6115A pressure sensor and micro SD card. The Assan receivers have an nRF2401 (or similar) RF chip whose 'DR1' pin goes high when valid data is received. This is monitored by the PIC. Assan transmit data every 4.5ms. If data is not received within 7ms (halfway between 4.5 and 9ms) data loss has occured (a 'lockout'). Data is saved every second so lockouts can be any duration up to that point after which the timer is reset. One lockout is recorded for each continuous loss of data within each second (not one per 4.5ms).

The PIC needs a connection to the RF chip and the receiver's Ground. For convenience, I disconnected the 'signal' from the receiver's MCU on the 7th set of servo pins by cutting the traces (one top and bottom). These are easy to repair later. The 5v regulator on the logger can operate two receivers but is not suitable for operating any servos. Other than the modified set of pins, the receiver can continue to operate as normal (with or without the logger).

The 16F88 is the smallest PIC with 'self-write' ROM. This is needed to accummulate 512bytes of data between writes to the SD card. Writes to ROM occur every second and to the SD card about every 30 seconds. These can can take several milliseconds so lockouts are not monitored during those writes (ie: lockouts are conservative).

A text file (.txt) is created automatically on the SD card on startup (ie: separate file per flight). csv files are possible too. Files are currently 64kb but can be bigger or smaller. This allows up to about 1 hour logging (per file). The file name is numeric and increments sequentially. The file can be opened in Pocket Word on a PDA so can be reviewed after every flight at the field. It can also be opened directly into Excel or any text editor on a PC. Altitude and number & duration of lockouts for both receivers is tab (or comma) delimited ASCII characters (so is correctly formatted into columns and needs no further conversion).

I am happy to release the source for the lockout code (written in C for the BoostC compiler) for personal non-commercial use (but not the ROM/SD parts). I may release full hex for people who want to build their own, and I can build working loggers for sale.


More information on 2.4GHz can be found here:
* Handheld 2.4GHz scanner
* Overview of 2.4GHz
* 2.4GHz diversity tests
* My 2.4 GHz Receivers

Top of page
Click on the links above for more info.
Click on the links below for other pages.

Home Art of the Possible Absence of Matter Links Updates Email
SITE MAP