Jeti have introduced us to 900MHz radio link as backup to 2.4GHz, let’s find out what it is about.
Is simple, to provide a backup radio-link to existing 2.4GHz frequency with additional 900MHz link. Well, technically it’s not 900MHz, in europe it is 863 – 870MHz and in US 902 – 928MHz.
Reason is also very simple, 2.4GHz band is very crowded and there are areas and flying sites where “air is too busy” for any longer distances. There’s a lot of discussions in teh interwebs about this so I’ll skip this part.
How Jeti made it
Jeti have chosen a quite clever way. Since they have long had the possibility to use satellites for both normal receivers with SAT-port and centralboxes basically all they had to do is the hardware. This is done with third transceiver in transmitter.
You can see the transceivers in the picture below, two shielded units on top are 2.4GHz and the one below is the 900MHz unit.
The idea with lower frequency is pretty clever. We do know that lower frequencies have better penetration-ability through obstacles so this makes the lower frequency very suitable to backup-use. My very basic range tests supports this, the 900MHz does go a bit further, of course some thorough testing still needs to be done on the subject.
The receiver is the size as R3, that means four outputs. Antenna cable is 28 cm long.
Antennas are of different shape than we are used to. There is two of them and measurements are 11 mm x 80 mm x 0.7 mm. Two-sided tape-strips to fasten them are included. The actual “wiring” is clearly visible.
I’m sure we will see some very different pictures on antenna placement with this receiver in future.
Looking at tech-specs between 2.4GHz and 900MHz we know that 2.4GHz is sending with 100mW, I did not find information on the 900MHz value. Comparing the receivers sensitivities for example R3 has -106dBm and RSat900 has -110dBm, receivers power outputs are 15dBm for the R3 and the RSat900 being with 14dBm. R9 has the values as R3. (These values are the same in US and outside US.)
Setting it up
I tested with an R9 receiver with firmware 3.24 Standard and the RSat900 with original firmware. Connection is straight from the manual:
I had issues setting this up in the transmitter, reason was totally myself, never used a satellite-setup before so some manual reading and Wayne from RC-Groups forums to the rescue, thanks Wayne!
First, we need to set the R9 receivers serial link to something else than PPM, so this is wrong:
I chose Jetibox here. After that we need to go to Wireless Modes/Trainer and enable 900MHz Backup and pair the receiver:
After that our device explorer should show both receivers:
Setup is basically easy, RSat900 is configured to PPM Output and R9 has the SAT-port in PPM Input-mode, here’s RSat900:
And here’s R9:
What we need also to make sure is that PPM settings match, in my case both are on PPM Positive, 8 channels and 20ms output period.
Thoughts, testing and results
Now, before testing we need to ponder on some things. First, latency. Different radio circuits on different frequencies, time it takes for the equipment to change the signal path, there has to be at least some time-issues here. And surely so, there is. I hooked a few of R9 receivers servo outputs to my oscilloscope, put the servo test mode on and watched the magically moving PWM-signals moving gracefully between 1.000 and 2.000ms width. A nerd get’s his kicks where he can find them, I know…
So, on with the record-button on scope and off to shut down 2.4GHz. Jeti gives us the opportunity for this in the Wireless Modes/Training screen’s lower left corner, we can shut down specific radio-links:
Choosing the last item, leaving only 900MHz on I did get some results. And I was impressed. The gap between movement on servo PWM-signal was only 700ns, that is 0.0000007 seconds. (Average on 10 measurements). What this practically means is that in case of signal dropping to backup frequency one does not lose the model due strange control surface movements.
When 2.4Ghz comes back the gap is shorter, I found it to be around 100 to 200ns.
There is another thing. There is a gap, and if right in the middle of the gap servo is moving it stops for the length of the gap and after the time it jumps to the correct position. I can imagine for a heli in tight spot (a lot of rapid movements, hard 3D for example) this will be noticeable. But, the model is saved so the main goal truly is there.
No quick test is perfect without some questions remaining.
For example, test is made with intentional shut down of 2.4GHz signal. Is there any delay in true life event where transmitter first has to notice the 2.4GHz link getting worse? To test this I tried to isolate R9 antennas leaving RSat900 antenna out but did not get it to work. (Too small house…) Also to test this I need a friend operating scope due limited record-time. I’ll try to get this sorted out.
I need to read on the subject. As I have helis and all of them are using sum-signals (PPM/UDI etc) I need to find out if I can use the RSat900 to backup those, if we can only backup pure PWM (servo) there is not much use of this for me.
Way to go Jeti, this was absolutely great!