Reptile S800 Sky Shadow V2 Mk2
The same day my original S800 became a 126mph lawn dart, I ordered the replacement airframe. In this page I will be covering the new build, including the changes I'm making and why as well as covering parts that were missed in the last build page. This page will grow as the model does, so stay tuned. As the last page was a bit of a monster, I'm going to put a contents section at the top so that people can jump straight to the appropriate section.
3D Printed Parts
Most of the 3D printed parts are the same as for the original S800 build. I have created a slightly different ESC cover with a lower profile and I will be 3D printing a hatch to allow easier fitment of 4S batteries.
I will have a link to the Fusion 360 model and, once completed, a link to the Thingiverse [Thing]. If there's no Fusion 360 link, it's not something I've designed. Please note that the files in the Fusion links are completely untested, so don't download them and print them. Once I have a proven, printed, and fitting design, I will post the Thingiverse link and feel free to print them as much as you like.
Building the Fuselage
This is going to be the main part of the build I'm going to gloss over as it's pretty much identical to the Mk1 build. The main differences is that I'm using different servos.
I have changed from the Banggood "tower pro" MG90S to the EMAX ES08MA II servos, which I'll also be running at 6v. The reason for this is one of the things that seemed to happen in the 126mph "#landing" of the original is that it didn't want to pull out of a dive. The entire flight was glitchy as I'd clearly changed something incorrectly in iNav, but I was also concerned that the servos may not have had enough grunt. Running at 6v will give the servos more torque, and should overcome problems with not being able to pull up.
Receiver 1/4 wave mod, and why you SHOULDN'T do it
Another thing that I'm going to mention is NOT to do the 1/4 wave mod on the R-XSR, or any antenna for that matter. The 1/4 wave of 2.4Ghz is 31.25mm, but that doesn't take into consideration the inductance of the material the signal is running through as well as other factors. There's also the fact that our TX and RX will be using frequency hopping in a range between 2.4Ghz and 2.5Ghz, so most antennae are tuned to be in the middle of this range. You should only tamper with your antennae if you have a frequency analyser in front of you, otherwise you are just going to enter a whole world of hurt.
In the last S800 build I suggested to do the mod. At the time I thought it was the right thing to do, and the video evidence seemed to prove that, however when they're not sure if a cat chewing off half an antenna would have any effect, I should have known better than to trust that video. I have now seen scientific evidence that shows categorically that just cutting the shielding to 31.25mm will not improve your distance, it may even reduce it slightly, the correct length of the exposed antenna will be closer to 28mm, depending on the impedance of the material used. For more information check out this video and this video. I apologise for the bad advice that I passed on.
Installing the Flight Controller
For this build I wanted to try something new, I wanted to give the Matek FCHUB system a go as it seems like a really good ideal. Matek have released the FCHUB-W PDB which is specifically made for iNav and fixed wing models, so that seemed to be the logical choice. It just seems like a good idea to me to separate the servos and ESC from the flight controller and have them on the PDB. Matek have also released the F405-WING which seems to be a combination of the 2 components I'm using in this build. While the combination will mean closer integration, there are a few things that make me think it's not the best solution. I will try one in the future, I'm sure, maybe the S1100 build. This is also acting as a test bed for my Mini Talon build where I'm planning on using the FCHUB-W PDB with the F722 flight controller. In this build, I will be setting the PDB to output 6V to the servos for more torque and speed.
There are a couple of things that could be better on the FCHUB-W, for example:
- They've included pads for connecting the VTx, but not the FPV camera. To me it would make sense to have them connected at the same point.
- They have included a UART (5) on the PDB, but why not make it UART 2 so that the receiver can be connected there?
To me, a the advantage of the FCHUB system is to have a flight controller that is quick and easy to remove, in which case it would make sense to put the pads for components the least likely to come out on the PDB.
What may not be obvious from this picture is that I actually installed the PDB backwards. Being new to the FCHUB system, I really should have tested out how the ribbon cable works. Installing the hub like this meant that when the flight controller was installed, the USB port was facing the back of the hatch. Of course, I discovered this after soldering in the servos and a bunch of other components. I did think about cutting the back of the hatch away to access the USB from the rear, but I thought better of it, got out the soldering iron, and did it properly.
Attached to the PDB I have the power and signal for the servos (running at 6v), the main power and signal for the ESC, and power to a few other components. I also added a 1000μF 35V low ESR capacitor between the main battery power connections.
The next step was to add the flight controller which meant soldering far fewer components than normal. The receiver is attached to the 4V5 pad (so that it will work with USB power), ground, SBUS to Rx2, with the SmartPort to Tx4 (for SoftSerial). The FPV camera and VTx signal cables are attached to the Cam and VTx pads respectively, they are both powered and grounded on the 10V and G pads of the PDB. The GPS is on UART 1 and the 5V and G pads next to them. The final connection on the FC is the signal for the LEDs, which goes to pad S5, again, powered by the PDB's 5V supply. The final step is to attach the ribbon cable and test.
Since building the S800, I've made a couple of changes to the spec. The first major change was the motor and ESC. The DYS/HobbyKing combo was removed and replaced with an EMAX LS2207 1900kV with a Racerstar Tattoo+ 42A ESC. I wanted to change the motor to get more efficiency as the DYS was going through a 4000mAh 3S pack in about 10 minutes. The LS2207 has pretty much doubled flight time.
- Cruise throttle has improved greatly, from around 10A draw to around 4.5A at 50% throttle.
- Max amps hasn't changed much, both peak at around 32A.
- Similar top speeds, 98mph compared to now 96mph.
- By the feel of it, the thrust is very comparable too.
- Old motor came down hot, new motor comes down normal with the same kinds of flight.
The ESC was changed as I was having trouble with noise in the FPV signal. This was the ESC that came out of the crashed original S800, so swapped it over for the Racerstar, which I run with the Multishot protocol (until iNav get round to implementing BLHeli_32 and DShot).
The next change I have only just done, and that's to swap the receiver from the X4R-SB to the L9R for a better signal. I want to get closer to and between trees, and was having RSSI issues, which hopefully the L9R will solve. The L9R was going to be in the Mini Talon build, but that will now be having Crossfire.
These changes have been a breeze to do, thanks to the FCHUB system that Matek have. Swapping the ESC was a breeze, as was the receiver. Think I'll be sticking with FCHUB and Matek for the foreseeable future.
Another change is that I've swapped the 7×4 APC propeller for a 7×5 APC. I've noticed slightly better cruise amps and a higher top speed, back to 98mph, which is where I was at with the old motor, only I'm still getting 20 minute plus flight times.