Introduction: RC Plane: FPV & INAV on a Z-84 Wing
What can you create merging one of the most appreciated flying wing frame, a proper electronic setup and a well set autopilot software? You'll have a nice and funny "toy" that can amuse different pilots with different skills.
This project started when i decided to put aside for a while other random ideas about rc planes, to dedicate some effort to design and build a model with some specs in mind. No custom, hotwired wing profile, no strange motor configuration and so on...
I've listed these design goals:
- The plane must "fly well" while being reasonably cheap, around 100€, the less the better. (well, purchase from china, without the transmitter-receiver).
- The plane must have an FPV system, with an OSD device that's easy to set up.
- The plane must being able to carry an HD action cam (foxeer legend, runcam 2, something light, with a low profile).
- The plane must have a reliable flight controller (not only a gyro stabilization). I don't aim to do autonomous waypoint missions (nevertheless the hardware would be ok), what i want is:
- Autolaunch feature, for no-brain takeoff and no need of additional tools like catapults or bungees (or a buddy that launches the plane for us).
- Autonomous flight mode, like "cruise" and "RTH".
- Reliable failsafe behavior in the event of video/radio signal lost.
- Battery voltage and current monitoring for additional safety.
You know, wouln't be nice to have a small and cheap rc plane, that you can always keep in the car's trunk to use it whenever and wherever you want? A plane that's also funny, with nice advanced features that can save you a lot of hassle, and with an HD cam to record your best places and moments? Well, i'm really sure it would, so i wrote this tutorial!
Teachers! Did you use this instructable in your classroom?
Add a Teacher Note to share how you incorporated it into your lesson.
Step 1: Choice of the Frame
One of the best plane i've ever had (and also heard of) is the Mini Talon. Nevertheless i didin't choose the mini talon for this project because it would't be an easy model to build and, more important, it wouldn't be a cheap model at all!
Another reasonable candidate could be the nano Skyhunter, which is a small and funny plane. Unluckily this frame is affected by some aerodynamic issues (it can start unwanted flat spins after tip stall and, without rudder, is not easy to save the plane) and for this reason it's not recommended for anyone.
Someone might like the traditional airplane design, with a fuselage, a couple of wing in the middle, with rudder and elevon behind, and a pulling motor on the front. Well, i really don't like to have pulling motors on these kind of planes (fpv). That's not beacuse the propeller disc will likely be in the middle of the recorded frame, that could be even nice. The issue is that the motor (and the propeller) will be more exposed during landing and, even more important, sometime something goes wrong. Remeber that there's always a chance to loose control, for whatever reason, and hit something or someone. With a pushing motor the chances to damage something or hurt someone are way less.
Another candidate could be a small delta wing frame, more or less as big as the nano skyhunter (80cm or 31in).
I have some of them with different setups: with or without the flight controller and FPV gears, with small or big motors and i have to say that this concept has several nice features.
First of all: they are tough and and crash resistant, you can easy glue them back into a fully working status several times. I've had full throttle crashes that required a bag to collect all the pieces spreaded on the field, but with a lot of glue i always managed to fly them again.
My final choice is a Wing Wing Z-84 frame, which is a cheap but really appreciated frame, with a nice and areodynamic shape with an efficient airfoil. I've considered also the Eachine Blackwing, but the Z-84 took the lead because of it's larger wingspan (then less wing load). Moreover, a guy called Mark Qvale also appreciates it, and you can believe that if he's happy with a frame then it's a good frame.
I suggest to buy the plane in kit version and the other specific components, separately. I don't trust the PNP planes beacuse of both the assembly quality and the components quality. I'm not to saying that every PNP plane is crap, it's just that i'm not sure that every plane is well assembled with good components! Moreover i still remember when i've put additional eletronics on my PNP nano skyhunter, it would have been way easier to do than with a kit -unglued- plane. Well, i have to admit that with this kind frame it's not a big deal to choose a PNP beacuse everything will be placed in the canopy.
For the records, the photo shows what you will receive with the kit version of the plane. Now we have to add some other components.
Step 2: Base Components
In order to fly you'll need other components: a motor (MT), an electronic speed controller (ESC), a battery (BT), two servos and a transmitter-receiver (TX-RX) system.
If you want to fly FPV you'll need also a camera (CAM), a video transmitter (VTX) and an antenna.
The last feature, the autonomous flight, requires a flight controller (FC) and a GPS module.
In the photo you can see what i used for my plane (i know, i forgot the camera).
MOTOR (MT): in order to keep the price low i've decided to use a quadcopter motor. It would be good enough and i also like to have the threaded motor shaft that doesn't need anything else to fix the propeller (the "airplane motors" usually have a smooth shaft that requires a prop adapter/holder). A good choice is the common 2204 2300KV motor. A 2205 or a 2206 motor could also work, but they would be more heavy and expensive for no particular reason. You could/would try to have less KV, in order to use bigger props without overloading the ESC or the battery. Bigger and slower props are more efficient, but you have to deal with your frame available space.
ESC: this component takes power from the battery and gives it to the motor, depending on the received signal. There's no reason to go big here, 12Amps are enough to fly without troubles, 30Amps allows unlimited vertical flight. I'll keep a nice balance, 20A ESC can provide plenty of power also for loops and other acrobatic manouvres. I'll test also DIY batteries made with 18650 li-ion cells, and that's another reason why i don't want to foresee a current consumption higher than 20A.
One important ESC's feature is it's BEC, battery eliminator circuit. It's basically a voltage regulator that can provide the right voltage to the RX, the servos and the other electronic. BEC is relevant because most of the expert buidlers think that servos should be powered by a dedicated circuit, which must not provide power to the FC and/or the RX.
You know, shit happens, and if there's something wrong with a servo it usually drains a lot of current (brownout).
This amount of current could easily be more than the amount that can be provided by the BEC, and therefore the bad servo can "steal" power to the other components. If you have the FC on the same power rails it could be forced to reset due low voltage, and therefore disarm the model mid air and a lot of bad bad things will happen. On the other hand, if you have a separated BEC, the brownout could still force you into an emergency landing, or maybe a crash, or maybe you can recover the plane, who knows? The idea is that in any case the OSD, the GPS, the telemetry and the buzzer will remain online, no matter what. That's clear how big this is in case you have to search your plane in the field, meters and meters away in an unknown place.
Anyway, since this plane has only two servo and since the suggested FC has its own (quite big) BEC, chances are not that bad. You'll gain few grams choosing "opto" ESC, without BEC, but imho is not worth the risk. I'll go with BEC ESC, i'll find other safe ways to safe few grams.
PROPELLER: the best compromise i've found so far is a Dalprop 5046C three blades. A two blades 6040 or 6030 can give you more efficiency but they will run very close to the frame (see the following image) and this make noise. It's not "taking off airbus" loud but it's definitely loud and i'd rather keep a low profile when i'm flying around places like golf club and other quiet places. Whatever, it's just a personal choice. You can't further increase the diameter to 7" or more, you'll hit the frame. You can reduce the prop diameter but you'll lose efficiency. A lot of efficiency. Anyway, any prop ranging from 5050 to 4045 should get you airborne. If you only have quad prop at hand you could still use them.
I haven't built a motor/esc/propeller test stand yet, it's on the schedule but for now you should trust my "feeling" of the full throttle thrust compared to its power consumption (i have a wattmeter, no guessing here) with different propeller. I have these props that are reasonably sized for this model [diameter][pitch]-[blades #]:
- 5056-3 -> too much current
- 5050-3 -> too much current
- 5046-3 -> my choice
- 5042-3 -> ok, but you can use more power
- 4045-3 -> little thrust and not efficient
- 6050-2 -> too much current
- 6040E-2 -> ok, but noisy
- 5030-2 -> ok, but you can use more power
BATTERY: Any battery that can provide 20-25A is ok, the capacity should range from 1000 to 1800mAh. I guess that a 1300mAh 20C 3S li-po battery would be the best choice, price/performance wise. I'll test also a 4S li-ion battery made from 18650 cells, i'll discuss that later.
SERVO MOTORS: these small components will move the ailerons, so they will control the plane's flight path. As you can imagine this is not the best component to save money with, giving up quality. If you have an issue with the esc or the motor you still can try to land doing a controlled glide, if you have a servo that doesn't respond the plane will be out of control. Good news: you need only two of them, so quality servos are not that big chunk of the budget. Any good digital servo with metal gears is a proper choice, the right size is 9g servo.
RADIO TRANSMITTER/RECEIVER (TX/RX): I won't talk a lot about this combo because the are mainly two options. Option one: you already have one, so you already know what i'm talking about and how to adapt this project to your radio. Option two: you don't have a radio and then you should just stick with the suggested one. A nice choice to begin with is the Flysky FS-I6 paired with an FS-X6B. These cheap radios have a lot of modding potential, you can see my other tutorials, so you can improve the performance and can add a lot of features just for few bucks. Anyway, with this receiver you can easily send the plane more than 1.5Km away, which isn't bad for a casual flight. Are other radios better? Yes, but the price is several times higher so there's no point in this. In the end the only requirement is to have a receiver that can output a serial bus (i-bus for the Flysky, but also sbus and others can work as well) or at least a PPM output. Serial bus is recommended!
CAMERA (CAM): this choice is really subjective and it could have a notable impact on your wallet. You "can" fly with cameras that range from 5 to 100 bucks and here the performance are strongly bonded to the price. Some reviews say that without the best one you can't fly. That's false, you actually CAN FLY, the comfort is worse, the colors might be faded and/or not real but that's still good enough. My choice is a Foxeer HS1177-V2, which is a well tested and appreciated FPV camera.
VIDEO TRANSMITTER (VTX): different places have different rules, check yours. My choice is a cheap but good enough Eeachne TX526. It works with in the usual 5.8GHz range, it can be set to different transmission power and it has one of the smallest footprint.
VIDEO ANTENNA: You have two main choices: linear and circular polarization. The stock antennas, the straight "router like" ones, are linear polarization antennas, and this is the best choice if you want the farthest video range but it requires your plane to fly mostly flat. The suggested "mushroom like" antennas use the circular polarization, and these ones are the best choice if you have lively flight. Linear polarization requires to have the two antenna pretty much aligned, circular polarization doesn't care. I want to do loops and tight turn (with high bank angle), therefore my choice is to use circular polarization antennas. It's very important to have the same polarization on the transmitter and the receiver, and pay attention that the circular polarization can be "left or right" polarization, don't mix them. Long story short: use the same antenna on the transmitter and on the receiver, you won't make mistake.
VIDEO RECEIVER (VRX): I can't tell you what to use in this tutorial beceause i'd need several pages only to cover the main options. Also, the price range is quite wide and you could easily invest more money just to buy FPV goggles than to buy all the other components of this tutorial. Use an FPV monitor, or FPV goggles, with or without diversity, the choice is your. I'll use cheap FPV goggles, you know, they will sleep in my car therefore nothing expensive. (if you really want a suggestion write a comment)
FLIGHT CONTROLLER (FC): there are a lot of different models, you can have the full-optional and expensive ones or you can have the basic ones but you would need additional components if you want some advanced features. My choice is the Omnibus F4 Pro V3, that's a good mix of embedded features for a reasonable price. First of all it's an F4 and not an F3 CPU, which is enough for everything so far. It has an embedded OSD chip that can be easily set up directly from the main configurator. It has a 3A 5V BEC, enough to easily power every other device. It has an embedded voltage sensor (common) and current sensor (not so common) . With three physical serial ports you can connect a lot of additional device like GPS, telemetry, serial bus RX input... You can save some bucks buying an SP F3 board or a Pico Blk board, but then you'd need to buy and set up an external OSD module, an external current sensor, the software can't reach the same cycle speed... Not to mention the additional effort to configure external components with their "external" software, the additional weight and time requirement for the additional wiring. Also, the external components are not for free so if you don't want to give up some features you shouldn't buy cheaper FC for numbers of reasons.
You can buy way more expensive FC, they are undoubtedly better, but they won't give you much more for this kind of project and they usually are bigger and heavier.
There's more: the suggested FC can also work with Arduplane! Arduplane is a more mature but more complex firmware, it won't be discussed in this tutorial but it's nice to have also this option for the future. You never can tell...
GPS: it's required for autonomous flights, we'll use it mostly to have our ass covered, and have a plane that can return to home in case of video/radio signal loss. The most common modules have also an embedded compass but that's for the copters. Since our plane must have at least a certain speed to fly, it's always easy to define its direction just by coordinate subtraction. The best choice is a BN-180 or BN-220 module, i've tested both and they are very similar, except for the fact that the 220 can update the firwmare. What this updated firmware can give to you? I don't know and i don't care about it, it works good enough with the stock one. I have a BS-880 on my mini talon, with integrated compass, but i don't use its compass for the same reason, it's not even wired. Actually, the BS-880 is faster to achieve GPS lock and it can see several satellites even in my basement (i'm not kidding!). It's worth to spend more money for the 880? Meh... Since i'm still unable to fly in my basement and since the 180/220 only needs a little more time to get the GPS lock, i'd go with the BN-180/220 module, wich is smaller and lighter (that's good!).
Edit: the GPS module i've ordered from China has have arrived faulty (dead), no big deal, they'll send me another one but i have to wait the delivery. Since i don't want to wait i'll use a spare BS-880, but that's not relevant for the setup.
Step 3: Optional Components
The project's premises are already nice, but with these components you can upgrade a good FPV plane into an awesome FPV plane.
1) I'll do some test with li-ion cells, my plan is to increase the flight time by using a 4S, 2600mAh, 7C, DIY battery. As i've already wrote in some comments on this very site: i don't trust people that claims that they can fly a drone with laptop reclaimed cells. I'll use 20A fully capable brand new cells, wich are usually used in vaping cigarettes, and this is barely "good enough" to fly an RC plane. A good laptop battery, even in pristine condition, can't provide more than 4-5A. They can't even manage the takeoff power consumption of a small and efficient plane, that's why i call a complete b.s. the claims that you can lift a quadcopter with them. Anyway, there are plenty of tutorial about DIY 18650 battery, just be sure to purchase cells from a reputable supplier. Edit: the idea is good but 4S is too heavy, you need 3S battery. Read last chapter...
2) I'll cover every surface with laminating film! It's not a difficult task, it's only a time consuming task... The process to put heat-shrinkable adhesive thin plastic everywhere on the model needs patience, but you'll end with a stronger, more rigid, more efficient and better looking plane. Anyway, i can imagine that only few people would invest 2-3 hours to cover everything. If you don't laminate the whole model you should put a stripe of tape on the leading edge of the wings. That's the most exposed part and is worth to spend few minutes to protect it. If you do laminate invest some money in a proper laminating iron, with such curved and small profiles a standard clothes iron is way too big and heavy, the small details will be a pain in the a.
3) Since i'll use 4S batteries i have to expect higher speed, higher speed means higher stress on the wings. I've decided to replace the stock glass fiber spar with a carbon fiber spar, wich is more rigid (and lighter, but that's not really relevant). I don't fear that the plane would broke in half in a tight turn, it won't, i just want to make it more rigid. The carbon spar (and the laminated wings!) will significantly increase the plane's rigidity and therefore the plane's handling at high speed and during high G maneuvres.
Step 4: Frame Assembly
Now it's time to glue something. The assembly is quite straightforward, there are no chances to glue something in the wrong place, but you have to use the proper glue. Generic "super glue" or cyanoacrilate glues are not recommended, you need a foam specific glue. The UHU POR is ideal, but also other foam glues are ok. One hint: glue the stabilizer fins at the end of the build, you'll have more room to move during the electrical wiring.
To glue the servos you could use hot glue, just few drops. I'm worried about the underwing servo linkage, which exposes them to shocks when you land on a not-flat surfaces (like every field), so i'd rather use a not-so-strong glue in case i have to change them. That's the price of better airfoil efficiency (due to the clean upper surface).
Take care to install the servo arm at 90° respect to the servo's case, when the servo's shaft is in the middle of its span. The servo arm should be able to move simmetrically in both direction, starting from 90°.
To mount the motor i've used a custom 3D printed part (thanks Mark), which is better than the stock one because it put the motor further back allowing bigger prop clearance (and slightly heavier batteries). It's ok to have the motor not horizontal when the frame is on the table, it will be leveled during flight.
The FC uses four nylon stand off, glued to the bottom of the fuselage. The ideal spot for the FC is exactly the CG, in this case it would be right under the carbon spar and it would be really uncomfortable. I've placed it right before the carbon spar.
The GPS, the VTX, the RX and the motor should be placed the as far as possible, especially respect to the motor, in orderto minimize the interference. I don't plan to do long range with this model, so this is not mandatory, and the drawback of installing components at the end of the wings is an increased inertia, which is bad for a short range fun wing. I've found a good compromise that doesn't bother me with a lot of long wiring but still keep the components quite apart.
Since i've laminated every surface i had to cut slots in the foam for the additional components. That's not mandatory if you didn't laminate the surfaces, but it's still highly recommended because it preserves the awesome plane's aerodynamics. The only component that shouldn't be cover is the VTX because it needs to be cooled by fresh air, i've attached it with strong and thick double side tape.
In the photos you can also see a telemetry module, i'll use it to do some tests with a ground station but i don't want to increase the complexity of this tutorial, just ignore it.
I put the cam in the nose (obviously) and i cut a dedicated slot in the fuselage foam. Please note that if you want the cam pointing straight forward during flight it should be aligned with the motor and not the table.
I've cut the top fin on the motor cowl, it doesn't do anything except make noise and reduce the air flux on the propeller (Mark's hint, and i totally agree).
Since i've laminate everything, also the fuselage belly, the air intakes under the fuselage are now sealed. The ESC and the motor have to be cooled so i've cut a small rectangular hole in the red canopy, right over the camera, to have a fresh air inlet. I suggest to do the same belly sealing, even if you don't laminate everything, because the air intakes under the fuselage will likely collect dirt, and if you have dirt inside the canopy it can exit only through the motor, and that's not good. Don't worry, you won't miss the grips to launch the plane, you'll be assisted by the autolaunch and the "frisbee launch" will work like a charm.
There's more: it's way more aredynamic and you really have to agree with me that its smooth and flat belly looks AWESOME! ^_^
The mechanical assembly is done, just paint it as you like and proceed with the wiring.
PS: spray paint IS HEAVY! Use as many as you need to make your model visibile in the sky but try to avoid a full body paint, and avoid multiple layers.
Step 5: Electrical Wiring
You'll need some tools, my basic DIY electronic kit is collected here.
First of all, sooner or later the FC must be calibrated, and it's easier to do that moving the PCB instead of moving the whole fuselage. So, if the FC is not calibrated yet, jump to the next step (software), calibrate the FC and come back.
Now heat up your soldering iron and use a drop of tin to bridge two pads under the FC, look at the photo. You have to connect the pad "RAM" with the pad "VCC", this will provide battery (unregulated) voltage to the VTX and the CAM. If you don't use my suggested components be sure that yours can accept 3-4S lipo voltage. In case they work with 5V, bridge the RAM pad with the 5V pad instead.
The other components have to be connected as shown in the attached wiring diagram. It's not difficult, just do one step at time and double check polarity and the welding quality. The most annoyng thing is to split 5V and GND cables (to provide power to bot the GPS and the RX), and to solder the negative/black wires of both the esc and the battery connector to the same GND pad (big and squared under the FC). This could be somehow difficult, use abundant soldering flux and set the iron on max power.
Hint: if you feel confident with the soldering iron you should remove the connectors in most of the cables and proceed soldering every wire right in place. I've done this for pretty much everything (except for the servos, i still want to being able to change them quickly) because of two reasons. One, you'll save few grams here and there. Two, a connector might get loose while its appearance is still ok, and you really don't want hidden issues in an rc plane. Try to keep the wires just as long as needed, it will also help with the weight reduction and the cable management.
Be careful and double check all connections, a crash could be fixed with some glue, a smoked component has usually to be trashed. Don't use a lipo to power it up the first time! You should use a power supply that can't provide a lot of amps, i've used a 12V 1A power supply (you can see the connector adapter in one of the previous step photos). Another good choice is the portable variable power supply that i've described here.
Step 6: Software Setup
Ok, the first FC setup could be overwhelming because of the huge number of new information but, with a little patience and doing one little step at time we will eventually complete the task.
First step: download and install the firmware (INav) configurator. I have Win 10, try to use the same OS to minimize the chances to get stuck somewhere. I'm using version 2.1.4, if you find something strange and you are using a subsequent version just revert back and install this specific version.
Now the worst part: the driver's installation. Sometimes is just a matter of plug the USB and let the OS do the trick, other times you need to use colorful words and a lot of profanity. The best Win10 strategy is to use this free software: Impulse RC driver fixer. Plug in the USB, run it as administrator and finger crossed.
If you open the INav configurator and in the top right field (look at the images, the "driver" rectangle) you see a new COM port, and you can connect to it, you are fine. Maybe the configurator software will tell you that your firmware version is not compatible but that's still good, we'll update the firmware anyway.
Now, unplug the USB, press the "boot" button on the FC (is near the buzzer) and keep it pressed while you re-plug the usb in. Now in the field "driver" you should show "DFU". If you don't see DFU try to run again the driver fixer. If you don't see "DFU" but can connect to the COM port, you can go to the CLI tab (read further), type "dfu" in the input bar and then press enter. The FC should reboot in DFU mode.
Without this DFU option in the upper right corner you can't proceed, you have to find a solution before trying to proceed further. I can't write a tutorial about all the possible driver issues on every OS, you have to google that by yourself.
When you see DFU, happy days, the worst part is already done! Proceed by following the numbers:
- click button Nr 1 to open the right tab,
- select the right option in the drop down menù: "choose a board" -> "OMNIBUS F4 PRO", "choose a Firmware version" -> "2.0.1 - OMNIBUSF4PRO ..."
- click button Nr 4 and wait, you need internet connection because it will download the chosed firmware,
- click button Nr 5 and you should see, after few messages, the message "firmware flashed".
Now the FC should reboot, and instead of DFU you should see a COM port again, if it doesn't appear unplug the USB and retry. Click "Connect" and you should go online on the FC.
Now you have to calibrate the accelerometers, go on the tab "Calibration" and follow the procedure. It's really easy, you have to hold the FC in 6 different angles and press the calibrate button. Now the board is calibrated and you can go back to take care of the wiring.
Now it's time to set up the new firmware, there's a huge number of parameters and it's impossible to explain everything in a single tutorial. The firmware developers explain (more or less) everything in their wiki, but for this project we'll exploit the CLI (command line interface). Don't worry, it's nothing more than a copy&paste task, and it's way easier to have the job done this way than following a to-do list made by dozens steps distributed on several pages.
You can donwload the full configuration file here, scroll down to the end of the page or search for "CLI OMNIBUS F4 PRO INAV 2.0.1", it's config file dowload link.
The file can be used as it is. You might want to change only one parameter, the airplane name, which is shown also in the OSD. Go to line 840 or search the string "set name", the current value is "Z-84 UBHF.EU", you can use whatever you want, just modify the string (don't use special characters).
Now select everything in the txt (ctrl+a on most text editor) and then copy it (ctrl+c).
Open the CLI tab in the INav configurator (look at the screenshot) and click on the input field (were it's written "save" in the image) and just paste (ctrl+p) the text, now press "enter" and it's done. The last string should be "save" and when you click enter the values will be stored in the FC, and the board itself will reboot (and disconnect).
Now you only have to check that everything was done properly, reconnect to the FC, go to the "receiver" tab and check that the different channels behave as supposed (bind the receiver! just press the right button, plenty of tutorial online). The GPS tab might not show a GPS lock if you are inside a building, but at least it must not show errors. The OSD tab should look like the one in the last step video, but feel free to modify the configuration in the OSD page (by drag and drop). Modify the layout as you wish, it's not critical for the flight. If you want to change the switch assignment on your transmitter use the "modes" tab.
The default behavior is:
- switch A (CH5) -> arm the plane
- switch B (CH6) -> force manual mode
- switch C (CH7) -> altitude hold mode / horizon mode / RTH mode
- switch D (CH8) -> failsafe killswitch
- var A (CH9) -> up = autotrim, left = autotune
- var B (CH10) -> down = buzzer and failsafe
The autolaunch is always active so you only have to arm the plane, increase the throttle, and launch it. If you want to disable the assisted launch you only have to arm the plane and move the roll or pitch stick.
That's all, now you have a nice plane to have fun with!
Step 7: MAIDEN FLIGHT & UPDATES:
Well, i've underestimated the li-ion weight. The 4S 18650 battery weights too much and i'd need to place it over the FC to have the right CG (which is impossible). Actually it's the 4th cell that gives extra weight issue, a 3S battery would be perfect. Unuckily i don't have spare 18650 cells at hand to build a 3S battery and this 4S battery is also used with other models and i don't want to unmount it. Another option, you could add 40-50g at the wing tips (which are behind the CG) to balance the cell's wheight. Now the issue is that you've also increased the roll intertia by a lot (not to mention the relevant ballast weight) and that's not desiderable in a fun wing. It's simply not worth it, it will cripple any positive effect that li-ion would give.
Beside that i'm very happy with this model, in the video you can see the maiden flight recorded from the DVR. I wrote some hints, you should follow what i wrote and not what i actually did (yeah, it ended well but it was quite overconfident and not the safest approach to a maiden flight).
The plane behaves very well at both low and full throttle. When the current sensor will be properly calibrated i'll try different propellers at high speed (they'll behave differently at high speeds respect to the bench test). So far the motor/esc/propeller combo seems to be ok. The plane doesn't suffer gusts that much (it's a light plane and there were gusts!) and it remains stable and precise. Max speed achieved so far is over 100Km/h but the plane can also glide quite well, it doesn't fear tip stall and the it reacts well even if stressed with some rough maneuvre (tight turn, wing over, loop...). It's not a proper trainer plane but it doesn't require a lot of skill to be flown peacefully.
The GC should be exactly on the forward edge of the carbon spar. With a spot on CG the flight characteristics are amazing, if the CG is 5mm far from the best point then meh... I don't like it that much. Anyway, 5mm is quite a lot, and in any case remember that a plane with a forward CG flies badly, a plane with a backward CG flies once.
Power: 20A @ 3S are ok, you can't do an unlimited vertical flight but you have enough power for acrobatic maneuvers. 12A @ 3S would be ok for relaxed flight, 30A or 4S should be reserved for skilled pilot and rigid models (carbon spar and lamniated).
I have to agree with other reviews, it's a really nice airplane kit that can be adapted to a lot of different tastes, it can be easily set up to fly at more than 130Km/h, or it can be a slope soarer, or it can be an extremely efficient long range platform (it's excellent).
Don't be as dumb as me: don't attach the HD camera on the canopy! OBVIOUSLY its plastic clip can't retain it in place during acro manouvres, it should have been cristal clear. Use a velcro or something else but attach it to the fuselage. Sorry, no HD video footage. :|
I don't have anything else to say except that i'm thinkig about a Z-84 Mark-II, I'd like to add the only feature that i miss on this plane: the capability of detach wings to be carried in a backpack (like my zohd dart).