whoop

Last time I left off at having finished the soldering and successfully tested that I could connect with my radio to the flight controller and receive a video signal on my goggles, and that the video signal was right side up. I tested that the VTX protocol was configured correctly so I could use my radio to bring up the VTX settings in the OSD on my goggles and change the VTX band and channel, and that the frequency the VTX was transmitting at changed appropriately.

Everything worked as expected, so aside from the one solder bridge issue that I had to fix the first part of the build was pretty trouble free. At this point I figured that the hard parts were done as the rest of the work was just assembling the pieces onto the frame, which normally isn’t any harder than building a Lego set.

First step is to put the rubber mounting grommets on the corners of the flight controller and then mount it in the whoop frame. The grommets are hollow and fit over plastic posts that have threaded screw holes in them. I put the screw in at the rear of the flight controller as that one wouldn’t be covered by the camera mount, and it left me get the flight controller partially secured into the frame.

The next step was to get the camera snapped into its mount and then fit that over the two side and front grommets of the frame, and get all three remaining screws in and snugged down. One of the major bad habits that I’ve had to overcome with building quads is over tightening screws. Especially in instances like this where doing so flattens out the grommets and cancels out any vibration protection that they might provide for the flight controller. Or worse stripping out the plastic posts so that the screws don’t hold at all.

Once I got the camera mounted, it was time to get the VTX on. I’d tried a couple different options before starting the build and I thought I’d figured the best way to get the VTX and antenna mounted. I started by cutting some clear heat shrink to keep the VTX from touching the back of the camera and shorting out. It also had the added benefit of holding the antenna in place so I didn’t have to worry about trying to secure that as well. With the heat shrink on then I zip tied it upright to the back of the camera mount with the status LED and button visible.

With all of that out of the way, all that was left was the motors and props. After doing the first motor, I discovered it was easier to plug them into the flight controller before screwing them into the frame. So that’s how I did motors two through four.

Putting the props on requires some attention to detail since props come in two versions clockwise and counter-clockwise, and they have to be put on in a certain order. For my setup, I prefer what’s known as reversed props or props-out. Which means that the front right and rear left props spin clockwise, and the other two counter-clockwise. There’s some aerodynamic benefits to setting up the props this way, Oscar Liang has a good article on that.

Usually I don’t have any problems figuring which prop is which and getting them all on the correct motors, but I ran into a novel problem with the new Azi props I was using. I got one of the motors put on upside down and didn’t realize it until all four props were on. I’d never had a problem with putting a prop on upside down before. To make things worse, these were push fit and didn’t need or use screws. Which meant they were extra hard to get back off in order to fix them.

In the end because the whoop frame was pretty flexible and the props were on so tightly, I ended up having to unmount the motors so I could get a good enough grip on them to pry off the props and get everything back on right-side up and in the correct order.

The last step was to hook the quad back up to my computer and launch the BLHeli utility to make sure that all four motors were spinning the correct direction. In this instance motors two and three were backwards, but it’s an easy software change to make and write to the ESC’s. Once that was done the build was finished.

All together the new whoop weighs 24 grams and 32 grams with a 350 mAh battery, that’s just 2 grams more than my brushed Acrobee.

The last step naturally is to fly it. I charged a couple of batteries up and plugged the first one in to do the maiden flight and ran into an issue where only three of the motors spun up after the quad was armed. I tried disarming and then arming again and all four spun up but not all at once. This is odd as I’ve always had all of the motors spun up at the same time.

At first I thought I might have a power issue or some bad motors or ESC’s or something. But I found that when I connected the quad to my computer and tested the motors individually they all spun up fine. It only seemed to be a problem when I armed the quad and all four would try to start at once. I looked online but wasn’t finding any good results, not until I thought to go to BetaFPV’s site for the motors I was using. There was a link there to a support page that covered the issue I was having, which led me to looking at the startup power setting in BLHeli.

This being the first quad I’d built or used with the combination of brushless motors and 1S voltage, it had never occured to me to think about the initial amp draw that happens when the quad is armed. But it made sense once I read through the page, since the motors were fine when tested individually, and sometimes when arming I’d get a couple of motors to spin up initially and then another one or two a moment later. I ended up increasing the startup power for all four motors to 1.5 going in 0.25 increments until I could consistently get all four motors to spin up simultaneously when arming the quad.

Altogether it took me about 28 minutes to do the initial soldering, another 11 minutes to fix that solder bridge, and nearly 50 minutes to do the assembly including having to redo the motors and props and figure out the motor power issue. So the whole process was about 90 minutes, which isn’t too bad.

I’ve flown a few dozen batteries on the whoop since then and it’s quite a bit of fun to fly inside. It’s just on the edge of being too much power for indoors at least in my home which doesn’t have a lot of open space to fly. I think when I take it to the local fly-in office space it will be a lot more fun than my old brushed quad.

I also tried it outside just recently when there was some good weather and it did okay, but it’s not nearly as much fun as my toothpick is, and I only did two batteries outside out of curiosity. I think most of the time that I can fly outside it’ll be my bigger quads.

It took me a while to actually sit down and do the build for the brushless whoop parts that I got for Christmas.

The first thing I like to do on a new build is test the flight controller. Especially before any soldering is done, it’s good to make sure that the flight controller works, so that later on if it doesn’t I know that it’s because of something that I did and can troubleshoot it instead of wondering if I just got a bad one from the factory. It’s easy enough to do. Plug it into a USB cable to make sure it gets power, and then make sure that Betaflight will connect to it. Test out the gyro by turning and tilting the board and make sure that the movements match up with the animated diagram in Betaflight.

At that stage, it’s not critical to do, but since it’s already connected, I like to download the latest firmware and flash it to the board. It’s just a couple of steps to do, and isn’t any harder or easier to do now or wait until later. But since it’s already connected I figure why not.

The last step I took care of before really starting the build was binding my radio. I don’t always do this part so early, usually because for anything bigger than a whoop build, I don’t use a flight controller with a built-in receiver. So there’s some soldering that normally has to take place first. But in this case since the receiver is built into the board, it’s much easier to do now where the bind button is easier to get to, than to wait and do later.

Binding the radio took me a bit to figure out since I had to find the built-in receiver specifications. I couldn’t find anything about telemetry being available or not, but it would only successfully bind when I picked the radio setting without telemetry, so I’m guessing it’s not supported. Not a big deal for a small whoop like this, but something I’ll have to keep in mind if I ever decide I need to go hunting for it outside.

With the flight controller testing out of the way, I was ready to start figuring out the build. This involves dry fitting the flight controller to the whoop frame and trying to get an idea for where the canopy, camera, and VTX were all going to sit. I found the wiring diagrams for the flight controller on BetaFPV’s site, and used those along with the printed manual that came with the VTX to figure out how everything needed to be wired together.

The only part that wasn’t immediately obvious to me until I’d studied it for a couple of minutes was that the signal wire of the camera needed to go to the camera pad on the flight controller. This would apply the OSD information, and then a wire from the VTX pad to the video in pad on the VTX would take the signal for broadcast. I also didn’t need one pair of the 5V and ground wires on the VTX since the camera already had its own.

Once I figured all of that out, then I rechecked it by fitting all of the pieces together to make sure I knew which wires to cut, which to desoldering, and what was getting soldered to what. I could always fix any mistakes, replacing or lengthening any wires that I cut too short, but I didn’t want to have to mess with that if I could avoid it with a little extra care up front.

After checking a few times though, it’s usually time to just commit and do it and hope I’ve thought through everything I needed to. In this instance that included trimming all of the VTX wires to be just a bit longer than the camera wire lengths. Trimming the plug from the camera, which I wouldn’t need, and then desoldering the two wires from the VTX that I wasn’t going to need.

Desoldering is quite easy. It just requires me to get the piece set up where I can get some pressure on the wire that I’m trying to remove, so that I can have a minimal amount of time for the soldering iron on the board. In this case the extra 5V and ground wires that were going to be replaced by the corresponding wires on the camera.

After getting the wires cut to length, I was ready to tin the wires and the pads on the VTX and flight controller. Tinning just means to prep exposed wires and any copper pads on PCB’s that I’m going to be using with solder. Ideally so that when it comes time to solder a wire to a pad, I don’t have to add any additional solder, I can simple touch them together with a bit of heat and I’m done.

Soldering a small build like this is a bit of a challenge just because of the sizes of the pads involved. I’ve found using the lowest heat setting along with a lighted magnifying glass, and a very small soldering iron tip helps tremendously for this. I’ve also applied some tricks I picked up in painting miniatures for keeping my hands steady by moving the piece to the iron and keeping my arms are my sides.

Once everything has been tinned and then soldered together, the easy part is over. With the camera, VTX, and flight controller all wired up, it’s time to check for shorts and then plugin in a battery.

Before plugging in a battery and risking releasing the magic smoke, I like to use a multimeter and check that there’s no continuity between the various connections. For example the yellow wire from the VTX to the flight controller should have continuity between the ends, but if I touch a probe to one end and then touch the other probe to the ground then I shouldn’t have any connection. If I do it means there’s a short somewhere, and on a tiny build like this it’s probably because two pads are bridged.

In my case I did have a solder bridge on the VTX between the video signal and the ground. But that’s just a matter of adding some flux to the board and using a soldering iron to try to clean up the bridge. It only took me a couple of minutes to get it cleaned up in this case, and afterwards all of the connections passed the multimeter test.

Which brings me to the last part before assembly and that was plugging in a battery and testing that everything booted up, that the VTX was broadcasting, and best of all that I was getting an image through the goggles from the camera. It’s good to do all of that before assembling everything just to make sure that there’s no more soldering fixes that need to be made, and it’s a good chance to make sure I know which way the camera needs to go so that up is up.