So this post going to be a little out of order, but I my progress on building a robot base has been much slower than I'd like, so I figured I'd write a little about my efforts to revive a broken "Hoverboard".
Some months back, I decided I wanted to buy a "Hoverboard" to turn it into a new robot base. (Reason for that in my next post). I went to Fry's electronics and bought one for less than $100. The only problem, it was dead on arrival... I tried to charge it, but the charger seemed to start and then immediately stop. So I did the logical thing, I tried to rip the unit apart to take a look inside. (Much more logical that going to return it right?) Anyway I discovered that the battery inside the "Hoverboard" was dead, like really dead. The nominal voltage on the battery said 36 volts, and I measured something like 30 volts...
For the uninitiated, even though the nominal voltage is 36 volts, a fully charged battery like this should be around 42 volts. I figured that the reason for the charger's behavior was that the battery's voltage was so low, that the charger detected some sort of over-current and stopped charging, BUT the charger seemed alive for a fraction of a second when you plugged it in. I decided to splice a relay into the charging line, and see if I could leverage that fraction of a second over a long period of time to see if it would raise the voltage to something more acceptable. Below are 2 videos of this.
After running it overnight, (I slept horribly that night) I raised the voltage to about 34 volts shown below. Not too shabby right? The problem though is that this method plateaued at this voltage, but the voltage was still too low and tripping the over current conditions.
I decided to go for the nuclear option / unsafe option? I hadn't realized initially but the overcurrent protections were actually on the "Hoverboard" itself. But regardless, I decided to try to connect the charger directly to the battery. (I unfortunately didn't grab photos of this.) I initially did this with a couple of resistors on the line to try to limit the current, but I was able to determine that the current coming from the charger wasn't high enough to warrant them. I could charge the battery faster with a direct connection. A few hours later, I was riding around on my hoverboard.
After that adventure, I did a little more inspecting of the "Hoverboard". I discovered the type of unit I had was not reversed by hardware hackers online, and I wasn't able to find instructions to convert it to a robot base just by flashing some new firmware (This was what I wanted to do). I thought it was still neat to see a slab of metal that they used as a heat sink for the MOSFETs, and I also learned that the footpads aren't actually buttons that activate the "Hoverboard". Instead, 2 pieces of rubber insert themselves into a "light gate?" sensor when you step on the device, that tells the balancing algorithm to start.
So I'm a little disappointed that I'll probably need to buy some custom ESCs to use the motors, but I think this processes was mildly entertaining.
Anyway I might make another post about why I wanted to get a new robot base soon.
Some months back, I decided I wanted to buy a "Hoverboard" to turn it into a new robot base. (Reason for that in my next post). I went to Fry's electronics and bought one for less than $100. The only problem, it was dead on arrival... I tried to charge it, but the charger seemed to start and then immediately stop. So I did the logical thing, I tried to rip the unit apart to take a look inside. (Much more logical that going to return it right?) Anyway I discovered that the battery inside the "Hoverboard" was dead, like really dead. The nominal voltage on the battery said 36 volts, and I measured something like 30 volts...
For the uninitiated, even though the nominal voltage is 36 volts, a fully charged battery like this should be around 42 volts. I figured that the reason for the charger's behavior was that the battery's voltage was so low, that the charger detected some sort of over-current and stopped charging, BUT the charger seemed alive for a fraction of a second when you plugged it in. I decided to splice a relay into the charging line, and see if I could leverage that fraction of a second over a long period of time to see if it would raise the voltage to something more acceptable. Below are 2 videos of this.
After running it overnight, (I slept horribly that night) I raised the voltage to about 34 volts shown below. Not too shabby right? The problem though is that this method plateaued at this voltage, but the voltage was still too low and tripping the over current conditions.
I decided to go for the nuclear option / unsafe option? I hadn't realized initially but the overcurrent protections were actually on the "Hoverboard" itself. But regardless, I decided to try to connect the charger directly to the battery. (I unfortunately didn't grab photos of this.) I initially did this with a couple of resistors on the line to try to limit the current, but I was able to determine that the current coming from the charger wasn't high enough to warrant them. I could charge the battery faster with a direct connection. A few hours later, I was riding around on my hoverboard.
After that adventure, I did a little more inspecting of the "Hoverboard". I discovered the type of unit I had was not reversed by hardware hackers online, and I wasn't able to find instructions to convert it to a robot base just by flashing some new firmware (This was what I wanted to do). I thought it was still neat to see a slab of metal that they used as a heat sink for the MOSFETs, and I also learned that the footpads aren't actually buttons that activate the "Hoverboard". Instead, 2 pieces of rubber insert themselves into a "light gate?" sensor when you step on the device, that tells the balancing algorithm to start.
So I'm a little disappointed that I'll probably need to buy some custom ESCs to use the motors, but I think this processes was mildly entertaining.
Anyway I might make another post about why I wanted to get a new robot base soon.
Comments
Post a Comment