Powering Teensy 3.1 off of a 11.1V (3S) LiPo

Status
Not open for further replies.
G

Gibbynator

New member
First off, I am building a quadrotor using the Teensy 3.1 as a controller (because I'm insane). I figure I'll have a few questions along the way since I'm new to the Teensy.

I am currently using an 11.1V 3S 30C 4500mah LiPo battery to power the motors. It will still have plenty of juice left to power the Teensy, but the question is what is the best way to go about this? I was thinking of using an L7805 voltage regulator off the power board to connect it, but allowing 1A through the Vin sounds a bit dangerous. I figure a resistor or two after the thing would be a good idea. Is this the safest approach to delivering a stable Vin to the Teensy throughout the flight (I have an alarm already attached that will sound when the LiPo starts getting low). If so, what resistors would be best to use? Otherwise, is there a better way to do this or should I just get a 3.7v 1S LiPo and run the Teensy off that? I've seen plenty of Teensy LiPo chargers, but nothing that charges the Teensy from a LiPo. Also, is there a max Amps in I should be concerned about if it's best to just get a second LiPo, or does the internal Voltage regulator in the Teensy take care of that?

Thank you for your advice.
 
I'd first ask what the Teensy is supposed to do. If the answer is almost nothing, then perhaps even a built in voltage regulator on a 3.2 night survive for a while.

However, if the teensy needs to do a serious amount of work (i.e. write to SDHC cards or some such) then I'd counsel you to get a DC-DC power supply that takes the voltage down to 5V or even 3.3V (if well regulated). This is not only for efficiency sake, it's also to keep the heat down inside your copter. Linear power supplies produce a lot of heat when the voltage input and output are far apart.
 
Constantin said:
I'd first ask what the Teensy is supposed to do. If the answer is almost nothing, then perhaps even a built in voltage regulator on a 3.2 night survive for a while.

However, if the teensy needs to do a serious amount of work (i.e. write to SDHC cards or some such) then I'd counsel you to get a DC-DC power supply that takes the voltage down to 5V or even 3.3V (if well regulated). This is not only for efficiency sake, it's also to keep the heat down inside your copter. Linear power supplies produce a lot of heat when the voltage input and output are far apart.
Click to expand...

The Teensy will do the following operations:

  • Read sensor data from a MPU9250 and BMP180 (sharing i2c ports, but I don't think there is any address overlap)
  • Stabilize sensor data using Mahony/Madgwick algorithm
  • Perform autonomous hover in place using sensor data
  • Receive controller input through an APC220 wireless transmitter

I doubt it will require massive amounts of power; I'm thinking I won't need to push the processor above 72MHz. Should be fine with only drawing 100-200ma. Weight is just as big an issue as heat, which is why I need to have the lightest possible solution.
 
You need this beauty.

By the way, ditch the BMP180 for a BMP280, much better performance in a smaller and cheaper package.
 
onehorse said:
You need this beauty.

By the way, ditch the BMP180 for a BMP280, much better performance in a smaller and cheaper package.
Click to expand...

Thanks for the advice on the part. Is this the same thing?: http://www.mouser.com/Search/m_Prod...jHzVIkrqUo2ijSvNMI1BHOwF7L09OqLle%2b1OkI/nQ==
Should I shoot for the 5V application or is the 3.3V sufficient?

I'll consider the BMP280, I already have the 180 and stuck it on the breadboard with the MPU9250 and the Teensy. Trying to get output through the USB to a GUI app so I can visualize the sensor data. My goal is to get this thing in the air by the end of November.
 
Yes, that's the part. I use the MP3620A not the MP3620. There's a difference in the power good pin.

I have Teensyduino sketches for the MPU9250 and BMP280 that you can adapt to your application pretty easily if needed.
 
Apologies for the slightly off-topic question: One Horse, would your module also have enough head room for running a Teensy LC from a 12v car battery or marine battery charged by both a windmill and a solar panel? As I understand it, these batteries can run as high as 16v, though it's not a good idea. I notice that the MPM3620A voltage regulator is rated for 24v.
 
The regulator can handle voltages higher than 12 V, as you say, but the resistor network is chosen to be optimized for 8 - 12 V . Above this voltage, I think they recommend different resistor values. Of course, this is easily changed with a soldering iron, or if you want I can assemble the modules with resistors (set to a voltage range) of your choice. Consult the data sheet to see the ranges available.
 
Last edited:
onehorse said:
Yes, that's the part. I use the MP3620A not the MP3620. There's a difference in the power good pin.

I have Teensyduino sketches for the MPU9250 and BMP280 that you can adapt to your application pretty easily if needed.
Click to expand...

As it turns out, I was actually using those exact sketches (MPU9250BasicAHRS_t3 specifically) to attempt to learn the necessary stuff, but it failed to compile for me, even after I commented out the lines for the Adafruit display. Probably because I was stupid and forgot to include the quaternion INO file :p I'll give it another try for tonight. I'll try to get the file to work with the display app found here: http://www.lucidarme.me/?p=5057 and hopefully I'll be able to get it running.

Is one of the other sketches compatilble (or close to it) with the BMP180/280? If I remember correctly both models use pretty much the same API, like the 9050 and 9250 do.
 
Last edited:
One option to consider is using the balance charger connector that most 3C Lipo's have. This is the small JST connector typically connected to a charger to allow it to level the cell charge levels. These allow you to tap into the battery back at 3.7 or 7.4 V levels. Surely a bad idea for significant loads, but may be fine for your purpose.

Most flight controllers take power from the ESC's that control the motors, most have a BEC (battery eliminator circuit) that supplies 5V. My quad uses a BEC to power the flight controller. I power the receiver directly from the 3C motor battery (it has an internal regulator) because it is the battery level monitor / alarm.
 
Last edited:
Just for the fun of it, I have been playing around with some neopixels from Adafruit and like some others up on Trossen Robotics, want to be able to use one or more of these on a few different robots... So far I have using an Adafruit Pro-trinket for this as their VR can handle the 12v... But I prefer the Teensy!

So I have been hacking around using Diptrace to build a much smaller board (more like a hat) for this. It is probably not great, but I ordered a batch of 3 today to see if they work!
teensy-Neo-3d-dpak.jpg

I played around with a few different types of Voltage regulators and the one that I found that fit, and that I could also potentially get maybe up to 1.5 amp version for more pixels like one of the rings, uses the layout type of DPAK.

Note: the board is setup with two AX Buss like connectors, one horizontal and one vertical so you can put them in the AX Buss, plus can use one of the Chip Neopixels or one of the Through hole versions and/or extend off with others. I added a few extra +5v/3.3V/Gnd pins to connect other things, and if I need more IO pins, will either use tall connections and the double type...

So in a few weeks I will try one out.
 
Last edited:
tranger said:
Most flight controllers take power from the ESC's that control the motors, most have a BEC (battery eliminator circuit) that supplies 5V. My quad uses a BEC to power the flight controller. I power the receiver directly from the 3C motor battery (it has an internal regulator) because it is the battery level monitor / alarm.
Click to expand...

How stable is the 5V output from the BECs when the motors are running? I was debating using that 5V source directly for my own flight controller, but wasn't sure if the output voltage from the BECs would be low-noise enough to not cause issues.
 
joe_prince said:
How stable is the 5V output from the BECs when the motors are running? I was debating using that 5V source directly for my own flight controller, but wasn't sure if the output voltage from the BECs would be low-noise enough to not cause issues.
Click to expand...

I don't know any specifications. However, they are widely used to power flight controllers. The ECS MCU's are powered from the same source, my particular units use Atmega 8's. Shouldn't be an issue.
 
tranger said:
I don't know any specifications. However, they are widely used to power flight controllers. The ECS MCU's are powered from the same source, my particular units use Atmega 8's. Shouldn't be an issue.
Click to expand...

I have been using them as well, and haven't seen any issues. It would be nice to be able to test the output voltage noise, perhaps with some data logging of some sort during flight. A scope would be ideal, but testing on the bench wouldn't be that representative.
 
Status
Not open for further replies.
Back
Top