Announcing APM 2.5!


I’m pleased to announce the new APM 2.0 revision, better known as APM2.5!

But before you start screaming that you just got an APM 2.0, let me explain that there’s NO difference in performance at all between the two, the board is still the same in terms of functionally and it runs exactly the same code as APM 2.0. We just took the liberty of throwing in a few new universal connectors in there (so accessories can be cross compatible with the new PX4 and protect the user from connecting the cable in the wrong direction), some production improvements that will allows to manufacture it faster and finally we added some protection features to protect it from those reverse polarity/short circuit lovers.

Here goes the full list of changes (you already know all its features):

  • No more shield/daughterboard: magnetometer and dataflash were moved to the main board, making it easier to assemble, repair and hack. 
  • No more on-board GPS, but we added a new 6-pins DF13 connector for our new GPS cable standard (PX4) and kept the old but reliable Molex (EM-406) connector for old GPS compatibility. 
  • The old radio-modem connector is gone and now is being replaced by our new 6-pins DF13 connector (like the GPS, because both are UART’s. Got it?), this will make it compatible with PX4 and prevents users from blowing up the UART mux. 
  • We added a second protection fuse and diode (you are no longer required to solder it yourself) for extra reverse polarity and short circuit protection. 
  • The old I2C connector is gone and has been replaced by our new 4-pins DF13 connector (yeah PX4 compatible. Oh wait correction… None of both (PX4 & APM2.5) are compatible with each other, we created this standard together as a team, but I have to give the average of the credits to PX4). 
  • Added a “mystery” 6-pin DF13 Connector that will be used later. I know you like surprises so please stay tuned. No big deal but is really cool and will save you some soldering here and there. 
  • Added “Magnetometer Jumper” that allows you to disable the on-board magnetometer, so you can attach an external one. 
  • Added extra LED attached to the PPM encoder (Atmega32-U2) on pin PC7. This will allows us later to use it as PPM status indication and leave those poor TX/RX LED’s do their intended job.


Those who ordered the APM2.0 and are still waiting for it (only if is not shipped yet) will have the option to upgrade to the new APM2.5, but you will be required to wait an extra couple of weeks for production to begin. For upgrades please send us an email to 

If you want to order one now you can get the assembled version here and the unassembled version here (for those who want their choice of RC connector: straight or right-angled). They will begin shipping in a week or so, but if there is a huge flood of orders, the backlog for any individual order may grow to 2-4 weeks. We're making them at the new Tijuana facility, so we hope that we'll be able to keep up with demand much better than we have in the past. We're committed to shrinking all delays, and our investment in the Tijuana facility is just one sign of this.

The APM 2.0 enclosures will not fit this new board, so new enclosures are on their way. If you buy an APM 2.5 now, you will get a $4.99 credit for a free enclosure with your next order (the enclosures should be in in about a month).  Also, APM 2.5 will ship with an adapter cable for the 3DR/Xbee radios, converting from the APM 2.5 DF13 connector to the radios' conventional 6-pin connectors.  Eagle files for the boards will be posted in the next few days.

E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones


  • whats size is the data flash chip?

  • @George,

    Don't be discouraged, if you have (or want to learn) some soldering skills you can easily accomplish some of the benefits found in the 2.5.

    The off board GPS is easily done by just adding solder between two pins and using the external GPS connector, it's even in the wiki.

    As for the compass, I have not yet confirmed myself (but will soon) that all you need to do is cut a single trace SDA1 near the compass to enable using the external I2C port.

    Then there is the Optical Flow sensor which is an open topic at this point on the 2.5 that you may be better off having the 2.0 for if you plan on using it.

    Now, back to having a good morning for you! :-)

  • Oh no :-( Jut received my APM 2.0 after waiting for 9 weeks :-( and now it is outdated already :-(

  • Optical flow:

    Just looked at the Eagle files some more, none of the ports have the 3 wires it takes for SPI (including the mystery port which has either 2 generic I/O's or ADC pins, VCC, and GND).

    It could be that landing the OF on the Atmega32U2's 5v SPT/ISP and adding a 3.3v <---> 5v logic level shifter and writing code to route the OF's SPI through the 32U2 is the only less-than-insane solution?

    Maybe the "mystery port" is for a new I2C OF sensor that kicks the snot out of this SPI one ;-)

    Still not getting this.... guess I'm going to plant this 2.5 in my heli as-is, funky data flash wiring and all. I've got some board mounting issues to solve before this weekend anyway, so maybe I'll get lucky and have an answer by then.

  • Can you still connect the FTDI pins instead of using the USB port?

  • Chris, any response on the Optical Flow connection?

  • 3D Robotics

    Max: That's fixed in the latest code (ArduCopter 2.7.2, ArduPlane 2.6). It's in the dev trunk now and will be pushed to the Mission Planner for public release in the next week or so. 

  • Was on-board flash logging bug fixed at 2.5? See discussion at . Now it's impossible to analyze on-board logs which renders APM 2 unusable if you try to go beyong simple flying. E.g. if you want to implement and debug any new features.
  • I'm also very anxious to see how Optical Flow is supposed to work. It looks like the code (in the library for 2.7.1) doesn't currently have any provisions for using any other pins on the APM2.5 just the same ones the APM2 uses. As Jeff points out, that's not workable.

    Can the ISP pins be used if we add a level shifter? It looks like UART2 might be available and easily solderable for this purpose (can all the atmega2560's USARTs be run in SPI mode?) but it would require code changes, and I haven't had a chance to test that yet.

  • No answers anywhere, so after MANY hours of testing and research I figured out the only place I could find to land the 3.3v SPI for the OF sensor was directly to pins 1,2, and 8 of the Data Flash card. SPI/ISP of the 2560 and 32-U2 are both 5v and though I tried neither of them would allow the OF to initialize. The 3.3v to the MPU-600 is also a no go, if you even touch the MOSI the MPU goes haywire...

    But, it IS working in the CLI at least where I have it now, 12+ hours later.

    Come on 3DR, I must be missing something right???? There is   no way   you expect someone to put wires where I did right???? Please tell me I am an idiot and I've just overlooked the most obvious of things. :-/

This reply was deleted.