I Can't boot my Omega2+ without a Dock



  • I received my Omega 2+ but I didn't buy a dock (thought it wouldn't be needed).
    So, to boot, I first tried using Arduino to provide 3.3V but I noticed it doesn't provide enough current.

    I then bought a YwRobot (protoboard power supply) that has an AMS1117 3.3 voltage regulator that can provide up to 1A output.

    My Omega2+ then starts, LED turns on and stay on for about 20~22 seconds. It then turns OFF and nothing else happens.

    So, I decided to use a Lab Power Supply, and the face the same behavior. It turns On for about 20secs and turns off. I was able to see that while the LED is on, Omega is consuming around 180 mA, and when the Led goes off, it keeps consuming around 60mA.

    The link to boot without a dock, tell to use LD1117, but uses no Capacitors around it. while the YwRobot has cap.

    Any other suggestions on how to boot it without a dock?
    Might I replace the LD1117 (I'm not finding it around), by a AMS1117? Should I skip the capacitors recommended by the regulator manufacturer?

    Thanks



  • I have a couple of 2+'s that are using AMS1117's but the Omega's are very fussy about the power feed they receive. Very small margin of error, use a decent PSU to feed the AMS1117's. I don't use caps.





  • I am experiencing that Omega2+ has difficulties in a densely populated area due to WiFi interferences / jam.
    It begins to boot, turns the adio on and shortly after simply completely turns off (Kernel panic ... poweroff).

    Try to move a few meters away and hope this will help.

    https://community.onion.io/topic/1595/kernel-panic-not-syncing-stack-protector-kernel-stack-is-corrupted-in-and-a-possible-reason



  • @Thiago-Gomes-Marcal-Pereira said in I Can't boot my Omega2+ without a Dock:

    Should I skip the capacitors recommended by the regulator manufacturer?

    Definitely not. Use the recommended capacitors, and potential even more.

    They are important both for the stability of the regulator, and because the main challenge that the MT7688 present is not the overall power consumption, but the abrupt pulse loads caused by the wifi.



  • I tried using only the AMS1117 without caps, and it had the same behavior.

    I could connect my Omega2+ to my PC using a FTDI and got the logs.
    It always ends like the following:

    [ 18.515046] EEPROM:Read from [factory] offset 0x0,length 0x400.

    I'm going to try using capacitors both in and out.



  • Does anyone know which Capacitor values are used in the original dock?



  • You'd be better off reading the recommendations of the regulator manufacturer

    The circuit design used in the AMS1117 series requires the use of
    an output capacitor as part of the device frequency compensation.
    The addition of 22µF solid tantalum on the output will ensure
    stability for all operating conditions.

    Most engineers would also put a .1uF capacitor with short leads right next to both input and output.

    and later

    To further improve stability and transient response of these
    devices larger values of output capacitor can be used.

    Transient response is exactly what you need here; however even then, you also need an upstream supply to the regulator that has good current delivery and response to pulse loads. A random USB port or phone charger may or may not be able to do that (technically, the USB spec only allows 100 mA without negotiation, though that is not often intentionally enforced).

    It's also worth considering that a linear regulator like the AMS1117 places a greater demand on the upstream supply than a switching regulator would. A linear regulator is basically a feedback controlled rheostat, which is a polite way of saying "miniature space heater" - it draws the same current at 5v as it supplies at 3.3v. Therefore its efficiency at converting 5v to 3.3v is at best 66%, which is to say that the upstream supply has to provide 50% more power than the Omega2 actually uses, the excess burnt as heat in the regulator.

    In comparison a switching regulator converts power meaning it actually draws less current at 5v than it is delivering at 3.3v. With possible efficiencies in the 80-90% range, there are upstream supplies that could run an Omega2 with a switching regulator, but fall over if trying to run it through a linear one.



  • @Thiago-Gomes-Marcal-Pereira said in I Can't boot my Omega2+ without a Dock:

    I tried using only the AMS1117 without caps, and it had the same behavior.

    I could connect my Omega2+ to my PC using a FTDI and got the logs.
    It always ends like the following:

    [ 18.515046] EEPROM:Read from [factory] offset 0x0,length 0x400.

    I'm going to try using capacitors both in and out.

    The WiFi radio will turn on after that line.
    So in this case probably you really have a power supply problem.

    BUT earlier you said:

    So, I decided to use a Lab Power Supply, and the face the same behavior. It turns On for about 20secs and turns off. I was able to see that while the LED is on, Omega is consuming around 180 mA, and when the Led goes off, it keeps consuming around 60mA.

    So that Lab Power Supply could not provide enough current either for your Omega.
    It's almost unbelievable.



  • @Thiago-Gomes-Marcal-Pereira The Mini Dock and Expansion Dock schematics are here at GitHub.



  • You also want to use moderately heavy gauge wire (say 22 gauge) between your regulator (or lab power supply) and the Onion Vcc and Gnd pins.

    Better would be to put the regulator and bypass capacitors on a small assembly directly connected to the onion pins. Then the wires to the regulator are much less critical.



  • Thanks all of you for the comments. They're being really helpful.

    @Chris Stratton,
    I added 2 caps (220uF in and 470uF out), to try and get that bigger transient response. Tried with a Turbo power charger (max 3A, as Omega would not request that).
    But also didn't work.

    Now, I noticed something strange. My regulator outputs 3.3V, but as soon as I connect my Omega to it, output instantaneously drops to around 2.9V. I have no idea why it being in parallel, would cause this drop.

    Any suggestions now?



  • re: 3.3V drops down to 2.9v
    It means either the wires in between the regulator and Omega2 are too thin, or
    Omega2 itself has unusual short-circuit somewhere.

    In all cases, ALWAYS visually inspect (using a magnifier) the Omega2 board first to see if there are solder splatter and solder bridge which shouldn't be there in the first place. Many had reported similar issues which lead to factory QA.

    ccs_hello



  • Thanks for the help.
    I bought the tantalum capacitors (22uF, 0,47uF).. I'll try with them.
    Also bought 20 awg (0,5 mm) wires.
    They may arrive on tuesday. I'll let you guys know of my results.



  • Solved. Using a 0.5mm wire, I could successfully boot.
    Now waiting some components to build my own dock, with USB functionality as well.


Log in to reply
 

Looks like your connection to Onion Community was lost, please wait while we try to reconnect.