Can't power up omega2 without dock: part 2



  • Ok, The ywrobot didn't work as mentioned on my previous post so I bought this:
    http://www.monarcaelectronica.com.ar/modulo-alimentacion-5v-3v3-800ma-ams1117-arduino-137974242xJM

    Plugged in an old power supply that outputs 5v and 800ma, the result was the same, 10-15 seconds the amber light turned on and then nothing.

    Later I tried with an usb cable as power supply because I read it on a post here, and it worked for a brief period of time, I was able to connect to the wifi and start the configuration process, but on the wifi step it turned off again. After that I wasn't able to power it up with the usb cable again. Strange. I thought it broke, but when I tried with the 5v 800ma cable again, the amber light turned on for 10 seconds and then off again.

    I also tried with a variable power supply to the regulator, from 1.5v to 12v and 1A and from 4.5v and below nothing happened, from 6v to 12v the amber light turned on but only for 10 seconds again.

    I don't know what I'm doing wrong. Maybe power stability?
    I used a multimeter to measure and the range from the regulator was from 3.25 to 3.30v depending on the power supply.

    please let me know if any picture may be helpful.



  • Hi,

    I've used a lab power supply to check my Omega 2+ voltage-current consumption profile.
    Note that by using my lab power supply, I can intentionally limit the max output current.

    With that, my first test is to set the max. output current to be 180mA. The output voltage is set at 3.3V.
    In this setup, I've observed exactly the same behavior:

    • In the first 2 second, amber LED stay lit. After that it's starting blinking. (In this entire 10 second or so duration, current consumption never exceed 170 mA.)
    • then all in sudden, amber LED went dark and the current consumption drops to only 60 mA which means it's no longer in booting state and probably hang.

    I then set the power supply output current limiting to 300 mA. My Omega 2 in this case, boot up successfully and logged in to my home Wifi AP.
    I checked about the 10-second time mark and found out there is a brief duration the current consumption
    shoot-up to 260 mA or so.

    I find my Omega 2+ nominal current consumption is about 180 mA, but need to handle the peak current demand.

    Summary is, the power supply need to be able to supply minimally 300 mA to Omega 2+'s Vcc and Gnd pins.
    If not, a brown-out condition will happen and it won't start-up or work reliably..

    The common solution is to make sure the power supply have a large output capacitor (serving as a reservoir) and make sure the 3.3V power cord is not too thin (I recommend 18 to 20 AWG, if no other docking circuitry.)

    BTW, your ams1117-3.3V power supply module (and many similar ones from eBay etc.) is well known for cheat out on both input and output filter capacitors. I would place a 470uF 10V electrolytic cap on its input side and a 1000 uF 6V on its output side. It will prevent the temporary voltage drop (due to burst power demand from the board) and you'll have a happily running Omega 2+.

    < note: edited to bump up the capacitor values >

    ccs_hello



  • @ccs-hello That pretty well all makes good sense.
    Matches closely to what I mentioned in https://community.onion.io/topic/1517/access-point-not-found/26



  • Hello, look at my post, I had the same problem with power supply...

    https://community.onion.io/topic/1476/omega-keeps-resetting-before-boot/10



  • @Agustin-Valeriani similarly I wanted to provide power without a dock. Using a 9V power supply I used the following device.

    https://www.youtube.com/watch?v=tU0ztOQP6PA

    If you like the supplier of the thing you bought that didn't work-- they sell the thing below that should.

    http://www.monarcaelectronica.com.ar/fuente-para-protoboard-5v-33v-salida-usb-mb102-arduino-137819247xJM



  • @Brad-Chesney 60 cents (+ PSU) looks a more reasonable price to power a "$5" Omega than a $15 dock (+PSU).



  • @ccs-hello etal
    I posted the results of some testing I did using a little different procedure than "ccs-hello" and saw indications of spikes to 420 mA on one of my Omegas. I've pasted that post below. I second ccs-hello's comments about wire gauge and filter capacitors and would add that you want the wires short and the 3.3 V capacitor close to the Omega power pins.

    Pasted post follows:

    @Mark-B said in Omega2+ hot hot hot:

    With my Omega2+ running on a proto-board with nothing but a linear regulator and 22 uF tantalum capacitor on 3.3 V, I got the following data points. This is monitoring current with a inexpensive USB voltmeter which has questionable accuracy at low current levels and updates about twice per second, so I'm probably not seeing the absolute peak current.

    • idle current ~= 110 mA

    • current running cpu intensive task* ~= 170 mA

    • current running wifi task** ~= 280 mA average, peaks around 350 mA - board is noticeably warmer to the touch, but not hot

    I think this suggests that CPU activity doesn't matter much. If an Omega2 is particularly warm (and not otherwise defective) it's likely because the wifi activity is high for some reason.

    * - Python for cpu task gives near 100% cpu

    import random
    import math
    x = 0
    for m in range(0,1000000): x = x + math.sin(random.uniform(0.0,1.0))
    print(x)

    ** - Python for wifi task run via ssh over wifi gives about 50% cpu load

    for k in range(0,10000):
    __ for m in range(0,k): print(m,k)

    Edit to add: I see similar numbers for my Omega 2 (not plus) except "wifi" numbers are higher, with and average of about 330 and peaks up to 420 mA.

    Also, the current spikes are apparent in the boot process where the serial debug suggests wifi activity with the first about 20 seconds into boot.



  • Thank you all for your responses. Sorry for the delay, I had a final exam.
    I tried the capacitors and they didn't work. Not even the starting amber light. I don't know if I didn't set them up the right way, but nothing happened.
    I'm going to try the other proposed solutions.



  • About @Kit-Bishop response, I think it may be the wires, I'll try with smaller ones (i'm using the ones proposed on the documentation).



  • @Agustin-Valeriani said in Can't power up omega2 without dock: part 2:

    About @Kit-Bishop response, I think it may be the wires, I'll try with smaller ones (i'm using the ones proposed on the documentation).

    If the "smaller" you wrote means shorter power wires then yes try as short as is reasonable. Thicker is better, too.



  • @Ken-Conrad Right, I mean shorter.


Log in to reply
 

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