Wanted: 3.7v regulator + backfeed protection circuit.

I am including an Adafruit Feather on a PCB with a propeller several propellers.

Adafruit Feather has two power options: USB or 3.7v/4.2v, 500mAh LiPoly battery.

I would like a technique for powering the Feather that does not involve stumping a bunch of USB cables and soldering about two inches of wire to the board. Or gluing a battery to the board. I notice in the PCB schematic for the Feather that the BAT pin is connected directly to the positive terminal for their battery.

My idea: Use a 3.7v or 4.2v regulator to power the Feather via the BAT pin. Include a feedback protection circuit. Maybe a brownout detector holding a transistor closed(current going through), listening to the line and disconnecting the regulator when more than 3.7v is detected?
Returning to Spin after two months of not coding micros at all, forgetting to use :=

Comments

  • Why not use a USB charger? After looking at the specs and schematic I would say that the on board circuitry power supply and monitoring is more than adequate. One of the ADC's even monitors the battery voltage, which could be modified to monitor the power supply if you are not installing a backup battery on the board.
    In science there is no authority. There is only experiment.
    Life is unpredictable. Eat dessert first.
  • Well, I would like the board to have a main power supply only. This prevents the need to check multiple battery packs or wall plugs. My idea will theoretically replace the battery entirely.
    Returning to Spin after two months of not coding micros at all, forgetting to use :=
  • kwinnkwinn Posts: 8,352
    edited 2019-09-27 - 14:50:16
    Well, I would like the board to have a main power supply only. This prevents the need to check multiple battery packs or wall plugs. My idea will theoretically replace the battery entirely.

    Wouldn't your "main" power supply need to be connected to the 120VAC wiring either directly or via a wall plug? If you want it to be in the same enclosure as the feather that can be done, but it will need some form of input power or charging if it is a battery. We are not at the stage where we have a "main power supply" that can provide power for unlimited periods of time without some form of input energy.
    In science there is no authority. There is only experiment.
    Life is unpredictable. Eat dessert first.
  • Main power refers to the wall plug, yes. It is the main supply for the entire board.
    Returning to Spin after two months of not coding micros at all, forgetting to use :=
  • Main power refers to the wall plug, yes. It is the main supply for the entire board.

    What I do in cases like this is to purchase a 120VAC in module that outputs the desired DC voltage and current, mount it in the enclosure, wire it to the board, and add a 120VAC cord to connect the power supply to the wall plug. For prototyping I may break open a USB wall wart and use the board and cable to power my project board. Thrift stores have wall warts by the dozens if not hundreds so there is no problem finding one for the desired output voltage.
    In science there is no authority. There is only experiment.
    Life is unpredictable. Eat dessert first.
  • Ummm, I wasn’t planning on an enclosure. This is for the VIPMC, which I plan to have in an Eval board sort of form.
    Returning to Spin after two months of not coding micros at all, forgetting to use :=
  • Ummm, I wasn’t planning on an enclosure. This is for the VIPMC, which I plan to have in an Eval board sort of form.

    Then for the sake of safety you should use an approved wall wart or other modular encapsulated power supply that does not have 120VAC exposed.
    In science there is no authority. There is only experiment.
    Life is unpredictable. Eat dessert first.
  • https://www.parallax.com/product/750-00003

    This was the idea. Power the board off this dude. Convert 12vdc down to 3.7vdc. Likely use boost converter/buck converter, such as to not lose amperage. Use another converter to power props, sd cards, USB ports, EEPROMs, and XBee.
    Returning to Spin after two months of not coding micros at all, forgetting to use :=
  • https://www.parallax.com/product/750-00003

    This was the idea. Power the board off this dude. Convert 12vdc down to 3.7vdc. Likely use boost converter/buck converter, such as to not lose amperage. Use another converter to power props, sd cards, USB ports, EEPROMs, and XBee.

    That's a reasonable approach, and I have used it on a few occasions where I have needed multiple voltages. Start with a switcher for the highest and use switchers or linear regulators for the lower ones as needed. Typically that would be 12V for relays or solenoid valves, 5V and 3.3V for electronics, and occasionally another for sensors or analog circuitry. Just make sure the primary supply can handle the total current load.
    In science there is no authority. There is only experiment.
    Life is unpredictable. Eat dessert first.
  • Ok, this approach seems solid. Now, how might I go about protecting the feather’s supply from the charging circuit?
    Returning to Spin after two months of not coding micros at all, forgetting to use :=
  • kwinnkwinn Posts: 8,352
    edited 2019-09-30 - 16:06:07
    Ok, this approach seems solid. Now, how might I go about protecting the feather’s supply from the charging circuit?

    If you are not using a battery you don't need a charging circuit, so no protection required. Since the battery voltage/current also comes to the input of the 3.3V regulator via Q3 the only way the charging voltage could damage the rest of the circuit would be if the regulator shorted out. Very little chance of that happening with the protection circuitry built in to modern regulator IC's.

    If you do want a battery the charging voltage/current comes from the VBUS, and so does the voltage/current input to the 3.3V regulator. That means that you will probably need to provide 4.8-5V if you use a 3.7V battery.

    Personally, I would use a 5V 5A supply (typical Raspi supply) to power this unless I needed the 12volts for something. That would eliminate the intermediate 12V to 5V supply. The simpler you can make something the more reliable it is.
    In science there is no authority. There is only experiment.
    Life is unpredictable. Eat dessert first.
  • I’m trying to mimic a battery and protect myself from the charging circuit because it is part of the Adafeuit Feather’s design, one which I want customers to be able to replace with their stock modules if needed or wanted.

    I can’t change the charging circuit, but if I can’t make a protection circuit between the module and my circuits, then I may need to look into using a different module.
    Returning to Spin after two months of not coding micros at all, forgetting to use :=
  • Thanks Tony! However, I have chosen to use a different module, one without the need for reverse blocking.
    Returning to Spin after two months of not coding micros at all, forgetting to use :=
Sign In or Register to comment.