Shop OBEX P1 Docs P2 Docs Learn Events
parallax propeller coin cell? — Parallax Forums

parallax propeller coin cell?

Hey guys,
Can I power a parallax propeller chip off a coin cell battery? The specific one I'm thinking of using is the CR2032 and it is 3 volts.
Since it's 3V, can I skip the voltage regulator and power the propeller directly until the battery runs out of juice? Or do I still need a regulator?

Thanks!

Comments

  • jmgjmg Posts: 15,148
    You can power directly, but keep in mind the P1 is not by default a low power device.

    The Prop is Static CMOS, and being SRAM based, has wide Vcc tolerance, so you might decide to regulate lower than VBat, in order to have lower system power.

    The HFOSC is default source, but you can switch to the LFOSC and get lower Icc values, but of course much slower CPU speed and less precision.
    The XTAL osc is a linear circuit, and consumes significant Icc.

    See other threads on lower power prop, and lower supply tests.
  • One thread to look at is this
    prop-limbo-how-low-power-voltage-can-it-go

    The CR2032 has 220 mAh capacity, but the current has to be kept low. You can operate most of the time on RCslow at <10µA, then wake up to RCfast or 5 or 10 MHz with occasional brief pulses of current up to 10mA, better <2ma, if necessary. Helps to put a capacitor in parallel with the cell, because the internal resistance of the cell can be quite high, especially at low temperatures and end-of-life.
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2018-01-15 22:31
    You will probably need to tie the brown-out pin high as otherwise with BOE normally low the Prop would reset if the supply dips below 2.7V. As Tracy mentioned you will need to watch the peak current as the battery voltage may dip too much and I'd recommend a high value ceramic if you are using SMD and/or tantalum cap of around 22uF roughly. Electrolytic caps leak current and do not have as good "a response" to peak current demands etc.

    Using a coin cell will probably rule out full speed operation and/or full cog utilization. A simple program running at 10MHz will still draw around 10ma. If you use a lower frequency crystal than the Prop PLL's is specified for then you can only use RCSLOW/RCFAST and XTALx1 settings for your clock. Personally I'd just use a standard 32.768kHz crystal for timing accuracy from which I could calibrate RCFAST if I needed that.
  • jmgjmg Posts: 15,148
    .. If you use a lower frequency crystal than the Prop PLL's is specified for then you can only use RCSLOW/RCFAST and XTALx1 settings for your clock. Personally I'd just use a standard 32.768kHz crystal for timing accuracy from which I could calibrate RCFAST if I needed that.

    Prop does not have a 32kHz oscillator drive setting, so the OP will need to add a Oscillator, and cheapest way to do that is probably a RTC like PCF85063 (or a small MCU with a 32kHz Xtal mode)
    I have persuaded HCU04 buffers to oscillate at 32kHz, using larger Rs, but the Icc at 3V is never great. Proper 32kHz really needs a uA-drive Amplifier+buffer, designed for kHz use.
    Prop curves stop ~ 4MHz, - you can get ceramic resonators below that, but crystals under 4MHz get rapidly larger and more expensive.
    Resonators have low accuracy, and I see even there, 2MHz is the lowest compact SMD offering. 455kHz/500kHz are bent-leg thru hole 'smd'

    Re Lower voltage Xtals, I see this post
    http://forums.parallax.com/discussion/comment/1251164/#Comment_1251164
    says
    This morning I was able to run from a 2.4576 MHz crystal down as low as 1.53v. Below that the oscillation appears to die and supply needs to be brought above ~2.2v to recover.
    Edit: Was later able to get a 1.84 MHz crystal to operate down to ~1.55v.

    So crystal oscillator startup seems the limiting factor, ~ 2.2v
  • mikeologistmikeologist Posts: 337
    edited 2018-01-16 15:19
    I'd use a CR2354 as it delivers 600mAh and a pulse current of 15mA. That will give you more power in reserve and it's only a bit thicker.
  • Tracy AllenTracy Allen Posts: 6,656
    edited 2018-01-16 22:15
    A nice thing about the CR2032 is that it is ubiquitous and cheap (unless it's a rip-off from a corner drug store, but at least they'll have it.)

    For reference, here's my chart of current consumption, a Prop running SPIN on various clock settings, one cog, either in a wait state, or doing simple math, or toggling one pin.


    662 x 393 - 75K
  • jmgjmg Posts: 15,148
    For reference, here's my chart of current consumption, a Prop running SPIN on various clock settings, one cog, either in a wait state, or doing simple math, or toggling one pin.
    Nice numbers...

    These show a consistent slope of appx
    (1.434m-1.212m)/(5*2-5*1) = 44.4uA/MHz
    (1.897m-1.212m)/(5*4-5*1) = 45.67uA/Mhz
    (2.814m-1.212m)/(5*8-5*1) = 45.77uA/MHz
    (4.641m-1.212m)/(5*16-5*1) = 45.72uA/MHz
    and that indicates the CMOS-Logic portion of the 5MHz-Xtal power, is (1.434-1.212) = 0.222mA, leaves 294uA for the Crystal Oscillator Amplifier+Buffer power
    equivalent Power Dissipation capacitance, looks to be ~ 18pF



  • Tracy AllenTracy Allen Posts: 6,656
    edited 2018-01-17 18:24
    For low power operation, account for the time taken at each clock speed, including the time taken to change clock speeds.

    (time@RCslow * 10 + time@speed * µA@speed) / (totalTime) = average µA

    Transition from RCslow to xtal 5MHz (no pll)
    ---SPIN: 67ms
    ---PASM: 3.4ms (crystal dependent)
    Transition from RCslow to RCfast ~12MHz
    ---SPIN:33ms
    ---PSAM: 0.83ms
    switching-from-20khz-rcslow-to-xtal1-5mhz-or-rcfast
  • RaymanRayman Posts: 13,898
    edited 2022-03-06 17:40

    @"Peter Jakacki" said:
    You will probably need to tie the brown-out pin high as otherwise with BOE normally low the Prop would reset if the supply dips below 2.7V.

    Just looking at Propeller Manual and Datasheet and don't see anything that says that tying BOEN high will prevent reset below 2.7V.
    It just says that it changes the nature of the reset input. Is this the same thing?

    Looks like I'd need to pull reset high with an external resistor if I tie BOEN high.

  • Tracy AllenTracy Allen Posts: 6,656
    edited 2022-03-07 01:16

    When BOE is tied high, the reset pin is a CMOS input with hysteresis, so you'll need a pullup or an RC circuit. https://forums.parallax.com/discussion/comment/1033397/#Comment_1033397 has graphs showing what happens below reset with BOE disabled. The entire thread "prop-limbo-how-low-power-voltage-can-it-go" is rife with low-power info.

  • RaymanRayman Posts: 13,898

    @"Tracy Allen" Thanks, that's good info.

    I am wondering what is the best way to handle BOEN. Since it's a CMOS input, seems could just tie directly to battery.

  • Yes, just tie BOEn direct to the battery +Vdd to disable the 2.7V reset threshold and allow operation at lower voltages.

Sign In or Register to comment.