Computing stuff tied to the physical world

Finished Low-power Supply

In Hardware on Dec 28, 2011 at 00:01

The Low-power Supply described previously on this weblog now has a PCB – and it’s about as small as a JNµ:

DSC 2827  Version 2

Here’s an assembled unit, ready for testing and hooked up to power a JeeNode (w/ disabled bootstrap):

DSC 2831

And here’s the whole test setup I had to create to check this thing out:

DSC 2830

Lots of stuff involved, including high-voltage probe and 230V isolation transformer (to the right, out of view).

Here’s a demonstration of how it works – summarized as one elaborate scope capture:

SCR42

Green is the mains voltage (235 Vrms), purple is the charge building up on the 100 µF reservoir capacitor, yellow is the regulated output, and blue is the JeeNode’s current consumption (measured as voltage drop over 10 Ω). Note how some of the voltages measured here differ more by than four orders of magnitude!

Anyway, that zoomed-in image is the clear signature of the second 8-byte RFM12B packet transmission. Current consumption varies from 23 to 26 mA. It’s a relatively coarse image, since it has been zoomed in 4,000 times.

The 3.3V supply level is reached ≈ 2s after power-up, with another 2.5s needed to fully charge the reservoir cap. You can see from the purple dips that this supply could sustain at least one packet transmit per second.

No surprises here, but it’s good to see that the PCB design works as intended. Next step: implement deep sleep on the ATtiny84 – hopefully this’ll take just some minor adjustments to the Sleepy::loseSomeTime() code.

  1. Very nice little board there JC (no comment about the scope, just gritted teeth and a nice shade of green).

    I know your target was minimal current consumption, which you have achieved superbly, but could the current output of board be increased with some component changes (obviously at the expense of efficiency) to make it capable of supporting a JeeNode on constant receive with occasional transmit bursts? Maybe by increasing the value of the X2 and/or placing a bridge rectifier on the input?

    If so it would make a far more compact solution to one of my projects which is using those quaint old bulky transformers.

    • You’ll need quite a bit more current (15 mA just for receiving, i.e some 50x as much). Probably need to lower the zener to say 6V (not much lower, they don’t work well). Cap 50x bigger (0.47 µF, X2!). The fusible resistor will need to be a bit smaller and higher wattage for startup peak – maybe 2.2 kΩ 1W (just guessing). The cap will be a lot bigger, physically. Probably 50x as much heat, i.e. 0.6W continuous power consumption.

      This supply really only works because the attached JN/JNµ is in microwatt mode most of the time.

      The other option, which I’d like to explore further at some point is the LNK302 approach (see this weblog post). That would support permanent receives and a small relay.

      Yet another option, which you’ve no doubt considered, is repetitive polling once or twice a second, but there too it’ll take some energy to wait for the ACK + info (still if the total is 10ms, that would be just a 1..2% duty cycle).

  2. Chapeau ! One carefully crafted picture tells so much.

    Tankslappa, limited scaling up is possible, but some of the component choices are finely balanced. The effect of a larger X1 on static conditions drives a higher useable consumption, but you need to look at some worst-case scenarios to compute the ratings of other components.

    For example, for the input surge when connecting the mains near a crest, X1 discharged and the reservoir “full”. X1 looks temporarily like a short circuit, so the fusible resistance value is chosen to limit the stress on the zener. As the X1 is increased, for efficiency, the fusible value should decrease, requiring a higher rated zener.

    You are right, substituting a bridge rectifier ~doubles the available coulomb draw, the only “penalty” is losing the output reference to the neutral input wire. In this case, halve the fusible value and put one in each L1/L2 leg. Other components can remain the same.

    A typical application locks the supply away in a wall box, directly mains connected (i.e. to a circuit fused at several AMPS). Hence fail safe after surviving the general ugliness of raw mains supplies for 1,000′s of hours.

  3. Oops – 4am blur :-(

    For X1, read X2/Y1….

  4. Great work! Are there any plans to make this PCB available in the shop?

  5. What changes would be required for 120V operation in North America? Awesome blog, btw.

    • Changing the 10 nF cap to 15 or 22 nF should be enough. The current setup appears to work down to 150 VAC @ 50 Hz. But at 60 Hz the cap’s reactance will be 20% less, so it might even work as is.

  6. Have we persuaded you to put this into the shop?

Comments are closed.