Solar IC charging a LiPo cell

Schematic used to build my circuit, provided by ST’s web-based design application. The Dout transient diode was omitted in this test.

After some frustrating nights, I finally have the SPV1040 solar charging IC working with my Iowa Thin Film Technologies photovoltaic strip. The PV puts out a nominal 3V at 25mA. Basically, the chip requires two voltage dividers for voltage and current feedback. The ratio of these dividers determines the output.

ST’s SPV1040 solar boost IC along with a 10µH power inductor and a Schottky rectifier.

While I have try and squeeze more out of it, the IC’s 100kHz PWM is able to knock this up to 5.5V at a peak of nearly 3mA. This is enough to power the TP4056, which is designed primarily for USB-powered charging. Given the relative gluttony of the charging IC, full sun is needed with the single PV strip in order to attain charging current. STM’s eval board uses a more efficient LiPo charging IC, which is something I am planning to investigate in my quest for efficiency.

Full sun charging the 3.7V LiPo cell via the TP4056 IC.

Ultimately, I’ve learned that while using alternative methods to solder an IC can be required in a pinch, it is more desirable to use known-working components and boards with the proper footprint.

Advertisements

4 Comments

Filed under Battery power, DIY, ICs, Soldering

4 responses to “Solar IC charging a LiPo cell

  1. Jonas

    Hello Mr.,
    I have the same chip – SPV1040 and it works normally if I connect battery first, but If I want to use this boosted 5V for another (Li-Po) charger, SPV1040 breaks down, even if 5V TVS is mounted. As I understand – boost spike kills internal N-type Transistor (maybe P-type too) and TVS does not protects it from burning down. It is very strange, because I this ST should think about it – using this chip with disconnected battery… Did You have something like this in Your experiments? Thank You for help.

    Best Regards,
    Jonas

    • Hi Jonas, thanks for your comment.
      I don’t quite understand your question. The boost breaks down when you put a load on it? Is it possible for you to post an image of your PCB layout? Perhaps your current sense filter is picking up drops beyond the 50mV threshold?
      I had a lot of trouble with stability on the breadboard, but was able to come up with a functional PCB layout after reading through the application notes (esp. AN3319 pp. 17,19). I hope this information helps
      All the best,
      Chris

      • Jonas

        Hello Chris,
        Thank You for the reply. The main thing, what I want to reach – is to use this chip for Lithium battery charger. So I also need a lot of protections for that – temperature turn off, over-charge protection, discharge protection and etc. That’s why I’m using extra chip for that and spv1040 is used just as simple step-up converter in my scheme. When I was debugging my circuit without second part (used just SPV1040 and load was battery), everythig worked just fine, voltage was stable, current limiting feature was working just fine (experimented with different shunt resistors). But – if my load is not battery(with no load – open circuit) – SPV1040 just brakes down immediatelly (already 3 chips dead, while experimenting in different conditions)… As far as I understand – SPV1040 should start softly – if it does not sence output voltage(it is not supplied by output voltage). But my guess was, that in first startup moments, when internal transistor drives choke(10uH) to the ground and closes down, huge voltage spike is generating and overvoltage occurs on the output (when no battery is used). And this overvoltage is not absorbed by TVS diode (clamping voltage is too high), that is why internal transistors just brakes down – and after braking down – SPV1040’s input impedance is ~9ohms… All 3 dead chips acted the same – now I’m waiting for TVSs with lower clamping voltage. I guess, that, when using battery as load – everything works just fine, because battery can esasily absorb this voltage spike because of low internal resistance.
        I didn’t find where to attach pdf with schematic and pcb, so published it in filedropper :), here is the link:
        http://www.filedropper.com/lisolarcharger

        P.S. Happy New Year 😉

        Best Regards,
        Jonas

  2. Jonas, thank you for sharing a detailed description of your problem with eh SPV1040. After reviewing it and your diagrams, I must admit I am a bit stumped. In my experiments, I did not encounter this problem, though I did fry a couple chips on the breadboard due to transients. In order to simplify layout, I chose a battery charging IC that contained overcurrent and other protections. Although this was my first board layout, I tried to emulate how STM arranged the voltage and current sense components.
    If I to troubleshoot your problem, I might start with the PCB layout. STM states in their STEVAL AN4123 document ( http://www.st.com/web/en/resource/technical/document/application_note/DM00056630.pdf : p.14) that the “output current sense
    circuit (RF1, RF2 and CF1) must be designed as symmetrical as possible, in order to guarantee the noise immunity on the voltage drop measurement across RS1.” Their designs also seem to use a lot of RF-cancelling vias around the board, to compensate for the high frequency switching (see also http://www.st.com/web/en/resource/technical/document/application_note/CD00292052.pdf ). Ideally, access to a DSO in order to capture what is happening when you zap your chips would yield better evidence to guide the troubleshooting process.
    Sorry I can’t help beyond this. You have me stumped! If you figure out the problem, please let me know. I would be interested to hear what was the solution. 🙂
    Happy New Year,
    – Chris

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s