I'm using a voltage detector for a solar powered ESP8266 project. Its purpose is to make sure that if for some reason the battery goes below a certain value (around 2.7v) then the MCU goes into "zombie" mode and does not power up as needed again. So I tried to use a voltage detector.

According the datasheet (see Test 5 figure)- there is a resistor RL which noted at the beginning as 200 ohm, and a capacitor, whose value is not noted anywhere.

When using without RL and capacitor, erratic behavior after sleep reboot occurs causing it to it to fail (MCU's led flickers very rapidly.)

Is there a rule of thumb for the capacitor?

**Edit 1 Circuit: ** Circuit

Although tried several combinations of R1(200,1K, 10K) and C1(3.3 to 1000uF), I still get the following behaviour :

  1. Dropping down input voltage from 5.5v to 3.3v works as expected (ON ), while going below 3.3v - cutting off the voltage to merely 0.1v ( measure at RST ).

  2. Going back up ( 0v -- ~2.5v ) - measures 0.1v anf then, until ~3.8v (!!) MCU's boot led flickers, while measures at RST pin rises from 1.6v at some point to 3.0v ... and when Vin reaches 3.8v RST pin is arround 3.3v.

  3. when MCU is not connected, going up & going down behaves as expected.

  • \$\begingroup\$ solar cells are current sources, so a std comparator with a capacitor load will oscillate near threshold. this effect is unavoidable but can be reduced with a battery or an ultracap across input and use some hysteresis with positive feedback to Vin- = show your actual cct and specs for power and load \$\endgroup\$ Commented Dec 27, 2020 at 17:00
  • 1
    \$\begingroup\$ Not noted anywhere? You mean the capacitor CL from output to ground, which is mentioned under test conditions where CL is 100pF? \$\endgroup\$
    – Justme
    Commented Dec 27, 2020 at 17:04
  • \$\begingroup\$ @TonyStewartSunnyskyguyEE75 this mentioned behavior occurs with battery only ( prior to connection of solar power ). \$\endgroup\$
    – guyd
    Commented Dec 27, 2020 at 17:14
  • \$\begingroup\$ @Justme fig 5 does not say, other gifs show 0.1 micro or 10 micro, so I'm not quite sure \$\endgroup\$
    – guyd
    Commented Dec 27, 2020 at 17:15
  • 1
    \$\begingroup\$ it is hopeless to analyze every fault without a detailed schematic and specs. e.g. what kind of battery& at what SoC. all batteries have memory, ie hysteresis and thus allows astable oscillation like a Schmitt trigger oscillator. so your comparator must have more hysteresis than the battery in that SoC, .2.7 is pretty lowV and might be high ESR. \$\endgroup\$ Commented Dec 27, 2020 at 17:30

2 Answers 2


The output of the KA75330 is an open collector - it can only "pull down." You must have a pull up resistor (RL) in your circuit.

enter image description here

Some microprocessors have a pull up built in on the reset pin, some don't.

You'll have to see if your ESP8266 module has a pull up on the reset pin, then you have to make sure that your reset signal (from the KA75330) meets the timing requirements of the ESP8266.

enter image description here

You'll have to use the value of your pull up resistor and the needed timing to figure out a capacitor value.

You'll need to use the RC time constant equation.

$$ \tau = R \times C $$

Where \$\tau\$ is in seconds, \$R\$ is in ohms, and \$C\$ is in farads.

The "Electrical Characteristics" table on page 2 gives the \$R_L\$ and \$C_L\$ values used in the test circuits. Look in the "Test Conditions" column.

enter image description here

You could try them and see if they are adequate for your needs.


Components of figure 5 depend mostly on the MCU requirements, so that is the reason the values are not given.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.