Thoughts on another EE blog

Having got into a browsing chain on EEVblog, I came across EEVblog 262 ‘World’s Simplest Soft Latching Power Switch Circuit’ (right), where David Jones makes a simple latching power supply switch operated by a single push button.

EEVblog 262

He talks it through in an excellent thorough and educational way.

Possible one button power switchPondering this, I wonder if one more transistor could be removed with the following circuit (left).

Although this one might:
Not work at all
Need a noticable gap between operations
Need a small resistor in series with the push button to save its contacts from being eroded




  1. Hi Steve. I remember that ‘steering diode’ bistable – although I can never remember it well enough to draw it without looking it up first.
    I think it always uses power in each state and therefore misses one of Mr Jones’ criteria.
    I agree with the CMOS comment – always the right answer for battery power – no power consumption, nice clean operation, and frequently smaller because no bias resistors.
    Much as I seldom make things, I do try to have a single push button for control, just to keep construction simple.

  2. Re: SCR’s.. yep, neat gadgets, and the basis for a lot of power electronics in the era before big mosfets and IGBTs. I don’t recall what voltages Dave was working with, or the currents. The voltage drop across the SCR could be an issue, and it certainly doesn’t offer the toggle function that Dave was looking for (and it took me a while to realize that the toggle function was one of his goals).

    Re: one button switch multivibrators.. the classic configurations don’t seem to get into a single switch bistable design. I did find a design on the web that fulfills the requirement…..
    (the second schematic)
    …but it is a bit more complex.

    Personally, I’m a bit inclined to just use a CMOS D flip-flop and be done with it. It does need a power transistor to switch the load, and maybe a debounce circuit for the switch. It does require that the voltage source be within the flip-flop’s operating range, of course. The best solution depends on the definition of the problem, of course.

  3. Hi Steve.
    Thanks for originally pointing out the EEVblog
    I think Mr Jones starts off by saying his criteria are zero ‘off’ consumption and power on-off with low-cost switches. And then he goes from two buttons to one by adding that other transistor. I like his circuit – I suspect my alternative might be a bit flaky.
    I am a great fan of SCRs – although they do have a bit of a high forward voltage. There are two of them in a little regulator I am pondering – and, it turns out, even these venerable components come in surface mount now.
    When you say multivibrator, can you switch one with a single button?

  4. I’ve been browsing through Mr. Jones EEVblog episodes myself, and found them to be rather interesting. On the soft-latching switch, I was scratching my head a bit because I didn’t really understand the specs… i.e. what the purpose was, how power was applied and removed, etc.
    I do recall thinking “why not just wire an SCR into the low side of the load and be done with it?”

    The biggest unknown is what causes current through the switch to cease. Then the other questions about how often it will be activated, etc.

    Is Dave’s circuit designed to toggle power?? That seems to be the only reason to have the npn in the lower right corner of the schematic. In light of that, it does appear to be similar to the classic bi-stable oscillator built with cross-coupled transistors.

Leave a Reply

Your email address will not be published. Required fields are marked *