#257 Emergency

Pimp the Boldport Club Emergency kit (Project #2) .. now a LED blinky with 3 passives and a transistor (Esaki Effect)

Here’s a quick video to show it working..

Notes

I joined the Boldport Club rather late, but manage to snaffle up the Emergency kit PCB in sale of “Just Less the Perfect” boards.

The tiny ‘engineer superhero’ emergency kit, second edition was Boldport Club Project #2.

As originally intended, it is a kit with a few components and a nice travel case that will possibly save your engineering superhero status at some point!

With all the components soldered into the (of course beautiful) PCB, it works as a battery/continuity tester (the LED lights up if you apply voltage)

Emergency_board_front

Emergency_board_rear

Circuit Modifications

Since I’m working from scratch (PCB only - not the full kit), there’s an obvious challenge to do something a bit different.

@mehradzie asked on Boldport Slack if it is possible to make it blink. Challenge accepted!

I was actually thinking along similar lines, at first using a blinking “joule thief” circuit. But I just couldn’t figure how to get it to work within the constraints of the PCB.

So then I turned to an interesting idea: Negative Differential Resistance (NDR) effect or Esaki Effect. Basically causing an Emitter-Collector avalanche, a behaviour noted by Leo Esaki and exploited by many since to create interesting effects with BJTs.

After a bit of experimentation, I was able to get something to work with only a few liberties taken with the original PCB:

I had to accommodate larger capacitor. Anything over ~100µF is workable, but best effects achieved with 330µF electrolytic. I selected the smallest package I could find, but it is still a bit of a squeeze to fit it in where the capacitor should be.
an NPN BJT replaces the n-channel MOSFET. I’m using an S9013. Note: yes, this is in “backwards” (emitter to +ve, collector to -ve)
moved one resistor to make wiring easier
cut and rewired a few traces

Transistor selection is crucial, as the blinking effect is really exploiting a “defect” in the transistor. Of all the small-signal NPN BJTs I had on hand, only two worked at all, or at reasonable voltages (S9013, S9018). I went with the S9013 because I could excite the effect conveniently around 9V.

The revised circuit essentially works like this:

battery charges the capacitor
over a certain voltage, the emitter-collector junction experiences an avalanche and inrush of current from the capacitor
the LED turns on
as the capacitor discharges, the avalanche breaks down and the transistor junction closes
repeat…

Here is a scope capture from a breadboard build/test: