Experiment to investigate the explosive potential of electrolytic capacitors
AJ PENTON, JD HALL, C REED, CE HOWMAN et Al. University of Cambridge. October 1999

Abstract

Electrolytic capacitors are known to be unstable when connected to dc power with reverse polarity. This investigation will attempt to prove the viability of such devices as explosive devices. Thus far a preliminary investigation has enabled the scale of the effect and the precautions necessary to be gauged.

Introduction

Electrolytic capacitors are composed of fine sheets of metal foil separated by an electrolytic dielectric. During manufacture a voltage is applied which "sets" the dielectric, causing it to become a good insulator. As a consequence, such capacitors are polarised and must be connected to power with the correct polarity. Failure to do so causes the dielectric to break down and hydrogen gas to be produced. Either the pressure build up or combustion then destroys the capacitor.

A University lecturer in electronics has described such destruction as generating a lot of gunk and bits of metal, but says that "everyone has to try it once". Upon such recommendation such an experiment is to be performed.

Preliminary

Since the scale of the effect was entirely unknown, a preliminary investigation was performed in order to enable further, large scale experiments to be performed successfully and safely. A 10uF electrolytic was connected to a terminal block, and in turn to a linear dc power supply capable of supplying 1A at 35V. Coloured wires were carefully connected in order to ensure that the capacitor could not mistakenly be connected with the correct polarity.

[Picture]
The capacitor connected and ready to go

A warning signal was given and the power turned on. After a few seconds the capacitor exploded with a pop. Examination of the remains showed that the metal case had been blown off the top, leaving rolled up foil and "fluff" inside. The plastic covering had been destroyed.

[Picture]
The remains of the capacitor

From this it was learned that a 10uF capacitor produces a relatively small pop. This result can be scaled up to predict the size of the explosion that will be generated by a larger capacitor.

Further experiments await the acquisition of some larger capacitors. Once again science is being limited by budget constraints.


Following the generous donation of some very large capacitors to the group, the follow-up work has been completed, and is available here.