The circuit demanded for a Real Time Clock with I2C. A DS1307 seemed a smart choice. It is powered with +5 V. As the datasheet specifies a battery voltage from 2 - 4 V, the initial design used 4 x LL4148 (because of the low leakeage current), in series, to get rid of approx 2 V and therfore land in the middle of the specified range. The charge is stored in a "gold cap".

The circuit worked nicely after switching on. But, after some days the clock seemed unhappy.

The reason is the following : The voltage over the storage capacitor will keep on climbing (up to 5 V), as the current drawn from the gold cap is very low - and for very low currents, the voltage drop across the diodes is also very low. Zero volts at zero current.


The voltage over the gold cap climbs ...

As the mechanism - which leads to malfunction - was clear, the solution was also clear. With only one diode, but this time connected to the +3.3 V, the voltage above the gold cap can never go above the +3.3 V - with a 1N5711 (Schottky Diode) it settles at about 3.0 V - just the value we aimed for ...

The webmaster does not read these comments regularely. Urgent questions should be send via email. Ads or links to completely uncorrelated things will be removed.