Cursed circuits: charge pump voltage halver

In the spring of 2023, when this Substack had only a few subscribers, I posted a primer on Voltage adjustment in electronic circuitsThe article began with a brief discussion of linear regulators, and then immediately threw them under the bus in favor of more efficient charge pumps and inductor-based topologies,

The basic charge pump architecture – a voltage doubler – is quite elegant and easy to understand. This is more common than many people suspect: circuits can be manufactured directly on silicon dies, so it appears inside many digital chips, from modern op-amps to MCUs. In case you weren’t a customer in 2023, or if you don’t have a photographic memory for random blog articles, a conceptual drawing of the pump is shown below:

In the panel on the left, we see court The capacitor that is located on top of the positive rail during a “flying” capacitor CF Charging from power supply. The charging process generates a voltage that is internal to the component: we can unplug CFPut it in our pocket, and then connect it to another circuit to power it for a while.

In the second panel (right), we see the second part of the cycle: CF is disconnected from the supply and then connected to the terminals courtThis action transfers some charge CF To cout, Until the voltage at the terminals of the capacitor becomes equal. After many of these round trips, V_Now should contact v supplyAbsolutely, V_BC is also equal to v supplyThis implies that the voltage between A and C must be the sum of the two, or 2 v supply,

In other words, the circuit is a voltage doubler; repeated motion of CF This ensures that the charge is inside court If we connect any load between point A and C then continuous replenishment is done. There will be a slight voltage ripple, but the amount can be controlled by sizing the capacitors and selecting the operating frequency to match the desired load.

Naturally, practical charge pumps do not mechanically move the capacitor around. Instead, they use transistors configured as switches to connect alternately CF Up to supply and output caps, an architecture that can be depicted in the following way:

The transistors themselves can be powered by a simple relaxation oscillator Or by a programmable digital chip.

A similar circuit can be used to generate negative voltage: we do this by simply hanging court from the negative supply rail instead of placing it on top of the positive supply rail. This modification effectively places the lower terminal of the capacitor -VSupply.

so far so good. But this leads us to a more puzzling flavor of the charge pump – the voltage-half topology, shown below:

You may ask what is that – a capacitor-based voltage divider? Well, yes and no. capacitor can do Can be used as voltage dividers for AC signals: They display Resistance-like effect known as feedbackSo if you have an alternating sinusoidal wave, you can attenuate it this way. As said, the divider doesn’t really work for DC voltages, because at 0 Hz, the reactance approaches infinity.

To understand design, ignore CF and attached load. Let’s focus only on the pair of series capacitors: C1 And C2When these two capacitors are first connected to the power supply, they can be analyzed as a single overall capacitance, with some common charging current that will flow briefly through this circuit branch, especially, if C1 = C2The normal current will produce approximately the same state of charge for each capacitor, resulting in V_Now ≈ V_BC ≈ V supply / 2.

This seems to be the result we are looking for, but once the normal charging current stops, there is nothing to keep the voltage the same. Specifically, if we connect a resistive load to terminals B and C, the bottom capacitor will discharge to 0 V; The reduction in voltage at point B will allow the upper capacitor to charge in a way that will make the difference. A temporary stream will flow, but the final state is V_Now , v supply, V_BC = 0 V, and i’m out = 0 A.

It sounds useless, but this is where the blown capacitor is – CF – Comes into play. If it is moved back and forth between C1 And C2It will charge from the higher voltage capacitor and then discharge into the lower voltage capacitor; In our example, it will constantly replenish the charge C2So that a steady current can flow through the load.

The stable equilibrium for this charge transfer process is reached when V_Now V_BC v supply /2 – so unlike conventional voltage dividers, the output voltage is always at the midpoint between the supply rails, with no dependence on the relative values C1 And C2pretty neat!