# Why don't we make negative output voltage like this?

Here you have a buck conveter where Vout1 = D*Vin which is positive for positive input voltage.
I'm wondering why don't we just reverse the connection of output like Vout2so we can get a negative buck output instead of using inverting buck-boost converter to generate negative output?

• Because usually you want your GND to stay your GND. If your entire circuit operated off a negative voltage then sure you could do that, but then you could do the same with any voltage source and any regulator. It would just mean your entire circuit uses the more positive voltage as the reference rather than the most negative (which is usually the case for unipolar circuits). Commented Jun 26, 2022 at 20:51
• @DKNguyen thanks, why would you want to ensure that? Commented Jun 26, 2022 at 20:52
• Because there's almost always other stuff in your circuit that is operating off a positive voltage and is using that GND as the reference. Commented Jun 26, 2022 at 20:52
• @DKNguyen if you have any circuit in mind could you post it where it is not possible to reverse the connection like I did above? Commented Jun 26, 2022 at 21:08
• For modest power requirements, and for single-digit voltages, the switched-capacitor method used by ICL7660 can generate -ve voltages from +ve input...GND stays GND for both +ve/-ve supplies. Commented Jun 26, 2022 at 21:38

You're on the right track for one fairly common method of generating a negative output voltage. Take a look at the circuit below:

simulate this circuit – Schematic created using CircuitLab

The main downside to this is that the controller is no longer ground-referenced, so unless it's designed to be used this way and has level-shifters built in, you'll need to level shift any signals that go between it and the rest of the system. Another downside is that the controller needs to be rated for Vin-Vout (a higher voltage, since Vout is negative) instead of just Vin. But if you can get past those hurdles, it's an effective way of generating a negative supply from a positive one.

Note the difference between this circuit and yours: in this circuit, the control IC and the low-side switch are no longer connected to ground, but to the output instead.

• Thanks, interesting to read. Commented Jun 26, 2022 at 22:22
• You haven't identified where the other end of Vin (Vin-) goes. To ground or Vout (GND). I know the answer but others may not. Commented Jun 26, 2022 at 22:58
• @Marla I wondered about that too. Is it connected to GND? Commented Jun 26, 2022 at 23:16
• @internet It's connected to ground. Sorry, let me edit to make that more obvious. Commented Jun 26, 2022 at 23:58

Your idea is correct, in that the polarity of the output of a single supply just depends on which one you "call positive". You could connect any component or circuit to the output to get one polarity, and simply flip the output connections to get the opposite polarity.

Problems arise when you try to combine buck converters (or any type of power supplies), to obtain dual, or multiple voltages, some positive, some negative.

Start by noticing that most DC-DC converters, like the one you show us, have a common connection between input and output. Usually it's called ground, or 0V, or "negative side". Your own circuit shows this as a wire connection directly from input-side ground to the output side.

You might think you could power two buck converters from the same source, and connect their outputs in series, to form a dual "+5V, 0V, -5V" supply, but what you would actually be doing is this:

simulate this circuit – Schematic created using CircuitLab

Following the blue path, you can see that I've accidentally short-circuited the output of the lower buck converter.

When combining supplies in series to obtain a negative voltage with respect to some common ground, you run into problems if the supplies' outputs aren't completely isolated (on both positive and negative sides) from the inputs.

The designer of a negative DC-DC converter might solve this problem by ensuring that it is the positive side that is common to both input and output. In that way, when combined with a normal buck converter (by connecting their outputs in series), that output short-circuit doesn't occur:

simulate this circuit

Perhaps the most important thing to notice in this last diagram, is that the lower converter produces a -5V output when provided with a +12V input. This implies that its output is not merely a reduction in voltage magnitude. If you took a voltmeter to measure the difference between the battery positive and the -5V output, you would measure 17V difference!

The change in sign is not just cosmetic. The negative DC-DC converter is producing a voltage difference which is not just 12V to 5V, it's +12V to -5V, a difference of 17V. It allows designers to declare a common ground, and produce voltages below that 0V level, even when the source is greater than zero volts.

• Note that the situation shown in your first circuit is not the same as the one shown in my answer. It's subtle, but by having the buck converter be "powered by its own output" in a way, you can break the short circuit. That is, no connection to ground on the input side at all, the power taken between the input positive and the output negative. Commented Jun 27, 2022 at 5:50
• I'm aware that you probably know this, but I want to point it out to avoid confusing the asker! Commented Jun 27, 2022 at 5:50
• Interesting info. I did think about using two converters like that to make negative voltage like what we usually do with two battery. Commented Jun 27, 2022 at 18:18