I have a high voltage power supply which can power up to 30 kV with respect to ground. However, for my experiment (something like the one used in photocopier machines) I need a high negative DC supply. So how can I convert a positive let's say 20kV into -20kV. I'm a material scientist with a limited knowledge about electrical and electronics machines.
Electronic – How to convert a very high positive DC voltage to negative DC voltage
dc/dc converterhigh voltageintegrated-circuitinverting-amplifierpower supply
Related Solutions
The outputs are almost floating with respect to ground but a tiny feedback current both +/- will show the voltage with respect to ground as an offset when only loaded with a 10MOhm probe.
My guess is that is normal for your old model and wont have any impact in most applications. I suspect newer models improved that leakage. But this wont stop you from using tracking mode or Independant or CV , CC or same or complementary polarity earthed at any one point.
Typical use in the old days was separate Analog and Digital grounds wires but the complementary 0~24V outputs grounded at the PSU only with the 5V logic ground, so they do not share logic current noise.
In theory a buck converter can produce any output voltage between the input and ground, while an inverting buck-boost converter can produce any output voltage below ground.
The relay approach
As the two topologies are very similar, you could switch between them using a DPDT relay. While not very elegant, this incurs only a modest increase in part count and board area. Unfortunately, you do need a custom controller for controlling this, but any simple microcontroller with an ADC would likely do just fine.
simulate this circuit – Schematic created using CircuitLab
Solid state
I later came up with a solid state (relayless) four quadrant variant, which can source and sink current at both positive and negative voltages, while maintaining the same ground reference.
Just like the relay circuit, this topology combines a buck converter with a buck-boost converter.
The benefits are instantaneous switching between step-down and inverting mode, the lack of moving parts, and the more compact size.
The main drawback I see is that the gate drivers of M3 and M4 must be able to swing their outputs below ground, while the digital control inputs are probably going to be between zero and a logic high (e.g. 3.3 V). This complicates the selection of gate driver chips.
Best Answer
Yeah seriously, the comment on your question is actually the right answer. Assuming the power supply outputs are isolated/floating (I'd be shocked if such a high voltage supply didn't have floating output -- and you would be too), then the easiest way is to connect the negative terminal of your supply to thing you want to energize to -20kV and your positive lead to whatever your reference is. The power-supply does what it always does and attempts to force 20kV between the positive and negative terminals. Voltage is always a differential measurement, and with the floating the supply you can choose your reference to be whatever you like.