Hello, I'm working on a high-voltage power supply design project for a load (10-15 kV). My goal is to control this voltage by first converting it to a 0-5V range to read it via an Arduino ADC. I feel like my circuit isn't isolated or safe enough. Need help please. Thanks!
image|450x202
Can you tell us what current capacity your 10-15 kV supply can deliver? It is unlikely in my book that it has sufficient current capacity to enable voltage dividers to work.
Also as the supply seems to be variable then a voltage divider is no good to you anyway.
As a citizen/public, you are only allowed to work with 220V (110V) source voltage and then use step-down transformer to get low voltage followed by full-wave rectification and filter to get unregulated DC voltage (range: 0V to 4.5V) proportional to 220V (110V).
By no way, you have access to kV high tension line.
Hi, @marie2808
Welcome to the forum.
Your circuit for all to see.
I certify kV level insulation testers.
The safest I have seen is one that produces up to 15kV with a transformer and Cockcroft type multiplier.
The voltage feedback is provided by a scaling measuring network on the output terminals, it is supplied by a DC to DC converter with 20kV isolation.
The monitor network PCB then communicates with the controller via two optical fibre cables.
Can you please tell us your electronics, programming, arduino, hardware experience?
What is the actual project and what is it for?
15kV is not a voltage to be just fed to a PCB and potential divider scaled down to digital 5V level.
Tom....
Hello everyone,
I’m working on a 15kV power supply for an ion thruster. It converts 28V input to 15kV output through two stages (28V → 800V, then 800V → 15kV). Control is done by an Arduino Nano generating PWM signals, which drive power MOSFETs via TLP250 drivers.
Current issue
To measure the output voltage, I initially used a high-voltage resistor divider (100MΩ / 10kΩ) to scale 15kV down to 0-5V readable by the Arduino ADC, with a feedback loop adjusting the PWM duty cycle.
However, I’m concerned about:
- The risk of HV reaching the Arduino if there’s a fault
- Possibly insufficient isolation between HV and low-voltage sections
- Long-term stability and reliability of the HV resistor divider
New approach I’m considering
Instead of scaling the HV directly to 0-5V, I want to use a dedicated industrial HV probe with proper isolation to measure the voltage externally (away from the Arduino). Then, I would adjust the PWM duty cycle based on this external measurement to maintain stable output voltage.
Questions:
- Do you think this method is safer and more reliable?
- Are TLP250 drivers sufficient for isolating the PWM signals to the MOSFETs at high voltage, or should I consider other isolation methods (optocouplers, isolation transformers, etc.)?
- Any recommendations to improve overall system safety without redesigning everything?
Thanks a lot for your insights and advice!
This link may help you: https://electronics.stackexchange.com/questions/246894/how-to-measure-an-15-kv-electric-fence-to-0-5-v-analog-input-of-a-arduino That was for an electric fence, operating at 15kV.
Be sure of your grounds before powering it.
That all depends on the type of HV resistor used, and your build.
A circuit board is likely not good enough for 15KV, because of creep.
Only current can damage an Arduino pin, and 100Megohm alone (if the 10k resistor to ground fails) can't produce enough current to damage a pin. But flashover voltage or creep still can.
The resistor shown in post#9 is a 5KV type. 15KV types are much longer.
Leo..