# Controlling high voltage devices with Arduino Relay module

Goal: I want to build a circuit that converts 700V DC supply to 700V pulses of 15ms duration.

From what I have learnt from online, is that Arduino controller can be used with Relay module to control such high voltage. But the voltage is limited to 250V AC or 20-30V DC.

Question: Is there any specific relay module that can control such high voltage (700) that can be controlled with Arduino. If not any other ideas are appreciated.

(Update on the question, adding details)

I am trying to do electroporation for zebra fish embryos. The details can be found here. For that i need two pulses of 0-700V of 15ms duration which will create 3500V/cm electric field. My lab has already bought Trek Model 2210 as DC source which will vary from 0-1000V. I have to design a pulse generator which will generate only two pulses of 15ms duration and 700V. More than two pulses will destroy the cells. My plan is to

Arduino----Buffer circuit---relay/power MOSFET (will be connected with 1000V source)---cell

When the Arduino will provide voltage to relay it will switch on and supply the voltage 0-1000V. How long it will stay on will be controlled by ARDUINO.

However, i am not getting any relay which can handle such high voltage. I am open to suggestions how i can solve the problem.

• solid state relay, or high voltage opto-coupler. probably not available as a module. – Jasen Feb 27 at 6:25
• please eidt you question to include how much current you need to switch. – Jasen Feb 27 at 6:33
• You would need a 1200V IGBT or SiC gate driver + transistor, of course. It's a very expensive module. – Marko Buršič Feb 27 at 7:45
• see addition to my answer. This is an example only. Do you need complex waveforms or simple on/off or ...? – Russell McMahon Feb 28 at 14:27
• See further additions to my answer. – Russell McMahon Mar 1 at 12:55

The voltage switched is completely unrelated to the switch controller, whether an Arduino or other device. The switchable voltage and the control voltage are characteristics of the relay or other switching device.

There is no reason why a 700 V rated relay should not exist, but, a quick look at Digikey's (a large electronic component reseller) catalog shows that they have no 700V or above rated relays in either mechanical or solid state versions.

While you could investigate other suppliers, this is a very good indication that what you are trying to do is "not a good idea". The problem is that you have told us HOW you wish to solve a problem . but not what the actual problem is.
"Tell us what you want and we'll tell you what you need".

What are you trying to achieve?
What is driven by the output voltage?

It is likely that a "solid state" solution will meet your need - once we know what it is.

Digikey sells MOSFETs rated at 1000V and low current for under $1 and devices rated at 2500V and several amps for under$10. Whether any of these meet your need depends on what the need is.

Update:

This is a non trivial application and circuit complexity depends on whether you require simple on/off control or more complex waveforms. I'd expect that the electroporation community would have addressed the requirement for lowish cost semi-DIY power sources. Digikey have a number of high voltage IGBT devices suited to switching the voltages involved. Circuit design is liable to neither be rocket science nor trivial.

An example IGBT is the IXYS IXBH2N250 pricing and datasheet. Rated at 2500V, 2A it seems a reasonably likely basis for a switch. There are a reasonable number of others.

This thesis appears to be highly relevant

Design of a MOSFET-Based Pulsed Power Supply for Electroporation by Jason R. Grenier -- thesis University of Waterloo Master of Applied Science Electrical and Computer Engineering

The 2SK3748 MOSFETS are obsolete but still available from ebay and amazon suppliers.

I include the following two circuits from the thesis to show how notionally simple such a circuit may be. Neither circuit shows any means of output voltage monitoring or waveform control. If a relay would have been adequate then a circuit of this degree of complexity may suffice. Other MOSFETs that suit this circuit are available.

This paywalled paper sounds ideal. - They claim a \$60 cost DIY device.
Their voltage is limited to 300 Volt maximum but this should be able to be easily extended.

Design and evaluation of an affordable and programmable mobile device, capable of delivering constant current and high voltage electric pulses of different waveforms for biomedical and clinical applications

• Thanks for the suggestion @Russell McMohan. I have updated my question and my need in details. – Nusrat Feb 28 at 5:38
• @Nusrat The answer is good, since it includes a scientific article based on the problem. BUT: I would still use galvanically isolated gate driver (CRD-001 1700V SiC) for your safety and then connect MOSFET in the same arrangement as described in the article. – Marko Buršič Mar 15 at 9:36

The article explains also two different type of pulses: square and exponential decaying pulse. This can be done with a half bridge.

For square both transistors are used:

• hi side off, lo side on - zero voltage
• hi side on, lo side off - for a pulse duration (hi and lo side signals do need dead time in betwen to prevent trough conduction)
• hi side off, lo side on - zero voltage, rapid transition from HV to 0V

For exponential decay, the low side transistor remains off all the time.

The Arduino is a toy MCU, it does not have dedicated timers for PWM control of a H-bridge with dead time. And Arduino is dead slow compared to ARM MCU for motor control. Moreover, the entire solution could be realized with FPGA or CPLD or a hybrid. The IGBT gate driver could be very costly and who knows how many IGBTs you gonna burn before making it work.

This is a very cheap board with a hybrid MCU+CPLD (DRSSTC SKP) it's a Tesla coil driver. It has a fibre optic trigger signal and it is a complete lo and hi side driver for pulse transformers. Unfortunatelly I think it won't work for your application, because it is made for high frequency pulses. If you need 15ms pulse, well that's quite long pulse. You woud get the best answer on some 4HV.org or Tesla coil builder's forum.

If the budget is not a problem then you should look to aquire a complete half bridge IGBT and gate driver circuit.

The cheapest board from reputable manufacturer that I have found is:

https://www.wolfspeed.com/power/gate-driver-boards/cgd15sg00d2

It's an eval module for CREE Sic MOSFET. You would need two pair of these gate drivers and a pair of 1200V Sic MOSFETS. Extra you gonna need a 12V PSU sharing the same Arduino GND.

Find some Aduino example to implement HI/LO side switch with dead time. Before testing you should also put a ballast load, it would prevent frying transistors if your Ardino program fails to output the correct sequence. You eliminate the ballast once you have good results.

simulate this circuit – Schematic created using CircuitLab

And here it is all in one evaluation board, if you want to stay out of shallow water Half bridge SiC :

A pulse with exponentnial decay can be made using only the high side switch:

simulate this circuit

• This started out well, but got diverted into false statements. The Arduino's ATmega most definitely has hardware timers capable of doing this, and quite a bit more. People drive three phase BLDC motors with them, while the question needs to generate a pulse on a single channel and then maybe counteract it. What's more the time scale would make a pure software approach viable, too. – Chris Stratton Feb 28 at 9:25