High Voltage Measuring

I am trying to measure a high voltage on an Arduino (Nano). The voltage range is 400-850V. It is for charging a battery so the rate of voltage reading is not important (like once every 30 seconds at worst). I am also not very concerned about the resolution so the ADC on the Arduino is fine, but I would prefer the measurement to be isolated. Are there standard circuits/components that convert and isolate high voltage to 0-5V?

kwithlub:
I would prefer the measurement to be isolated.

Yes, I'd say that keeping away from 800V is to be "preferred".

It's why a pair of Cat I Fluke probes alone cost more than many complete meters.

I've never measured voltages that high but if I wanted to I would build a measuring circuit that was not isolated from the HV. I would put that in some kind of isolation box, or better still one box inside another. I would have this circuit send the data by some safe means such as Wi-Fi or optical fibre to the actual logging electronics.

My mains voltage and frequency monitor electronics is direclty connected to the mains and is in an earthed aluminium box. It is powered through isolated DC DC converter with a rated isolation of 1500V and the out out is via opto-isolators.

Hi,
What is the application that has a battery that operates from 400V to 850V?

What is the application?

Thanks.. Tom... :slight_smile:

A Google of DC Voltage Transducers should get you plenty of hits like this one:

https://www.ohiosemitronics.com/categories/dc-voltage-transducers?gclid=Cj0KCQjwka_1BRCPARIsAMlUmEowKTByfuEjqlSsAeSDxMp9GDmOSA4aDsRpGDofSWSaw_sa2WVM4OYaAgdDEALw_wcB

I have used Ohio Semitronics and CR Magnetics transducers in the past with good results. Most transducers like this also offer you outputs of 0-1mAdc, 4-20mAdc, 0-5Vdc, 0-10Vdc and of course isolation. Other companies also manufacture the same type devices.

Ron

I like the idea of isolating the high voltage in a box and safely transmitting the signal elsewhere. For my application, I plan to control the charger from about 30ft away. So I will have one arduino next to the charger (with a simple voltage divider feeding an analog input) and another arduino in my control box. I would like to use the TX and RX pins for communication but of course a wired connection would eliminate the isolation.

Fiber optic seems like a good idea and I started looking for a (cheapish) breakout board that converts the TX and RX pins to an optical transmit and receive but have come up dry.

Would I be able to use something like this Photo-link Light Transmitter Unit and this Photolink- Fiber Optic Receiver to communicate safely through a fiber optic cable? Am I reading the datasheet wrong or should I be able to simply put a transmitter and receiver on the TX and RX pins of each arduino?

Would I be able to use something like this Photo-link Light Transmitter Unit and this Photolink- Fiber Optic Receiver to communicate safely through a fiber optic cable? Am I reading the data sheet wrong or should I be able to simply put a transmitter and receiver on the TX and RX pins of each Arduino?

My reading of the data sheet is that you can do exactly that. Note that I have not read it in detail and I have not used these myself. Are they cheap enough to buy and play around with?

You still need to solve the problem of getting power to the measuring circuit.

The fiber optic cable I have used all have metal sheathing for protection of the fiber.

Paul

Digikey sells them at $1 for the transmitter and $1.84 for the receiver so that's a nice price.

I decided to go with a digital isolator. This way I don't have to worry about fragility of bending fiber optic cable and I only need the one chip.

As far as powering the measuring circuit, I have some Meanwell power supplies that take 110-240VAC and convert to 5VDC. For measuring the high voltage battery, I am thinking of two inexpensive voltage divider ideas:

  1. High wattage resistors in the kOhm range and I deal with the heat loss (not a huge deal since there is >1.5kW flowing into the batteries so a little energy waste isn't a huge deal)
  2. Higher resistors in the MOhm range. The limit here is getting current to fill the ADC cap for getting an accurate reading. But if I only measure once or twice per second and use a cap between the analog input and ground I should be fine. I haven't done the math yet but this should work.

kwithlub:
I like the idea of isolating the high voltage in a box and safely transmitting the signal elsewhere. For my application, I plan to control the charger from about 30ft away. So I will have one arduino next to the charger (with a simple voltage divider feeding an analog input) and another arduino in my control box. I would like to use the TX and RX pins for communication but of course a wired connection would eliminate the isolation.

Hi, what charger and how will you control it?
What is your overall project?
Can you post a block diagram of your concept?
What are you batteries?
Can you tell us your electrical, electronics, programming, arduino, hardware experience please?

Thanks.. Tom... :slight_smile:

I decided to go with a digital isolator. This way I don't have to worry about fragility of bending fiber optic cable and I only need the one chip.

They will do the job. You need a slot in the PCB beneath the isolator so the current can't creep across from one side to the other.

I don't think fibre is particularly fragile, but as Paul warns make sure you use fibre that does not have metal in the sheath.

Digikey sells them at $1 for the transmitter and $1.84 for the receiver so that's a nice price.

At that price I'd be tempted to buy some to play with!

  1. High wattage resistors in the kOhm range and I deal with the heat loss (not a huge deal since there is >1.5kW flowing into the batteries so a little energy waste isn't a huge deal)

Ugh! To power my mains monitor I used these (or similar) Isolated DC DC converter

Higher resistors in the MOhm range. The limit here is getting current to fill the ADC cap for getting an accurate reading. But if I only measure once or twice per second and use a cap between the analog input and ground I should be fine. I haven't done the math yet but this should work.

The limit is not what you think it is. If you have, for example, a 1MOhm resistor and a 10kOhm resistor in series then the source impedance at the junction is not 1MOhm or some such, it is 1MOhm in parallel with 10kOhm, so 9900Ohms, this assume that the source impedance of the supply is low enough to ignore. In any case, for your application I doubt there is any need to read more than once per second or so.

Paul_KD7HB:
The fiber optic cable I have used all have metal sheathing for protection of the fiber.

Paul

Black-jacketted 1mm plastic fibre doesn't conduct electricity and its short range is a non-issue for this kind of application - the datasheet for the Photo-link explicitly mentions 1mm plastic fibre, which is robust and doesn't
require protection.

Something wrong here - 800v to charge a battery ??

This a bit specialist

hammy:
Something wrong here - 800v to charge a battery ??

This a bit specialist

Sounds like it is a grid connect, or power wall unit, but the OP will not give us and answer to post #3 and post #9.
Tom.... :o :o :slight_smile: :slight_smile:

The project is creating a trickle charger for an onboard large lead-carbon battery system. It is a temporary solution until the industrial charger is available. There is already an onboard battery management system which can transfer power between batteries but there isn't a way to add power to the system. I have a couple Meanwell CSP-3000-400 (3000 watt 400 volt) in series so the upper limit is really 800V. The power supplies can be voltage or current controlled with a 0-10V input. I currently have two pots for power control but would prefer using something compact, isolated, and digital and with a display (I am using a 4x20 LCD display). I set one pot to limit current the other to limit voltage. Since the arduino puts out a 0-5V PWM I'll be using a level shifter and an isolated power source (the power supplies each have an isolated 12V power source that I am utilizing).

It may be easiest to find an rs232 isolator or even rs232 to rs485 isolator and adapter for longer distance and cleaner communication.

Hi,
Thanks for the explanation.
Have you googled

isolated I2C bms arduino

This may help
https://www.youtube.com/watch?v=BlQDPORL8pI

Tom... :slight_smile:

The project is creating a trickle charger for an onboard large lead-carbon battery system.

Thanks, I learned something, I did now know about lead carbon batteries.

kwithlub:
I am trying to measure a high voltage on an Arduino (Nano). The voltage range is 400-850V. It is for charging a battery so the rate of voltage reading is not important (like once every 30 seconds at worst). I am also not very concerned about the resolution so the ADC on the Arduino is fine, but I would prefer the measurement to be isolated. Are there standard circuits/components that convert and isolate high voltage to 0-5V?

That was your original post. What was wrong with the isolated transducer suggestions I offered up in post #4? You can have your isolated 0 to 5 volts ideal for a uC analog input. Simple and done.

Ron