Measuring high current and voltage in 3 places

Thread Starter

Bogdan.m

Joined Apr 20, 2019
57
Hello everyone, i am building an arduino based solar controller and i am looking for ideas on how to measure high current. I made a quick sketch of my idea. My initial thought was shunt resistors and ads1115, but from what i read about it it cannot measure negative voltages. maybe you can advise me on a different configuration or something.
2020-04-18 14.48.05-1.jpg
 

nsaspook

Joined Aug 27, 2009
13,086
Ratiomatic hall sensors are nice but they should be supplied with precision Vcc voltage for both device/ADC ref input and used with at least a 12-bit ADC for rated ACCURACY PERFORMANCE in the higher current models where the signal voltage is 40 (mV/A) on +- 50A units. For a ACS724 measuring battery charging DC the filter function should be used to reduce device noise to a lower level to improve measurement resolution.

For +- 100A I use this sensor CSLT6B100 : https://sensing.honeywell.com/CSLT6B100-open-loop-current-sensors

Accurate, stable with low noise.

Honeywell CSLT6B100 100A zero current voltage. SD of 86 microvolts, very good.
 

Thread Starter

Bogdan.m

Joined Apr 20, 2019
57
So the acs comes in acs758 as well, and it is bidirectional, and it goes up to 200A as well, could be what i need so i ordered a few of thees, and i am gonna play with them after they arrive, so thanks for the idea.
LE: also i a thinking of putting them trough an ads1115 adc as well as the voltage readings so i can get a better accuracy. what do you guys think ? Combining all ideas.
 
Last edited:

nsaspook

Joined Aug 27, 2009
13,086
So the acs comes in acs758 as well, and it is bidirectional, and it goes up to 200A as well, could be what i need so i ordered a few of thees, and i am gonna play with them after they arrive, so thanks for the idea.
LE: also i a thinking of putting them trough an ads1115 adc as well as the voltage readings so i can get a better accuracy. what do you guys think ? Combining all ideas.
If you want to actually be able to use that ADC resolution just be sure to run both the radiometric sensor and ADC ref at 5.0 volts using a separate precision supply (regulator chip like a REF02) from the main digital supply. A digital or analog LP filter will also be necessary to reduce sensor noise.
 

nsaspook

Joined Aug 27, 2009
13,086
Last edited:

Thread Starter

Bogdan.m

Joined Apr 20, 2019
57
The input will be a regulated 12V with decent amperage. I was wondering if it is a good idea to put some trimmers between the adc and sensor so i can manually increase the accuracy even more.
 

MisterBill2

Joined Jan 23, 2018
18,176
Igot a number of industrial current measuring packages that use a Burr Brown isolation amplifier and a 100amp shunt in one nice package. They are great in that the output is 1-10 volts for 0-100 amps, totally isolated. The isolation amp is a type 3656HG from Burr Brown, Both the input and the output are isolated from the power source. They provide good application notes on how to use the IC, so no challenging design work needed.
 

nsaspook

Joined Aug 27, 2009
13,086
The input will be a regulated 12V with decent amperage. I was wondering if it is a good idea to put some trimmers between the adc and sensor so i can manually increase the accuracy even more.
Trimmers?

You can and I did this time but it's a source of drift unless somewhat expensive units are used. With proper precision resistors it's easily possible to have software calibration on all ADC channels using a internal voltage reference as the calibrations source. For my latest solar monitor design I have a analog header that connects all inputs to the internal precision voltage source for trim scaling adjustments. Once the voltage levels are calibrated another routine sets the zero offset for current sensors first and then calibrates the current read-back using a 10A source. You will need the scale the raw ADC numbers to standard volt, current units, so software calibration to EEPROM is usually the better way.
 

Thread Starter

Bogdan.m

Joined Apr 20, 2019
57
so this is a sketch of what i have so far, it is incomplete and a lot more changes will come, i am sure.If you guys have any recommendations, i am open to ideas.
Untitled.png
 

danadak

Joined Mar 10, 2018
4,057
For future reference there are processors with onboard A/Ds that and internal front ends that can
handle measuring signals outside supply rails by 100 mV, usually enough for high or low side current
sensing. This one can do this up to 20 bits. This one can also use internal analog mux so that multiple
channels can be measured.

This also has a precision onboard Vref.


1587640573343.png
 
Last edited:

nsaspook

Joined Aug 27, 2009
13,086
any thoughts ?
How about a schematic so we can see the circuit. The RE02 is a reference chip that's current limited to about 20mA per device. Each of your current sensors uses about 10mA with the ADC module using about 150uA each so you should review the power budget as you will need two ref chips. The ADS1115 has no provision for external voltage reference for rail to rail measurements so the internal 4.096 ref will be used with a PGA gain of 2/3 to get 6.144V max scale. (you can't exceed VDD @5.0v at the pin) PCB test points (for test pin easy access) for reference voltages, analog reference ground is a good idea with the normal set of circuit testing points for digital and analog signals.
 
Last edited:

nsaspook

Joined Aug 27, 2009
13,086
During voltage calibration (using a RE02 for the internal voltage standard value readout on the OMEGA calibrator mV scale) you should see results like this if your ADC measurement system is stable.

 

Thread Starter

Bogdan.m

Joined Apr 20, 2019
57
The schematic is non existent unfortunately since i add things to pcb once i need a modification, i added another ref02 on the board as well as test terminals for testing the voltage and current inputs.
Untitled.png
 
Top