Hi guys,

I'm working into the design of a battery monitor, battery array voltage goes from 42v to 56v aprox.

I want to use a hall effect current sensor feeded by -15VDC, 0, +15VDC.

I am thinking to use the LM5013DDAR to step down the battery voltage to 30VDC to get a middle step and have the 0 +-15VDC for the hall current sensor, but I dont know how to get the step at 15v from the 0v - 30v, I wonder if it is possible to achieve the with a mosfet or something with few components.

Thank you in advance.

 

A Hall-Effect-Current-Sensor doesn't care what the Voltage is ( within certain limits of course ).

Instead of telling us about your solution,
state all of the specifications of your Battery,
and what problem You are trying to solve.
.
.
.

 

Edgar,
this splitter built on an OPamp may work:

https://www.eleccircuit.com/power-supply-splitter-circuit-using-op-amp/

 

A Hall-Effect-Current-Sensor doesn't care what the Voltage is ( within certain limits of course ).

Instead of telling us about your solution,
state all of the specifications of your Battery,
and what problem You are trying to solve.
.
.
.

It is a gel battery array, 8 * 6v series, cuttof at 42v.

I want to watch the bidirectional current, the hall effect sensor that I've found has -15v 0v +15v input, and output -4v +4v, I want to put the electronic (ADC and uC) at that step 0v (or close) to make it easy the measurements by a voltage divider at the hall effect sensor output.

That is why I though that it could be achieved by a mosfet, like it is done with the BSS138, to have the step that can handle the current to the sensor and the electronic, but I dont know how to work with mosfets.

Kind regards,

I hope I've explained myself

 

Edgar,
this splitter built on an OPamp may work:

https://www.eleccircuit.com/power-supply-splitter-circuit-using-op-amp/

Will take a look for sure, thank you very much.

 

You might save yourself a lot of trouble if you chose a current sensor with a 5V supply.
There are plenty available:
https://www.lem.com/en/current-transducers
https://www.micro-transformer.com/ZMKD16SCP-Series-Hall-Current-Sensor-pd45960156.html
https://www.allegromicro.com/en/pro...ty-amp-integrated-conductor-sensor-ics/acs712

 

You might save yourself a lot of trouble if you chose a current sensor with a 5V supply.
There are plenty available:
https://www.lem.com/en/current-transducers
https://www.micro-transformer.com/ZMKD16SCP-Series-Hall-Current-Sensor-pd45960156.html
https://www.allegromicro.com/en/pro...ty-amp-integrated-conductor-sensor-ics/acs712

I think it worth a try to deal with the electronic once and have it working in order to save into the sensor investment, I've seen multiple sensors and the cheaper ones has that dc step between +- vcc (mostly the bidirectional ones).

One direction sensors could work too, but I would need two, one for charging and other for discharging the battery, not discarted option but just doing the research and asking about, thank you for your suggestions.

Amps could reach close to 100A (@50VDC for 5KVA output inversor), so I think it should be a donut around the wire and not a PCB sensor.

 

I think it worth a try to deal with the electronic once and have it working in order to save into the sensor investment, I've seen multiple sensors and the cheaper ones has that dc step between +- vcc (mostly the bidirectional ones).

One direction sensors could work too, but I would need two, one for charging and other for discharging the battery, not discarted option but just doing the research and asking about, thank you for your suggestions.

Amps could reach close to 100A (@50VDC for 5KVA output inversor), so I think it should be a donut around the wire and not a PCB sensor.

MicroTransformer quoted me $3.09 for theirs (I don't remember the exact model). 3.3V supply, bidirectional, with the output at 1.65V with no current flowing

 

Standard Analog Devices hall effect ACS772 is the easiest/quickest solution (datasheet here). I use these for my LiFEPO4 180Ah wheelchair battery pack for both charge and discharge "Coulomb counting" to track true SOC. VCC = 5v, 0A = Vref = VCC/2, Vref+2v = +100A, Vref-2v = -100A

Here's the layout I use, the grey squares are stick-on 30mm sq BGA heatsinks, onto 50mm sq 1mm thick copper pads - in my use it barely gets warm when charging at 40A

 

MicroTransformer quoted me $3.09 for theirs (I don't remember the exact model). 3.3V supply, bidirectional, with the output at 1.65V with no current flowing

Wow that would be such useful solution, thank you!

 

Standard Analog Devices hall effect ACS772 is the easiest/quickest solution (datasheet here). I use these for my LiFEPO4 180Ah wheelchair battery pack for both charge and discharge "Coulomb counting" to track true SOC. VCC = 5v, 0A = Vref = VCC/2, Vref+2v = +100A, Vref-2v = -100A

Here's the layout I use, the grey squares are stick-on 30mm sq BGA heatsinks, onto 50mm sq 1mm thick copper pads - in my use it barely gets warm when charging at 40A

View attachment 298036

I appreciate your suggestion, but I prefer to avoid dealing with Joule effect (mainly for size restrictions) and the fact that the circuit has to be intervened for installation, need to cut the wire to install the device, I rather based hall effect sensors by ease installation.

For this particular application I'm working on is not feasible, but for sure I will have it present, I guess this could be pretty accurate at low currents, is it? Im curious.

 

A Hall-Effect-Current-Sensor doesn't care what the Voltage is ( within certain limits of course ).

Ummm ... yeah, they do. Hall effect devices, and the circuits that process them, have clearly specified operating voltage ranges.

ak

 

this splitter built on an OPamp may work:

https://www.eleccircuit.com/power-supply-splitter-circuit-using-op-amp/

That circuit can work very well. AND, if the Hall device operating current is low enough, you might be able to eliminate the two current-boost transistors for even better performance.

Separate from that, my experiences with both TI and Maxim integrated switching controllers with a bottom-side power pad were not good. Replacing the chip when it fails is a real pain. We wound up putting the circuit on a small daughter board and treating the whole circuit as a single component.

ak