ground isolation

[ranganatennakoon]

I planed constant power load control circuit. All parts are fully operational in computer simulation. But I want to Isolate High voltage DC ground (GND 2) and low voltage DC Ground (Gnd).  What can be done for this? How to solve this problem?



This circuit works very well when one part is taken.
The system works very well when all ground  are connected.

This system is flawless.The current  controlloing is very smooth and stable.
he current flowing through the shunt could also be measured very accurately.
So there are no errors in it,
I took the help of EEv bolog, Greatscoot, electronoobs YouTube channels for this. There are no errors

[f4eru]

hmm, crispy mosfets in sight
take care, linear driven FETs can go short at the slightest mistake.
You might want to add some protections.

For isolation, perhaps two analog isolators on the control signals....
You*ll have to add a second PSU

[NiHaoMike]

ESP32 is a wireless chip, I assume you want it isolated since you have something else connected to it that needs to be isolated? Might it make more sense to isolate that? Easier to isolate digital signals than analog, especially if linearity is important.

[srb1954]

I planed constant power load control circuit. All parts are fully operational in computer simulation. But I want to Isolate High voltage DC ground (GND 2) and low voltage DC Ground (Gnd).  What can be done for this? How to solve this problem?

Not easily. With the configuration you have drawn up the low voltage ground and the high voltage ground have to remain connected. The op amp U1B is comparing the sense voltage across resistor R4 and the reference voltage at the junction of R1 and R2. For this comparison to be valid these 2 voltages must share a common reference level and hence the 2 grounds have to be connected.

If you really have to have this circuit isolated from the ESP32 then it would be far easier to use an external DAC and ADC chips and connect these through a isolated digital interface e.g. I2C or SPI to the ESP32.

You can also simplify the circuit a bit by ditching the separate LM358 U2A and its separate 3.3V supply and using the remaining section of the LM324 op amp. The output of the LM324 can then have a resistive divider to divide the higher output swing down to safe levels for the ESP32 and the gain of the LM324 can be increased to compensate for the divider loss.

The resistance of this divider will also provide some protection for the ESP32 ADC input by limiting the current fed back into it when the MOSFET fails. When the MOSFET becomes overheated it will likely fail with a source to gate short, which will take out all the other circuitry in a quite spectacular fashion if you still have 185V applied to the circuit. You should also add some resistance in series with the ESP32 DAC output to limit potential fault currents.

Finally, you should not rely too much on getting valid computer simulations to design your circuits. You should start by checking, on paper, the expected voltage current and power levels in your circuit to make sure everything in working within its rated limits. I think you will find that it is practically impossible to safely dissipate the amount of power in the MOSFET that this circuit would seem to require. The computer simulation doesn't care about device power dissipations and you can have MW power dissipation in simulated components without them exploding or catching fire. However, real components care very much about levels of power dissipation and junction temperatures etc. 

[ranganatennakoon]

This circuit works very well when one part is taken.
The system works very well when all ground  are connected.

This system is flawless.The current  controlloing is very smooth and stable.
he current flowing through the shunt could also be measured very accurately.
So there are no errors in it,
I took the help of EEv bolog, Greatscoot, electronoobs YouTube channels for this. There are no errors


Is it possible to use ground isolation like B1212 or any other method for this?   






Thanks

[srb1954]

This circuit works very well when one part is taken.
The system works very well when all ground  are connected.

Off course it does. Most simulations work better than real-world circuits as they don't bother to fully model the real-world behaviour of circuits. Consequently, real-world circuits quite often disagree with simulations and no amount of arguing that your simulation was perfect will make the real-world circuit work any better.

This system is flawless.The current  controlloing is very smooth and stable.
he current flowing through the shunt could also be measured very accurately.
So there are no errors in it,
I took the help of EEv bolog, Greatscoot, electronoobs YouTube channels for this. There are no errors

You are brave to claim your design is flawless. You are going to get a big surprise when the real-world disagrees with your assertion of perfection!

Maybe you should read up on the Dunning-Kruger effect before making such claims.

Is it possible to use ground isolation like B1212 or any other method for this?   






Thanks

Draw up a circuit showing how you propose to utilise a B1212 for isolation so we can comment.

[ranganatennakoon]

No, it works well I made on veroboard

[srb1954]

No, it works well I made on veroboard

Then show us a picture of the unit and the results of your measurements proving that the design is flawless as you claim.

[ranganatennakoon]

The circuit works fine.





watch this 
https://www.youtube.com/shorts/oaF8soCysAA
https://www.youtube.com/shorts/xBlz2WFeGcY

this circuit has 3 main units

1) esp32 DAC --> non inverting amp gain is 4 (12 V DC )
2) current control loop (185V DC )
3) measure shunt voltage --> non inverting amplifier to gain 6 ( 3.3V DC) --> esp 32 ADC

These three sections are powered by under different voltage levels! I want isolate Grounds of 12V and 185V for safety
thanks

[NiHaoMike]

I suggest adding isolation to external I/O, trying to isolate precision analog is asking for a lot of difficulty. You could also isolate the front panel but it would more more sense to just properly design it to comply with double insulated standards.

[srb1954]

The circuit works fine.

Hard to tell anything about the performance of your circuit from these videos. However, it is notable that there is no sign of an oscilloscope in the measurement setup so how do you know if the circuit is even stable?

These three sections are powered by under different voltage levels! I want isolate Grounds of 12V and 185V for safety
thanks

You obviously didn't take notice of my first post as to why it is not possible to separate the grounds in the circuit configuration you have chosen so I will repeat it in bold:

With the configuration you have drawn up the low voltage ground and the high voltage ground have to remain connected. The op amp U1B is comparing the sense voltage across resistor R4 and the reference voltage at the junction of R1 and R2. For this comparison to be valid these 2 voltages must share a common reference level and hence the 2 grounds have to be connected.

If you do not understand the above statement I suggest you take some basic courses in electrical network theory and feedback theory before attempting to design such circuits.

[David Hess]

The best option is to isolate the ADC and DAC on the digital side, but isolation could be done on the analog side using isolation amplifiers.