high DC voltage sensing

[omax]

Hello every one!
can i use this circuit to sense the output high DC voltage with ADC of LPC1768?

i have connected it to lpc like this:

two of my LPCs have burnt:shock::!: with this circuit.
what is the problem?!
thank in advance

 

[KlausST]

Hi,

the problem is that you connected two individual nodes with "GND" --> short circuit

Why the FET?
Why not sensig referenced to 220V_GND? (Why high side referenced?)

Klaus

 

[omax]

Hi,

the problem is that you connected two individual nodes with "GND" --> short circuit

Why the FET?
Why not sensig referenced to 220V_GND? (Why high side referenced?)

Klaus

where will be short circuit?

if i sense 220-GND two GNDs(i.e. circuit,and micro's and also optocoupler‬‏'s GND) will be connected together,right?
and as you know ADCs maximum input voltage should be less than 3.3V
should i isolate the adc part of micro from power circuit?

 

[FvM]

How about a complete schematic showing the 220V DC source and it's relation to "GND" node(s). Also actual TLP250 connections.

Presently it looks like the DC source is GND referred and you are supplying 220V to processor pin.

 

[omax]

this is my complete schematic

as you see i have two voltage feedback
i think i have found the problem,represented with red arrow in below picture, is it correct?

but now how can i sense two voltages and connect them to two ADCs?
please check it connections to lpc,
thank u all.

 

[KlausST]

Hi,

according your schematic all pink is GND.



The resistor (red) is completely useless then.

Klaus

 

[omax]

thank u for ur reply.
sorry,i forgot one part in the last picture. also note that we have two GNDs in the circuit

 

[KlausST]

Hi,

impossible.

--> The LPC is referenced to GND2...it can not output PWM to TLP which is referenced to GND1
--> Voltage measurement is referenced to GND1.. you can not connect it to ADC od LPC which is referenced to GND2.


Klaus

 

[FvM]

Presumed GND1 and GND2 are actually different nodes, the circuit still won't work, because you have ADC1/ADC2 shorting the secondary inductor/transformer winding. I also wonder if the processor has an isolated power supply as required.

You need some kind of high voltage differential amplifier, e.g. differential voltage divider followed by an instrumentation amplifier. Or a single OP differential amplifier with respective large input resistors.

 

[schmitt trigger]

What exactly are you attempting to do?

What is the purpose of the FET? Why the high side voltage sensing? Why the weird transformer connection? (or is it a coupled inductor).

 

[omax]

Hi,

impossible.

--> The LPC is referenced to GND2...it can not output PWM to TLP which is referenced to GND1
--> Voltage measurement is referenced to GND1.. you can not connect it to ADC od LPC which is referenced to GND2.


Klaus

but GNDs in first side and second side of tlp must be different,(lpc has one supply and tlp has another one for itself).
about the voltage measurement the GNDs from resistors to lpc have different voltages,so i think this is the problem, on the other hand if I measure the voltage from ground to the upper node all the GND will shorted together,yes?
maybe i should isolate the lpc ADC from circuit using linear optocoupler or sth such as lm331.
do you have any other idea to solve this problem?
thank u

- - - Updated - - -

What exactly are you attempting to do?

What is the purpose of the FET? Why the high side voltage sensing? Why the weird transformer connection? (or is it a coupled inductor).

this is a LED driver that i wand make a pwm switching for mosfet using two voltage feedbacks,
this two feedback used to find the proper duty cycle for current regulation.
(i know that i can sense the current or use other methods but i have to sense voltages to control the circuit)
thank u.

 

[KlausST]

Hi,

this is a LED driver

LEDs seem to be harmless. But with your non isolated supply they may hurt and/or kill people.

--> I strongly recommend to use an "isolating" power supply ... and if you really want to build it on your own you first have to go through the safety regulations of your country. And for sure follow them.

Don´t hurt or kill yourself or others!

Klaus

 

[omax]

Hi,



LEDs seem to be harmless. But with your non isolated supply they may hurt and/or kill people.

--> I strongly recommend to use an "isolating" power supply ... and if you really want to build it on your own you first have to go through the safety regulations of your country. And for sure follow them.

Don´t hurt or kill yourself or others!

Klaus

:clap:thank u for ur advice;
but this is not my own circuit,
https://ieeexplore.ieee.org/document/6755501/
i have designed an isolated one before,but not published yet.

 

[KlausST]

Hi,

but this is not my own circuit

Own circuit or not .... in any case you are reliable for safety.

Klaus

 

[schmitt trigger]

:clap:thank u for ur advice;
but this is not my own circuit,
https://ieeexplore.ieee.org/document/6755501/
i have designed an isolated one before,but not published yet.

I tried to open the link, but it does not allow me to access it without purchasing it.
Would you at least post an image of the original circuit?