Is this chart working this way or I must connect the both grounds???

[stoyanoff]

I want to use a product and in it`s datasheet the manifacturer was drown the chart from the pdf. This is a cirquit for current measurement. I`m not if it`s working because the voltage I`m going to measure and the supply voltage hasn`t common ground. So I think the amplifier will not change it`s output.
Is that true or the chart is fine???View attachment sheme.pdf

 

[D.A.(Tony)Stewart]

YOu can measure current in any path of the loop using a differential amplifier to sense the voltage drop.

 

[sutapanaki]

why don't you show the circuit?
малко повече информация е добре да има ;–)

 

[erikl]

... the supply voltage hasn`t common ground. So I think the amplifier will not change it`s output.
Is that true or the chart is fine???View attachment 72684

You're right: you need a "common" connection.
I'd suggest to connect the 5Vdc midpoint (i.e. the common node of R2 & R3) to either the inverting or the non-inverting input of U1A.

@ sutapanaki: You seem to speak Bulgarian, but can't read a PDF? SCNR ;-)

 

[D.A.(Tony)Stewart]

stoyanoff< You need to use negative feedback gain circuit, not open loop full gain. You can use -ve input for second stage with Rf/Rin gain control , However your best Common mode rejection design is called an Instrument Amplifier using (3 Op Amps) configuration and can adjust gain with a single resistor., But it is possible to do it with only 1 Op Amp and suitable input Bias resistors to V/2 and dual input R and 2 ratio R's

 

[sutapanaki]

You're right: you need a "common" connection.
I'd suggest to connect the 5Vdc midpoint (i.e. the common node of R2 & R3) to either the inverting or the non-inverting input of U1A.

@ sutapanaki: You seem to speak Bulgarian, but can't read a PDF? SCNR ;-)

) yes, you're right. Just now I noticed that there was an attached pdf in the original post. And I also think this circuit is not going to work open loop. A feed back is needed.

 

[stoyanoff]

The common ground is a problem. I`m using a microcontroller to measure constant current. An etalon resistor is connected to the power ground through the load. And here is the problem : the power voltage is 450V DC. I don`t want to connect the both grounds because I think there is a chance of an electrical burst which can destroy my controller.
Any suggestions?

 

[erikl]

The common ground is a problem: the power voltage is 450V DC. I don`t want to connect the both grounds because I think there is a chance of an electrical burst which can destroy my controller. Any suggestions?

Use an isolated/isolating sensor method, e.g. a Hall current sensor, or a magnetoresistive sensor:

 

[D.A.(Tony)Stewart]

Usually the smart thing is to do on current sensing of high voltage is choose;
a) put current sense R on the common ground side not B+
b) use an instrument amp design that supports very high common mode voltages by using input resistor divider to attenuate the high voltage in range of amp and then choose very high differential gain. (low offset)
c) use a floating power source to measure current
d) use optical interface using PCM or some other method.
I suggest a) & b)

 

[mister_rf]

LTC6101fh =dedicated circuit for high-side current shunt monitor

http://circuits.linear.com/239


Also check for MAX4172.
But here for higher voltages, need to re-size R1 as suggested in the text
**broken link removed**

**broken link removed**