LTC6106 Current Sense Conversion

[chandlerbing65nm]

I couldn't convert the ADC reading to its respective analog current value. I have tried the solution provided in this post but using 3v3.

(8-bit @ 3.3V) and LTC6106 for current sensing. Here's my schematic.

The input voltage is 3.3V.
The output ADC reading is HEX16 or DEC22
My expected value is 172mA

Here's my calculation:

3.3 V / (2^8 - 1) = 12.94 mV.

Sensitivity = V_out / I_sense = R_sense * R_out/ R_in,
= 0.025 Ω * (4999 Ω) / 100 Ω = 1.25 V/A

I_out = 12.94 mV * 22 / 1.25 V/A = 227.744 mA

What's my mistake here? Why I did not get 172mA?




Here are useful info from the datasheet of LTC6106

 

[KlausST]

Hi,

My first idea: offset

what is your ADC reading with no load?

Klaus

 

[chandlerbing65nm]

Hi,

My first idea: offset

what is your ADC reading with no load?

Klaus

With no load, the value is HEX00.

 

[KlausST]

Hi,

did you use kelvin wiring (4 wire sensing) ?

Show photo of your complete wiring and PCB.

Klaus

 

[chandlerbing65nm]

Hi,

did you use kelvin wiring (4 wire sensing) ?

Show photo of your complete wiring and PCB.

Klaus

The Vout is connected directly to a MCU that has ADC function.

 

[KlausST]

Hi,

open the LT6106 datasheet with a PDF viewer. Do a search for "kelvin".
And show us your photos.

Klaus

 

[chandlerbing65nm]

Hi,

open the LT6106 datasheet with a PDF viewer. Do a search for "kelvin".
And show us your photos.

Klaus

I don't even know what kelvin is. Please don't assume people asking in the forum to know anything. If I am knowledgeable, I wouldn't even try to ask here. Thanks.

I can't show the wiring because of some reason.

I connected a microcontroller in the Vout to read the current, it will give an 8bit HEX value.
The load is another microcontroller.

 

[KlausST]

Hi,

Don´t know why you are upset now.
No need to know everything from the beginning. I tried to show you where you can find the information.
I don´t know everything, by far not. Thus when I do an design I need to read datasheets, application notes, do internet search...
I´m sure I read many hunderd pages per week. In my opinion it should be nothing special for an electronics designer.

Klaus

 

[mtwieg]

open the LT6106 datasheet with a PDF viewer. Do a search for "kelvin".

The datasheet does not actually describe kelvin sensing.

Consider the two diagrams below of a current sense resistor and its connection to a current shunt amplifier. On the right, the sense traces are made far from the resistor terminals. The effective resistance of the sense resistor will be increased due to the resistance of the trace between the resistor and sense connections. On the left, the sense connections are made directly at the resistor terminals. This is a simple example of kelvin sensing.

 

[KlausST]

Hi,

The datasheet does not actually describe kelvin sensing.

True.

Just an objective view. Without being upset, without sarcasm...

Before post#7 I was not aware that he does not know what "kelvin wiring" is.
I did neither expect that he knows nor did I expect he does not know.
I just used the same keyword as the datasheet.
And I will do the same in future, because I think using different keywords is just confusing.
(example: if four people are talking about "the resistor", "R2", "the load", "100 Ohms" ... can one be sure all talk about the same part?)
I even added the synonym "4 wire sensing"...

In this case the recommended PDF search for "kelvin" will lead to the datasheet section "Sense resistor connection".
At this point he should knew that it is about "sense resistor connection" ... and the section does not mention "VOut, MCU, ADC" as he replied.
Also at this point he could do an internet search for "kelvin connection" which gives 5 million hits. Even 30.000 hits for videos.
So it´s one of the most often explained wiring techniques. Nothing special. Often asked. He´s not the first one. He is not dumb. No need to be upset.

And in case the internet information is not sufficient, he could still ask for an explanation.

If he feels dumb because he needs to do an internet search .... then probably I am the most dumb person, because I very often do internet search.

Klaus

 

[FvM]

The circuit in the original post has no direct relation to ADC conversion. Why don't you report voltage measurements at various circuit nodes to let us check, if the current sense amplifier is working correctly or not. ADC conversion is a differrent topic.