Hall sensor output voltage not zero when sensed current is zero?

Status
Not open for further replies.
T

treez

Guest
Page 20 of the ACS724 Hall sensor datasheet says that its output when sensed current is zero is vcc/2.
Why should this be so for the unidirectional version?…surely it should be 0v output for 0A sensed current?

ACS724 Hall sensor datasheet:
**broken link removed**
 

Hi,

There are several different types of hall sensors in the datasheet. You should exactely say wich one you refer to.

I suppose you refer to a bipolar one.
Then the input is bipolar and for sure the output needs to be bipolar, too.

With a single supply sensor you have to add an offset (here Vcc/2) to the output, otherwise it is impossible to represent negative signals at the output.

Klaus
 
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Hi,

Datasheet says for all unipolar sensors it is 0.1 x VCC...

It makes sense, because it is not possible to regulate exactely to zero. Only "near" zero.



Klaus
 

thanks, must admit 0.1xvcc is not as near zero as would like.
 

Hi,

What's the problem?

It just decreases the "dynamic" output voltage a little bit, but it enables to precisely read currents close to zero.

Klaus
 
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
the zero current voltage does not linger when the current goes above zero?...i mean , that voltage does not remain when there is a few amps flowing?......ie its not a constant positive offset on the reading?
 

Hi,

0.1 × VCC output means that there is no current.

As soon as the current rises, the voltage will rise, too.

Klaus
 
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
thanks, yes the voltage will rise, but does it still have that offset on it?, the offset of the voltage that existed when there was no current
 

Hi,

Is there any alternative?

V_out = 0.1 × Vcc + gain × I

Klaus
 
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Thanks, but yes, why do we assume that the 0.5V offset persists when one goes above 1.25A?...I mean, its seems likely, but by no means a certainty.

eg, 1.5A could Just give a 0.6V output...not necessarily a 0.6+0.5V output.
 

Is it because it has two op amps in it?

My ("well-made") ammeter using a current shunt monitor had a minimum of noise guaranteed at the output, even with excellent layout techniques explained in the datasheet, along with a formula to estimate the permanent offset there would be, and in my hands and layout on FR4 with bits of cable and so on, that turned into an additional 4 - 5 mA on top of all readings.
 
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
thanks, you see the ACS724 HALL sensor does not call the "0.1xvcc" an "offset"....so i wager that it is not an "offset" at all...it is just a voltage that exists at the zero current point, and up to the point where the current is 1.25A and greater....if it was a 0.5v "offset" then surely they would call it a "0.5v offset"?...but they dont.
 

Any o/p op amp has a min o/p voltage, one of the best is the LM324 which can pull down to 10mV on a good day, your o/p circuitry may only be good enough to pull down to 100mV, you'll have to live with that... unless you design your own ckt to subtract the offset and you use an op amp that can go to 5mV or so at its output, good luck...
 
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
thanks, your explanation suggests that the 0.5V "thing" doesnt get added to outputs for currents above 1.25A.....the hall sensor has 400mV/A output, but outputs 0.1xvcc at zero current...i suggest that at 1.5A of sensed current, the output will be 0.6V, and not 1.1V.....do you agree?
 

thanks, so for our system which has Vcc = 5V......
-for sensed current of 1.5A, the output will be 0.6V?
-for sensed current of 1A, the output will be 0.5V ?
-for sensed current of 0A, the output will be 0.5V?

(its a 400mV/A hall sensor)
 

Hi,

Yes, I think you are right. Now that you put it that way, if the Hall sensor is like a current shunt monitor, only regarding that aspect, and I seem to recollect some-one here who is an engineer explaining that to me a few months ago. So, again, yes, I believe you are right.
 
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Hi,

thanks, so for our system which has Vcc = 5V......
-for sensed current of 1.5A, the output will be 0.6V?
-for sensed current of 1A, the output will be 0.5V ?
-for sensed current of 0A, the output will be 0.5V?

(its a 400mV/A hall sensor)

I'm very sure this is not the case.
0A --> 0.5V
1A --> 0.9V
1.5A --> 1.3V
...
10A --> 4.5V

*****
For this sensor the 0.1 x VCC is very precisely "adjusted".
For the bipolar sensor 0.5 x VCC is very precisely adjusted.
It is trimmed to be within +/-7mV. This is not done with typical "output-to-rail" problem.
Both giving an output range of 0.1x VCC .... 0.9 x VCC.

A sensor, where you loose the lower 1/8 of the measurement range I'd call useless.

I recommend to
* either contact the manufacturer,
* or buy a sensor and do some tests.

Klaus
 
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
I don't understand the problem.

It's a high side current sensor using differential Hall sensors to reduce stray magnetic effects with a single supply for unipolar DC current using a ratiometric supply reference added to the output to avoid the converted signal inside being near ground.

It is digitally compensated for linearity and tempco.

Did your want to convert this to a 0~4V signal? or ?
 
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Similar threads

Cookies are required to use this site. You must accept them to continue using the site. Learn more…