Help understanding current transducer datasheet

[gnif]

Hi All,

Just a couple of questions that are likely very simple ones to answer .

I am using a LEM HY 15-P with a +- 12V DC power supply and feeding it's output through a precision rectifier circuit that is then scaled to 3.3V and fed into a STM32F103. The datasheet for this part shows a "primary nominal RMS current" of 15A, and then proceeds to state that the output of the device is +- 4V. It also shows that its "primary current measurement range" of 45A.

From what I have been able to determine through experimentation is that the stated +-4V is for the nominal RMS current where 4V = 15A, and obviously it can measure higher currents that are not "nominal", what are the ramifications of doing so? For example, if I were to use the attached circuit to read values > 4V would I have to apply any corrections in software other then the obvious scaling when reading the other range values?

Circuit: https://tinyurl.com/yda3rljp

[jonpaul]

Bonjour we used LEM and Tektronix current probes since,1980s.

The LEM spec to is vague on RMS vs peak  for  V/A scaling.

I would query LEM engineering département and make your own tests.

I would think exceeding the rated 15 A RMS will give an unspecified result.

Jon

[capt bullshot]

The datasheet specifies 4V output @ nominal primary current.
So if you put in + or - 15A DC, you'll get + or -4V DC at the output. For AC, you'd need higher peak values, at least 1.414 times the nominal current. These sensors provide 3 times the nominal current as a peak value, so they can copy with 15Arms at a crest factor of three (which is 45A peak). It'll put out up to 3 times the nominal output (say -12V to +12V peak) with such input.
So 45A is the measuring range, intended for peak AC. You shouldn't run more DC or ACrms than the nominal current through it, as it might overheat or overload then.

For your correct uC measurement, you'd want a +/-12V full scale input to cover the full range. If you intend to measure DC only, or pure sine wave AC, you can design your input range smaller, e.g. +/- 8V peak for a 15Arms AC, or +/-5V for +/-15A DC input (always with some reserve)

Scaling (e.g. 4V -> 15A) is always the same.

.

[gnif]

hello

you can read this post for more understanding: https://www.airmar.com/uploads/brochures/understanding_xducer_specifications.pdf

This has nothing to do with the topic, and looking at your other posts quite a few of them look spammy... care to explain?

[jmelson]

The LEM sensors I am familiar with have a split toroid core with a Hall sensor in the gap.  Then, a many-turn winding on the toroid is given a current so that it nulls out the effect of the single-turn wire whose current is to be sensed.  The output is provided by a resistor in series with the multi-turn winding.  I don't think these are intended for AC use, but maybe 40 Hz is low enough.  The inductance of the multi-turn winding is the issue, how fast can the current in that winding be changed due to the inductance?
Jon

[gnif]

Sorry but I left out that I am measuring a slowly changing DC current. Attached it the circuit I ended up using which works well.

Circuit: https://tinyurl.com/y7sftx25

The datasheet specifies 4V output @ nominal primary current.
So if you put in + or - 15A DC, you'll get + or -4V DC at the output. For AC, you'd need higher peak values, at least 1.414 times the nominal current. These sensors provide 3 times the nominal current as a peak value, so they can copy with 15Arms at a crest factor of three (which is 45A peak). It'll put out up to 3 times the nominal output (say -12V to +12V peak) with such input.
So 45A is the measuring range, intended for peak AC. You shouldn't run more DC or ACrms than the nominal current through it, as it might overheat or overload then.

For your correct uC measurement, you'd want a +/-12V full scale input to cover the full range. If you intend to measure DC only, or pure sine wave AC, you can design your input range smaller, e.g. +/- 8V peak for a 15Arms AC, or +/-5V for +/-15A DC input (always with some reserve)

Scaling (e.g. 4V -> 15A) is always the same.

.

This makes sense! Thanks!

[Terry Bites]

Try this; use a passive attenuator that is >10Kohm. A small gain (filter) stage. A simple fw active recifier runs on uC power to protect inputs. Output is nominally 3.3V FSD on each scale. This allows you to scale From Inom to  Ipeak*3 and Ipeak /3. If  you derive the 3.3V from the uC then its fully protected from over voltage ot the ADC input.

Current settings shown are peak. [ Specified attachment is not available ]