DRV421: TechNote: RCD leakage module using the DRV421

Martin,

You do not have to run the demag operation if you simply cancel out the offset due to the magnetized core. The magnetic core can get magnetized and the amount is based on the material used, the air gap size and the transient or large external magnetic field is applied.

If you think of any calibration you have to remove any outside error sources or influences to calibrate. That is why it is required to calibrate and demag in this condition. The DRV421 cannot distinguish the field coming from a magnetized core or the primary current (residual). If the primary current is zero (or magnetic field) is zero then you can use the demag function. This could also work if you know that the residual current is equal to zero. Please keep in mind that there still could be an offset from the fluxgate.

Hi Javier,

Thanks for the reply. So in our case if we cannot use a demag circuit, then it is important to produce a core that presents a little offset as possible (lower remanence core?).

Is there any way we could obtain a reference core, so we could verify our circuitry? I saw that there is a SC2912 and SC3113 from Sumida, but cannot see any availability.

Kind Regards,


Martin.

Apologies - I meant cannot use demag operation, did not infer using and extra demag circuit.

Hi,

A more general question - if we are making an RCD with a 30mA trip threshold, we would want the offset to be ideally less than 1mA, but no greater than 2-3mA. With all the offset errors within the circuitry and an optimised core, what do you think the best case offset at startup without de-mag at start-up is achievable?

Kind Regards,


Martin.

Martin,

At the moment I do not have a core for you to use for this. Sumida does no have those cores available either. There is a tradeoff with High permeability material which is preferred in order to get the magnetic coupling of two conductors optimized. The lower the permeability the lower the coupling from the two current carrying conductors couple into the core. This could create an error at the sensor as one conductor can magnetically couple directly to the sensor and not captured in the core. This would create an imbalance and the location of each conductor matters much more.

For lab work I have got a regular iron core and cut it myself. I then placed the DRV421 next to that gap. This is not optimal as you as you are measuring much lower currents and it would be best for a smaller air gap to get a higher magnetic gain in your core.

The demag is needed. Even with this function the DRV421 offset is max of 8uT.

I placed an example calculation on a previous post. Here is the example. Note I assume a 500uT/A gain seen from the core. This means that for every 1A you see through the primary it forces a field of 500uT at the sensor.

8µT/(500µT/A) = 16mA (Primary current error)
This they calculates as 16mA/1000 = 16µA on compensation coil
With Rshunt = 1kΩ there is a 16mV at the input of the differential amplifier
Therefore after the differential amplifier you can get 64mV.

Javier,

Understood re cores. And thanks for the example calculation.

However I think what we are saying here is that a 30mA RCD that uses the DRV421 should/must:

1) Have a core with a magnetic gain as high as possible to minimize the offset.

2) Have control over connection of the primary, being able to enable / disable via a relay etc, while de-mag is happening?

3) Wait 0.5s before enabling the primary.

This obviously limits the potential functionality of an RCD, where some applications must begin operation as soon as power is applied?

Kind Regards,

Martin.