TEA2208 Switch Speed / Waveform

[mspalluzzi]

The TEA2208 gives examples where it is attached to AC mains voltage but what is the capability of this part in terms of switching frequency / compatible waveform?

I see a max slew rate of 50V/ns which our design will meet.

We would like to use this part in a wireless power transfer application where the signal to be rectified would be a -36V to +36V 30kHz square wave. I assume the square wave will not be an issue? 

I understand the switching speed capability will be highly dictated by the gate capacitance of the external mosfets given the gate driver outputs have a 200mA typical sink/source current but wanted to make sure the gate driver itself will be able to switch in the 10's of kHz range.

[mspalluzzi]

Hey @TomasVaverka,

Reviewing the datasheet, it looks like 32V is the max threshold voltage not the minimum. With a typical threshold voltage of 22V I believe 36V should be sufficient even over temperature? Granted the threshold value is actually at the typical level.

[mspalluzzi]

Thanks for the response Tomas. Signal amplitude is something we can play with so increasing it to an acceptable voltage would be no problem but the two other points are concerning so thanks for clarifying. Since this part has only been tested at 1kHz I wouldnt want to commit this part to 30kHz.

On your last point, (TEA2208T not outputting DC), I understand the slew causing the dips in voltage. But I would expect this waveform if there is no smoothing capacitor:

With the smoothing capacitor I would think that the output would still be DC as in the waveform you show above with either the peak or full-bridge rectifier:

Is that not true?

[TomasVaverka]

Hi Michael,

Operation at 30kHz will most probably be a problem because of the gate losses and the associated current. But there might be also other limitations. The part has never been designed or tested for these frequencies and such low voltages. If we look at the gate currents then the resulting supply current is 4*Qg*30k. With Qg=100nC this results in 12mA. The self-supply IC cannot supply more than 1mA. So, this will not work. Applying external supply is considered to be too risky.

With a smoothing capacitor TEA2208 still gives:

This is different from a standard diode bridge.

Best regards,

Tomas

[mspalluzzi]

Gotcha, clear this part wont work for us. Thanks for clarifying Tomas.

- Michael

[TomasVaverka]

Hi Michael,

A square-wave is not a problem. The maximum frequency we tested is 1kHz. But we only tested with square-wave signals generated by an Agilent 6811B generator which does not have steep slopes. The rating of 50V/nsec is a non-damage rating for the high-side drivers which is in line with normal power FETs. Proper functionality of the total IC at 50V/nsec  is not relevant and not tested for a mains application which has an EMI_filter in front.

The 36Vpeak voltage is too close to the minimum voltage of 32V that is required for operation at 25C. Over temperature this will even need more than 32V. The TEA2208T is not intended for such low voltages. I would not recommend it.

Also note that that this is not a peak rectifier, but a full-wave rectifier. Looking at the picture below you see that a diode rectifier will output a DC voltage with ripple on it depending on the capacitor and the load current. With the TEA2208T you always get the full-wave rectified input, independent from capacitor or load current. If DC-voltage is required after the bridge, the TEA2208T should not be used.

Best regards,

Tomas