This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM74700-EP: Initial and subsequent EN delay timing and requirements of input capacitor?

Part Number: LM74700-EP
Other Parts Discussed in Thread: LM74680

Tool/software:

Hi,

I am developing a solution that uses the LM74700-EP in a low frequency rectifier circuit. I have a need to bring the anode voltage rapidly to 0V during every half cycle of the already rectified input voltage. According to the timing diagrams in the datasheet there is an initial EN delay of ~80uS. Due to the fact that I will bring the Anode voltage all the way down to 0V during every cycle, I would like to know if this EN delay is independent of the charge on the VCAP. That is to say, does this EN to gate turn on delay timing include the time required to charge the VCAP? If this is the case does this timing in fact include the time required at cold start for the anode to rise above the UVLO threshold, and then if EN is already above EN threshold, the time required to charge the VCAP above VCAP UVLO threshold?

If this is the case if we operate the IC in this rectifier mode, then the voltage on VCAP should be maintained during these short cycle by cycle dips down to 0V on the anode. Does this in fact reduce the EN to gate turn on delay? Or is the EN turn on delay more a function of some internal logic settling within the IC, where this is independent of the conditions on the VCAP pin? 

Furthermore, is there a strict requirement for an input capacitor. I understand that most application examples are concerned with automotive reverse current protection and that a TVS combined with an input capacitor makes a great deal of sense. However in my particular application any input capacitance would effect the rest of the circuit I am developing which actually requires the Anode voltage to rapidly follow the rectified source down to its zero crossing point. Is it therefore safe to operate the LM74700-EP without any input capacitance?

All the best

Aidan Walton

  • Hi Aidan,

    80uS delay will be present at every startup. This is not avoidable.

    Input capacitance is good to have if you need to filter out any noise of the power line.

    If you specifically looking for a rectifier, did you check LM74670? Or LM74680?

    Regards,

    Shiven Dhir

  • Hi Shiven,

    Thanks for your quick response. It is a good suggestion regarding the use of LM74670, however in fact I have a problem space that can be contained within the limits of Max Vgs of a PFET NFET bridge. My approach is to follow this simple structure with a single LM74700-EP to prevent reverse current flow. This works well in simulation and is significantly lower cost. 

    Having said this, if you could help me by summarising, if possible, the relevant differences between the LM74700 and these devices specified as 'rectifier controllers'.

    I assume and have found parts in the datasheet that would allude to these fact; the LM74670 is a fully floating solution. Its charge pump and reverse protection sinks and sources current from the Anode pin. Unlike the LM74700 which I believe uses its ground reference at least for the charge pump. However this is not entirely clear. Perhaps you can clarify.

    The AC source for my project is not so straight forward. It is in fact a field shaped energy harvester. During operation its phase changes are cyclically low frequency, but the edges are fast. I do not need very high voltages peak-to-peak on the rectifier, also current levels are not excessive, but I must limit any voltage drop across the rectifier as this affects MPP tracking and of course overall efficiency.

    So if you have any more to add regarding the alternatives, I am very pleased to understand.

    Many thanks

    Aidan

  • Hi Aidan,

    Your understanding is correct that LM747400-EP uses its ground reference for charge pump whereas LM74670 is fully floating.

    LM74670 takes power from the body diode drop of external MOSFET. For this reason, it works with duty cycle to recharge the charge pump time to time.

    LM74700-EP on the other hand has 100% duty cycle but needs certain voltage level on the ANODE pin.

    I believe LM74670 will have better efficiency because LM74700-EP will be turned-off every cycle and then have 80uS delay + time to charge its charge pump back.

    Regards,

    Shiven Dhir

  • Thanks Shiven, its a very fair point. I shall try and calculate these differences per cycle and see if the cost benefit is worth it. The rectifier controllers are both more expensive and considerably larger footprint, which does matter somewhat.

    Thanks for the help.

    Aidan