SM72295 MOSFET Driver Question

Hi Paul,

 Thanks for considering the SM72295.

1. The SM72295 does not have a specific frequency limit, however depending on the load on the output of the driver, at higher frequencies there may be thermal limitations.  The  thermal losses depends on the Qg of the MOSFETs that are being driven.  The LM5101A datasheet describes how these losses can be estimated, and based on the loss calculation there, the temperature rise can be calculated using the 60degC/W thermal resistance for the SM72295. This driver and others like it have been used at up to 500kHz, but again, there is no specific limit.

2. There is no protection against these conditions in the SM72295, since the SM72295 allows independent control of all four gate drives. Deadtime and protection against shoot-through (if needed) must be provided by the device creating the PWM signals that are input into the SM72295. In fact, in some applications this driver is used in a converter topology where overlapping HS and LS signals are required.

3. For the same reasons as mentioned in Question 2, the deadtime must be created by the PWM signals.

4. The current sense amplifiers have a bandwidth that varies depending on the magnitude of the signal going into the amplifier.  With a 4mV signal, the bandwidth is approximately 380kHz, and with a 40mV signal, the bandwidth is approximately 1.1MHz. So the bandwidth of the current sensing can be adjusted to some degree depending on the size of the current sense resistor and the DC amplitude of your current.

I hope these answers address your questions.  Feel free to let me know if you have more questions.

 Perry

a few more questions please:

1. Regarding my original question #4, what is the amplifier’s step response delay from a sudden large change of current at the input to an equivalent step response output? This should be in units of seconds.
2. Is the part’s behavior at all affected by the measurement of the current sense amplifiers, or are these just for convenience to the outside circuit (i.e. reduces the need to place additional amplifiers and that’s their only purpose)?
3. Are there any more application notes or white paper which go into more technical detail about this part? Things like step response is a standard part of amplifier datasheets, so I’m a bit concerned about the lack of information about this part’s behavior.

1. The step response delay is directly related to the bandwidth of the amplifier.  With the case of a 40mV signal and 1.1MHz bandwidth, a step from 40mV to 41mV would have a time constant of tau = 1/(2*pi*1.1MHz).  This means that for the output signal to settle to 95% of it's final value would take 3*tau=43us. What is the settling time you need?

2. The behavior of the gate drives is completely independent of the measurements from the current sense amplifiers.  These were combined in one circuit because of the convenience of replacing the external amplifiers.  Unlike standard amplifiers, the amplifiers integrated in the SM72295 have a high common mode range enabling them to sense the current from a high-side current sense resistor.  This has has some layout advantages over low-side sensing becuase it does not require breaking the power ground plane.

3. Unfortunately we do not have any further documentation available. If there is particular aspect of the circuit you need to verify, you can ask us here and we will see if we have data for it, or we can supply you with samples for your evaluation.

Perry