Elementary question about IRF540, from ST

Hello there guys,

I have a doubt here regarding the interpretation of the maximum rating of ID current.

In the datasheet we can see that the drain current (continuous) @Tc = 25ºC is 22A and @ 100ºC is 15A.

On other hand, there is IDM current, that is the Drain current (pulsed), which can go up to 88A.

My question is, imagining that we are designing a BUCK converter, for example, we have to look to the IDM current or the ID current, for example @ 25ºC?

There is a note which says that the "Pulse width limited by SOA". This means that the current will be function of the width of the pulse? Supposing that I want 20A, @ 100KHz, the 100KHz operating frequency must have a maximum width of 100us?

I have another question, regarding the Ciss capacitance. The graphic that the datasheet presents is only @ 1MHz. Supposing that we will operate the converter at 100KHz how do we know the Ciss capacitance at that frequency? Or the Ciss @ 1 MHz is the worst case, then we can compute, for example, losses with the Ciss value @ 1MHz which will be the worst case?

Best regards.

There is a note which says that the "Pulse width limited by SOA". This means that the current will be function of the width of the pulse? Supposing that I want 20A, @ 100KHz, the 100KHz operating frequency must have a maximum width of 100us?

Click to expand...

The SOA curve is clearly dedicated to single pulses.

For repeated operation, you can refer to average power as a rough estimation. Heat sink thermal resistance must be considered of course. For an exact calculation of permitted peak currents, the transient thermal impedance curve can be helpful.

Ciss is frequency independent in a first order. To calculate the effective driver load, you should better use total gate charge.

Okay, thank you for the answer.

So, what you mean is that "On other hand, there is IDM current, that is the Drain current (pulsed), which can go up to 88A" is only for one pulse? Supposing that there is any current transient, the maximum peak that can occur without damaging the device is 88A?

I say, the permitted peak current in repeated operation will depend on pulse width, PWM frequency and case temperature. For 100 kHz you can most likely calculate with average power dissipation.

The high Vce,sat values for currents above 20 A makes it unlikely that you really want to operate the transistor in this region, I think.

My confusion is for what current parameter I must look in order to see if I can use this transistor to my desired current.
About the average power, I can't see the graphic on the datasheet.

Dissipated average power is avg(Id*Vds), respectively Irms²*rds.

Permitted average power depends on total thermal resistance and maximum ambient temperature.

I see.

So, Id(avg)*Vds(av) + Irms²*rds. Is that it?

I still stuck in the current.

So, Id(avg)*Vds(av) + Irms²*rds. Is that it?

Click to expand...

I don't think so.
avg(Id*Vds) written more verbosely is avg(Id(t)*Vds(t)), the averaged product of instantaneous voltage and currents.