10
\$\begingroup\$

Looking at Diodes Inc. datasheets, I am having trouble following their power dissipation limit calculations for their MOSFETS.

E.g. for DMG4496SSS http://www.diodes.com/_files/datasheets/ds32048.pdf

They specify on page 1

  • I_D(max) = 8A @ V_GS=4.5V (with a R_DS(on) = 0.029 ohm)

But then the datasheet also gives on page 2:

  • Power dissipation P_D = 1.42 W
  • Junction temperature T_J = 150°C
  • Thermal resistance R_\theta = 88.49 K/W

And on page 3:

  • R_DS(on) @ V_GS=4.5V, I_DS=8A approximately 0.024 ohm

To me this looks like one big mess:

  1. P = 0.029 ohm * (8A)^2 = 1.86 W which is significantly larger than the permissible power dissipation of P_D = 1.42 W from page 2
  2. even with the R_DS(on)=0.024 ohm value from page 3, P = 1.54 with is still larger than the permissible power dissipation
  3. the permissible power dissipation figures are at least self-consistent: P_D =(T_J-T_A) / R_\theta = (150°C-25K) / 88.49 K/W = 1.41 W
  4. However, the R_DS(on) vs V_GS and I_D vs V_DS graphs appear to be inconsistent: Looking at the case of V_GS = 3.5 V: In fig 1, the tangent at the point (V_DS=0.5V, I_D=10A) is about 6A/0.5V which seems to imply a R_DS(on) = 0.5V/6A = 0.083 ohm. Looking at fig. 3 however, the R_DS(on) is more like 0.048 ohm at 10A.

How to use Diodes Inc datasheets?

So given the datasheet, how would one calculate I_DS(max) provided some V_GS and some V_DS? E.g. V_GS = 6V and V_DS = 12V.

\$\endgroup\$
  • 2
    \$\begingroup\$ Have a +1 from me, solely for reading a datasheet in that detail. \$\endgroup\$ – PlasmaHH Feb 12 '16 at 20:34
  • \$\begingroup\$ Nice related article:mcmanis.com/chuck/robotics/projects/esc2/FET-power.html \$\endgroup\$ – jippie Feb 12 '16 at 20:41
  • 1
    \$\begingroup\$ @jippie Thanks for the reference, unfortunately that explains why the power rating of the MOSFET is LOWER than suggested by the P_D and R_DS(on) figures. In the datasheet I referenced, the power rating is HIGHER than suggested by P_D and R_DS(on)... - The first is completely logical, the latter should not be physically possible! \$\endgroup\$ – ARF Feb 12 '16 at 21:07
  • \$\begingroup\$ 1. I_Dmax is usually specified at V_GS = 10V or perhaps 5V for a logic level MOSFET. 2. I_Dmax is not limited by power dissipation in the way that you think--imagine 100ns pulses with a duty cycle of 1%. In such a case it would be possible to pass 30V/0.024Ohm = much more than 8A without ever exceeding the power dissipation limit, yet still destroy the device. The first-page specifications are often typical rather than guaranteed values, so I wouldn't take them too seriously if they're slightly contradicted elsewhere. Does that help a bit? \$\endgroup\$ – Oleksandr R. Feb 13 '16 at 5:39
  • \$\begingroup\$ I should also say that thermal resistance is not a static quantity, but time-dependent because the MOSFET has a certain heat capacity and thermal diffusion rate. Infrequent, powerful current pulses will heat it basically only to the extent of their RMS value, not instantaneously to incandescence. See also (35352). \$\endgroup\$ – Oleksandr R. Feb 13 '16 at 5:48
6
\$\begingroup\$

Yup, that's the way MOSFET datasheets work. The maximum current rating really means "This is the maximum current you can ever possibly get thru this thing, if you were to somehow not violate other specs in the process, although we have no idea how to do that. We put this here because we think it's cool, and maybe someone is dumb enough to buy a truckload of them before realizing they can't actually run the part at this value for any set of real world conditions".

Basically, each of the limits of the device are specified separately. You have to look at what you're doing and carefully check each one. The real limit on current is usually die temperature. To check that, look at the max Rdson for your gate drive level, compute the dissipation due to your current, multiply that by the die to ambient thermal resistance, add that to your ambient temperature, and compare the result to the maximum die operating temperature. When you figure all this backwards to find the maximum current the device can take before overheating, you'll usually find that's well below the absolute maximum current spec.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.