2
\$\begingroup\$

In a design, I have Zener diode used for the protection of a MOSFET's gate and source as shown in the picture below.

I'm comparing two diodes for this purpose, and most of the spec values are equal, except for the "Voltage - Forward (Vf) (Max) @ If", where one says 1.2V @ 200mA, and the other says 1.2V @ 1mA.

In this case, would there be any consequence in choosing one over another?


Edit : The gate and source of this MOSFET is connected to the high-side of a gate driver, where the source is not the lowest potential in the circuit. Here's the datasheets of the two diodes: diode 1, diode 2

(picture taken from here)

enter image description here

\$\endgroup\$
8
  • 2
    \$\begingroup\$ It's impossible to say without knowing what's behind 'GateDrv'. Please edit your question and add that part of the circuit, without any ambiguous source signals. Thanks. \$\endgroup\$
    – TonyM
    May 3, 2022 at 20:57
  • 2
    \$\begingroup\$ While editing, it would be also interesting to share the link to the two datasheets. \$\endgroup\$
    – devnull
    May 3, 2022 at 20:59
  • 1
    \$\begingroup\$ Why do you need to protect the Gate of the MOS-FET? You have to share more information about the environment, the whole picture of the circuit. \$\endgroup\$ May 3, 2022 at 21:39
  • 1
    \$\begingroup\$ Your edit doesn't tell us anything - you:re still just showing a copy and paste of a circuit fragment. Please edit your question and add the full circuit info detailed earlier. VTC as that doesn't seem to be forthcoming, unfortunately. \$\endgroup\$
    – TonyM
    May 3, 2022 at 21:44
  • 1
    \$\begingroup\$ The forward current given there isn't the current rating, that's the test current at which the forward voltage is specified. \$\endgroup\$
    – Hearth
    May 4, 2022 at 13:31

2 Answers 2

4
\$\begingroup\$

Note that for an N-channel Enhancement-Mode MOS (NEMOS), 1.2V Source-Gate is quite low; most NEMOS will not work appreciably at this low of a voltage. So if indeed using it like this, ensure the chosen device is truly "on" at that gate voltage (has a VGSthreshold of less than 1V likely.)

I'm comparing two diodes for this purpose, and most of the spec values are equal, except for the "Voltage - Forward (Vf) (Max) @ If", where one says 1.2V @ 200mA, and the other says 1.2V @ 1mA. In this case, would there be any consequence in choosing one over another?

This all boils down to the differences of the two Zeners. Note that this question is about forward conduction, so in this circuit, would be clamping negative input. Most MOSFETs can withstand an equal amount of negative gate voltage as their positive limit (such as +/-15V) but always check the MOSFET datasheet to be sure.

If GateDrv has a 1k impedance, then the "1.2V @ 200mA" one will have a lower forward voltage at 1mA versus 200mA. This is because a Zener (all diodes) are non-linear; at it's forward transition voltage (perhaps 0.7V) it starts conducting current, but this is not a hard-switching point. The voltage will rise if current increases - but since GateDrv has a relatively high resistance, it won't rise much at all.

Now if GateDrv has a 10R impedance, then the "1.2V @ 1mA" one will have a higher voltage, perhaps 1.6V or so, because more current is being dumped through it. The details for reverse breakdown are in the datasheets, since these are Zener diodes. What you're after is the forward current, which is part of standard diodes:

1N4001 forward current at temperature

Image courtesy Wikipedia 1N400x General-Purpose Diode.

This image shows a typical forward current for the venerable 1N400x series diode, along with the corresponding forward voltage, and for three temperatures. Zeners behave similarly to this, but typically have a slightly higher forward transition voltage.

So in summary, the forward voltage of the Zener isn't going to matter to the NEMOS for clamping purposes. It could matter, if that line were connected to a microcontroller input however. Because a reverse spike on GateDrv will allow it to go below ground by nearly a volt, which many chips will not like at all. Research latch-up for more info.

\$\endgroup\$
6
\$\begingroup\$

There are some problems with your assumptions.

  • A Zener is chosen to protect the gate from exceeding its Vgs max or static max Vgss usually +/-20V so a 15V Zener is suggested.
  • Since the static gate current is low xx microamps, a 5W Zener is overkill. - all large power diodes also have large capacitance (nF) at 0V which can slow down PWM risetimes if fast rise times are required.

Consider the following FET gate requirements:

  • Vt=Vgs(th) only defines the knee Threshold voltage at some small fraction e.g. 0.1% of the rated current
  • RdsOn is rated at some Vgs that is at least 2 to 2.5 times Vt according to the square law of \$I_D=beta(V_{GS}-V_t)^2\$, k=beta is not usually given

Zener diodes are also rated at two current levels.

  • One at a small fraction of the rated power (5W in your case is overkill) often called the Knee threshold or \$V_{ZK}\$ where the incremental resistance, Zzk can be 100 times larger than Zzt at this lower voltage and thus offers poorer load regulation, but is useful to know.
  • The other is the suggested voltage Threshold, \$V_{ZT}\$ with a standard current optimized for good regulation where the incremental resistance \$Z_{ZT}\$ is often given.

Thus you can choose a < 100 mW Zener at 15 V if required at all. If your driver path is active and not open collector with inductive traces and less than 15V, then no zener is needed. In high voltage applications there may be series resistors to limit current and Zener clamps to limit voltage.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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