The main reasons for the heating of MOS tubes with small internal resistance include the following:

Power loss: when the MOS tube flows a large current, even if the internal resistance is very small, it will produce a large power loss, resulting in heating. In switching power supplies, if the MOS tube does not turn off and on fast enough, additional power losses are also generated12.

Linear operating state: If the MOS tube operates in the linear region instead of the switching region, too long a conduction time will result in operating in the linear region, increased equivalent DC impedance, and increased voltage drop, which will increase power loss and heat generation23.

Switching frequency is too high: in the pursuit of miniaturization, increasing the operating frequency will lead to increased losses on the MOS tube and significant heat generation problems23.

Insufficient heat dissipation design: If the heat dissipation design is insufficient, even if the current does not exceed the nominal value, it may lead to serious heat generation due to poor heat dissipation2.

Improper selection: When selecting MOS tubes, improper selection of internal resistance can also lead to increased switching impedance and increased power loss, thus causing heat generation23.

External factors: External short-circuit or disconnection, failure to cut off the circuit after the over-current protection action, excessive load, high ambient temperature and other factors can also lead to MOS tube heating4.

Source, Drain, Gate corresponds to the three poles of the field effect tube: source S, drain D, gate G (in this does not speak of gate GOX breakdown, only for the drain voltage breakdown).

I. Which kinds of MOSFET breakdown?

First of all, the test conditions, are the source gate substrate are grounded, and then scan the drain voltage until the Drain terminal current reaches 1uA. so from the device structure, its leakage channel has three: Drain to source, Drain to Bulk, Drain to Gate.

1. Drain - “Source through the breakdown

This is mainly Drain plus reverse bias voltage, making Drain/Bulk PN junction depletion area extension, when the depletion area touches the Source, the source leakage between the need to open the formation of the pass, so it is called through (punch through).

How to prevent the punch through? This goes back to the diode reverse bias characteristics, the depletion zone width in addition to the voltage, but also with the doping concentration on both sides, the higher the concentration can inhibit the width of the depletion zone extension, so there is an anti-punch through injection inside the flow (APT: AnTI Punch Through), remember that it has to hit the same type of specis and well.

Of course, the actual encounter WAT BV ran and sure from the source side, may also depend on whether the PolyCD or Spacer width, or LDD_IMP problem.

So how do you rule that out? It depends on whether you have both NMOS and PMOS running?POLY CD can be verified by Poly related WAT.

For penetration breakdown, there are some characteristics as follows:

(1) The breakdown point of penetration breakdown is soft, and during the breakdown process, the current is characterized by a gradual increase, which is due to the wider depletion layer spreading, which generates a larger current.

On the other hand, the depletion layer spreads wide and is prone to the DIBL effect, so that the positive bias of the source substrate junction is characterized by a gradual increase in current.

(2) The soft breakdown point of penetration breakdown occurs when the source-drain depletion layer is connected, at which time the carriers at the source end are injected into the depletion layer and accelerated to the drain end by the electric field in the depletion layer.

Therefore, the penetration breakdown current also has a sharp increase in the point, this sharp increase in current and avalanche breakdown current sharp increase is different, this time the current is equivalent to the source substrate PN junction positive conduction current, and avalanche breakdown current is mainly for the PN junction reverse breakdown avalanche current, such as no current limitation, avalanche breakdown current to be large.

(3) penetration breakdown generally does not appear destructive breakdown. Because the penetration breakdown field strength does not reach the avalanche breakdown field strength, does not produce a large number of electron hole pairs.

(4) Penetration breakdown generally occurs in the channel body, the channel surface is not prone to penetration, which is mainly due to the channel injection to make the surface concentration than the concentration caused by the large, so the NMOS tube generally have anti-penetration injection.

(5) In general, the concentration at the edge of the beak is larger than the concentration in the middle of the channel, so the penetration breakdown generally occurs in the middle of the channel.

(6) The polycrystalline gate length has an effect on the penetration breakdown, as the gate length increases, the breakdown increases. The avalanche breakdown, strictly speaking, also has an effect, but not so significant.

2. Drain - “Bulk avalanche breakdown

This is simply a PN junction avalanche breakdown (avalanche Breakdown), mainly under the drain reverse bias voltage makes the PN junction depletion zone broadening, then the reverse bias electric field added to the PN junction reverse bias above, so that the electrons accelerated impact lattice to produce a new pair of electrons hole pair (Electron-Hole pair), and then the electrons continue to hit the avalanche so that the avalanche doubled down to lead to the breakdown, so this breakdown, so this breakdown is strictly speaking, but not so significant. Breakdown, so this breakdown current increases almost rapidly, the I-V curve almost vertical up, very easy to burn. (This is not the same as source-drain-through breakdown.)

So how to improve this juncTIon BV? So mainly from the characteristics of the PN junction itself, we must reduce the depletion zone electric field, to prevent collision of electron-hole pairs, reduce the voltage certainly can not, that can only increase the width of the depletion zone, so we have to change the doping profile, which is why the breakdown voltage of the Abrupt junction (Abrupt juncTIon) than slow-change junctions (Graded) JuncTIon.

Of course, in addition to the doping profile, there is also the doping concentration, the larger the concentration, the narrower the width of the depletion region, so the stronger the electric field strength, which will certainly reduce the breakdown voltage. And there is a law that the breakdown voltage is usually by the low concentration of the concentration of the side of the greater impact, because of the depletion zone width of the side of the large.

The formula is BV = K * (1 / Na + 1 / Nb), from the formula can also be seen in the Na and Nb concentration if the difference of 10 times, almost one of them can be ignored.

Then the actual process if you find that the BV becomes smaller, and confirm that it is from the junction away, then check your Source/Drain implant.

3. Drain - “Gate breakdown

This is mainly between Drain and Gate Overlap caused by the gate oxide layer breakdown, this is somewhat similar to GOX breakdown, of course, it is more like Poly finger GOX breakdown, so he may be more care poly profile and sidewall damage. Of course this Overlap there is a problem is GIDL, this will also contribute to the Leakage makes the BV lower.

The above is the three channels of MOSFET breakdown, usually the first two cases of BV are mostly.

Above are Off-state breakdown, that is, when the Gate is 0V, but there are times when the Gate is open under the Drain plus voltage is too high will also lead to breakdown, we call it On-state breakdown.

This situation especially like to happen in the Gate lower voltage, or tube just turned on, and almost always NMOS, so we usually WAT will also test BVON.

The above is the MOSFET breakdown of the three channels, usually BV case of the first two are mostly.

Above are Off-state breakdown, that is, when the Gate is 0V, but there are times when the Gate is open under the Drain plus voltage is too high will also lead to breakdown, we call it On-state breakdown.

This situation especially like to happen in the Gate lower voltage, or tube just turned on, and almost always NMOS, so we usually WAT will also test BVON.

II. How to deal with mos tube with severe heat up at low current?

Mos tube, do power supply design, or do the driver side of the circuit, it is inevitable to use the MOS tube. mos tube has many kinds, also has a lot of role. Do power supply or drive use, of course, is to use its switching role.

Regardless of N-type or P-type MOS tube, its working principle is essentially the same, MOS tube is added to the input gate voltage to control the output drain current.

MOS tube is a voltage-controlled device it is added to the gate of the voltage control device characteristics, does not occur as a triode to do switching due to the base current caused by the charge storage effect, therefore, in the switching application, MOS tube switching speed should be faster than the triode.

We often look at the PDF parameters of the MOS tube, MOS tube manufacturers use the RDS (ON) parameter to define the on-state impedance, for switching applications, RDS (ON) is also the most important device characteristics.

The datasheet defines RDS(ON) in relation to the gate (or drive) voltage VGS and the current flowing through the switch, but for adequate gate drive, RDS(ON) is a relatively static parameter. MOS tubes that are always on are prone to heating. In addition, a slowly increasing junction temperature will also cause an increase in RDS(ON).

The MOS tube datasheet specifies the thermal impedance parameter, which is defined as the ability of the semiconductor junction of the MOS tube package to dissipate heat.The simplest definition of RθJC is the junction-to-case thermal impedance.

1. The reason for the heat generation of mos tube with small current:

1) The problem of circuit design: it is to make the MOS tube work in linear operation, not in switching state, which is also a cause of heat generation of MOS tube.

If the N-MOS is doing the switching, the G-level voltage has to be a few V higher than the power supply in order to turn on completely, and the opposite is true for the P-MOS. Not fully open and the voltage drop is too large resulting in power consumption, the equivalent DC impedance is relatively large, the voltage drop increases, so U * I also increases, the loss means heat. This is the most avoidable error in the design of the circuit.

2) Frequency is too high: mainly because sometimes over-pursuit of the volume, resulting in increased frequency, the loss on the MOS tube increases, so the heat is also increased.

3)Not good enough heat dissipation design: the current is too high, the nominal current value of the MOS tube, generally need good heat dissipation to achieve. So the ID is less than the maximum current, it may also heat up badly, need enough auxiliary heat sink.

4) MOS tube selection is wrong: the power judgment is wrong, MOS tube internal resistance is not fully considered, resulting in increased switching impedance.

2. Mos tube small current heating serious how to solve:

0 do a good job of MOS tube heat dissipation design, add enough auxiliary heat sinks.

Paste heat dissipation adhesive.

III. Why does the MOS tube prevent the power supply from being reversed?

Power reverse connection will cause damage to the circuit, however, power reverse connection is unavoidable. Therefore, we need to add a protection circuit to the circuit to achieve the purpose of not being damaged even if the power supply is reversed.

Generally can be used in the power supply of a diode series into a diode solution, however, because the diode has a voltage drop, will cause unnecessary losses to the circuit, especially battery-powered occasions, the original battery voltage of 3.7V, you use the diode to drop 0.6V, so that the battery use time is greatly reduced.

MOS tube anti-reverse connection, the advantage is that the voltage drop is small, small to almost negligible. Nowadays, MOS tubes can do a few milliohms of internal resistance, assuming 6.5 milliohms, through the current of 1A (this current is already very large), in his above the voltage drop is only 6.5 millivolts.