new MosFETs on the block...

[ViennaTom]

Since Infineon has now pushed its OptiMOS series up to 200 and 250V, there seems to be a new high performance 200V MosFET:

Infineon IPP110N20N3 200V 11mOhm, gate charge only 65nC (looks like a new world record for TO220)

i guess it could be well used for half and full bridges up to ~ 180V.

For even higher bus voltages there is a 250V 20mOhm FET too...

(no I m not an infineon sales guy, but I am always curious whats new on the market since I still have an unfinished class D project ongoing)

anybody has experience with these parts or is considering to use them, e. g. instead of IRFP4668 or the like?

[darkfenriz]

Looks nice, but the miller capacitance of just 5pF must be a typo

[ViennaTom]

Originally Posted by darkfenriz

Looks nice, but the miller capacitance of just 5pF must be a typo

..looks like they specifiy this kind of low miller C throughout the series (just look at IPP200N25N3, IPP320N20N3 and so on...).
looking back at their track record, it wouldn't be the first time in history for Infineon to be miles ahead of their competitors in terms of innovation.. hmm..
..i think i ll try to get my hands on some samples..

[StevenOH]

Capacitances always need be referenced to some voltage:
5 pF @ 60V, but
200 pF @ 10V
and then some...
That's the price for high transconductance/low Ron and high current capability.

[Eva]

Most ringing problems in class D output stages arise from series resonance between the output capacitance of the MOSFET that is turned off and with high Vds, and all package and layout inductances (the MOSFET that is on and supply decoupling capacitors are closing the circuit). Snubbers are often required to damp this resonance to a reasonable degree.

200pF at 10V is nothing to worry about, and 5pF at 150V with a good layout will resonate at such a high frequency that snubbers won't have anything to damp (above 100Mhz non-RF-specific circuits tend to be self damped due to high losses). Snubbers are probably still required, but just to control dV/dt at the switching node in the voltage transition that happens after body diode recovery.

Interestingly, Infineon does not mention dv/dt and di/dt strength of the body diodes, at least not for IPP110N20N3. But IR neither mentions this data in IRFB4227 datasheet, yet they have proven to be very reliable below 750A/us, 15V/ns and 40A.

EDIT: With these MOSFET, due to the low capacitance, amplifiers with extremely low dead time and idle losses can be made.

[ViennaTom]

Originally Posted by Eva

Most ringing problems in class D output stages arise from series resonance between the output capacitance of the MOSFET that is turned off and with high Vds, and all package and layout inductances (the MOSFET that is on and supply decoupling capacitors are closing the circuit). Snubbers are often required to damp this resonance to a reasonable degree.

200pF at 10V is nothing to worry about, and 5pF at 150V with a good layout will resonate at such a high frequency that snubbers won't have anything to damp (above 100Mhz non-RF-specific circuits tend to be self damped due to high losses). Snubbers are probably still required, but just to control dV/dt at the switching node in the voltage transition that happens after body diode recovery.

Interestingly, Infineon does not mention dv/dt and di/dt strength of the body diodes, at least not for IPP110N20N3. But IR neither mentions this data in IRFB4227 datasheet, yet they have proven to be very reliable below 750A/us, 15V/ns and 40A.

Eva, are you effectively using the slow body diodes of IRFB4227. Body diode dv dt capability is only an issue if the body diode is hard recovered from a conducting state. Based on what I 've heard, at this voltage level its quite wise to use external freewheeling instead of slow high Qrr body diode thus saving reverse recovery losses at the expense of series diodes in the drain path and its losses. The new Infineon part is worse in therms of its internal diode...

[Eva]

IRFB4227 body diodes are not slow at all. In one of my circuits a pair of them fully recovers from 80A conduction in 80ns.This is not worse than an ultrafast diode, but without consuming extra space or heatsink area, and benefiting from Rds-on shunting the diode most of the time.

With the Infineon part and proper drive, Rds-on (.01r) would prevent body diode conduction almost completely.

[ViennaTom]

Originally Posted by Eva

IRFB4227 body diodes are not slow at all. In one of my circuits a pair of them fully recovers from 80A conduction in 80ns.This is not worse than an ultrafast diode, but without consuming extra space or heatsink area, and benefiting from Rds-on shunting the diode most of the time.

With the Infineon part and proper drive, Rds-on (.01r) would prevent body diode conduction almost completely.

ok slow or fast is relative. I mean in comparison to Vishay 200V schottky they are still slow.

[Eva]

What you fail to understand is that with external diode there is always a diode recovery process taking place, and conduction losses are increased substantially. On the other hand, with Rds-on shunting the diode, reverse recovery only happens during high current peaks when driving low impedances, and only for the amount of current that the diode is shunting from Rds-on, not for the full current. The amplifier I mentioned can do full power into 4 ohm without diode recovery, which is only required for 2 ohm operation.

In terms of efficiency (power efficiency, cost efficiency, space efficiency), the Rds-on shunting method wins with current state of the art 200V and 250V MOSFET. For 500V/600V I would still go for external diodes or IGBT.

[darkfenriz]

Doesn't flatter Crss mean less pulse width distortion?