Thank you for the confirmation :-) But that still leaves me uncertain of the
best way to create this high gate drive voltage. The only way I can see at
the moment is to create an additional isolated supply and reference its
ground to the emitter of the high side IGBT. And duplicate this again for
the second IGBT. This seems like an excessive hardware solution, is there a
better way?

-Denny

----- Original Message -----
From: "Russell McMahon" <apptech@PARADISE.NET.NZ>
To: <PICLIST@MITVMA.MIT.EDU>
Sent: Monday, September 01, 2003 4:51 AM
Subject: Re: [EE]: High Voltage H-bridge design


> > I *thought* (correct me if I'm wrong) that an IGBT needed 5V-15V,
> depending
> > on the device, gate-emitter to switch on. In the on condition it should
> have
> > less than 2V emitter-collector, unless I'm missing something (I usually
am
> > ;o)  ) that would necessitates a gate drive 3V-12V higher than Vdd.
> > (assuming standard H-bridge with the high side collectors connected to
> Vdd)
>
> That is essentially correct. The following ramblings essentially reinforce
> your assumptions:-)
>
> For practical purposes an IGBT appears as a FET input driving a bipolar
> output.
> The FET input requires a specificed Vgson to operate and this will
typically
> be, as you note, in the 5 to 15 volt range.
> The output is a bipolar transistor and will have Vsat(uration) voltage
> appear across it when fully tuned on. Unlike a FET which is essentially an
> ohmic channel when fully enhanced and has an on resistance Rdson and
> therefore a voltage which varies with current, IGBTs have a saturation
> voltage which is somewhat constant across a current range. IGBTs are
usually
> more appropriate at higher voltages and lower currents, and the on voltage
> may be rather higher than 2 volts.
>
> To turn the IGBT fully on the applied gate-emmitter voltage needs to be
the
> requisite Vge above the emitter. When fully on the emitter is vsat = a few
> volts below top rail so the gate will need to be most of Vge above emitter
> voltage.
> Depending on how high Vge is the gate will PROBABLY have to be driven
above
> the high rail.
>
> In some cases you MAY be able to design suich that the gate is driven only
> to high rail and still have enough drive, but this is liable to be the
> exception. Taking a psudo randomly selected IGBT (data sheet was to hand)
> HGTP10N40C1D, 10A, 400v,. Vceonmax = 2.5v. Vgeth =2v typical, 4.5v max BUT
> at Ic = 1ma
> ie by connecting gate to high rail you MAY get turn on but only with
minimal
> current flow.
> For full performance a Vge of 7v is required - in this case 4.5v above top
> rail.
>
> A topside FET driver with an opto coupler may indeed be an adequate
solution
> in some cases BUT a high side supply above positive rail will be needed
:-(.
>
>
>
>         Russell McMahon
>
> --
> http://www.piclist.com hint: The PICList is archived three different
> ways.  See http://www.piclist.com/#archives for details.

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.