I still think the problem is as someone else mentioned - not enough inducta=
nce. Now joe is using a larger diameter wire the results did seem to improv=
e, but it still needs sufficient inductance that the current through the in=
ductor doesn't drop to zero during the transistor off time. Remember Ldi/dt=
..

>=20
> On 29 Jan 2017 at 21:40, IVP wrote:
> > > It seems a bit strange to me to drive the gate of the mosfet from a
> > > supply which is referenced to +16V and GND when the source of the
> > > fet is at
> >
> > Well, as it's a P FET I was thinking upside down, but evidently not
> > thinking hard enough and it's come out as part N FET drive
> >
> > I've restored the circuit to how it originally was
> >
> >
> https://www.flickr.com/photos/97814409@N04/albums/72157675978832174
> >
> > Output is 2.2V into 6R8
> >
> > Gate voltage is from 1.3V to 8.8V. The test circuit in the ATP106 d/s
> > shows gate drive from 0V to -10V, with source at 0V, which would be
> > 0 to +10 for an N FET. As I figure it then, my gate is reaching only a
> > relative -7.5V
> >
> > So I should be taking the emitter of the BC327 to at least a couple of
> > volts below 0V if source is at +9V ? That would make the gate 11V
> > below the source
>=20
> Hi Joe,
>=20
> ATP106 datasheet talks about gate drive as low as 4.5V gate drive, though
> 10V obviously gives better Rdson. Gate cutoff voltage is specified at 1.5=
V min
> to 2.6V max (for 1mA drain current @ Vds =3D 10V), so your revised circui=
t &
> measured gate voltages of 1.3V and 8.8V seem within workable range. Even
> if interpreted differently and the 1.3 and 8.8V were measured wrt 0V, the=
n
> translating that would give Vgson =3D 9-1.3 =3D 7.7V, and Vgsoff =3D 9-8.=
8 =3D 0.2V, in
> other words still fine.
>=20
> Seems like drive circuit should be good enough. Sure some tweaking might
> help but doesn't seem like it would make the difference between working
> and not-working.
>=20
>=20
>=20
> --
> http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist

--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.