Hi Roger the FET should be rated to have a minimum operating voltage of at
least 4 times the supply volts......... I guess you are using 12v?? This is
quite inefficient with switching fets and the design impedance is very low.
Are you using the right grade capacitors? This is most important, they
should be Pulse Rated as used in Switch-Mode power supplies. Ordinary high
voltage caps will not pass the level of RF current and will often fuse the
connection between a foil and lead-out wire leading to a steadily lower
value of capacitance and a mis-tuned PA. This is the case even with low
power (a few watts) the RF current at low impedances can be several amps.
Alan G3NYK
----- Original Message -----
From: "Roger Lapthorn" <[email protected]>
To: <[email protected]>
Sent: Tuesday, November 24, 2009 8:27 PM
Subject: Re: LF: IRF510 500kHz PA stability issues - advice please
> Andy,
>
> Thanks for the quick feedback. The 500kHz antenna is a 5m long vertical
with
> a spiral top hat. It is fed via a (thickish) ferrite rod wound with the
> required inductance to resonate and this is tapped at the 50 ohm point
(very
> low down - just a couple of turns). Well that is the general idea!
>
> Yes the IRF510 is rated at 100V.
>
> 73s
> Roger
>
>
> 2009/11/24 Andy Talbot <[email protected]>
>
> > If you've studied the drain waveform when operating inrto a correct
> > load, and assuming its all within design guidelines, then the trouble
> > is because of nasty impedances presented by the antenna during the
> > tuning process.
> >
> > What sort of antenna matching are you using? The often suggested idea
> > of an an adjustable tap onto the loading coil can lead to damage as,
> > when the antenna is detuned, the small number of turns of the link,
> > with an untuned coil, appear not far different from a short circuit.
> > A Class E PA, when presented with a slightly inductive short circuit
> > can get a bit stressed.
> >
> > On my antenna over the years on 73, 137 now 500kHz the system has
> > always been fed in series, using a ferrite tapped transformer to match
> > load impedance. I kept well away from tapped or link windings as
> > they are not easy to calculate or predict reliably. WIth a series fed
> > system, a detuned antenna results in a high impedance. This may not
> > always good news either to some class-E designs, but all my LF/MF PAs
> > are completely open-circuit-proof voltage fed designs. The only
> > true class-E one here was for 1.8MHz, and can be seen at
> > http://www.g4jnt.com/TopbandPA.pdf . This was completely imune to
> > open and short circuits and seemed very robust; but I did use the
> > meatier and more robust higher voltage IRF540. Doesn't the IRF510 only
> > have a 100V rating?
> >
> > Andy
> > www.g4jnt.com
> >
> > This email has been scanned for damaging side-effects by the health
> > and safety police
> >
> >
> >
> > 2009/11/24 Roger Lapthorn <[email protected]>:
> > > I've been using a single IRF510 FET PA stage on 500kHz and it produces
> > > around 6-7W of RF. But, when adjusting my antenna I've managed to
damage
> > > several devices now. I think the stage is running somewhere towards
> > class-E
> > > looking at the PA drain waveform on the scope and I believe either the
> > stage
> > > is going unstable or it is running in a condition in which too much
> > current
> > > (and heat) is taken and the device fails.500kHz teh ant
> > >
> > > Are there any simple "rules of thumb" or design/biasing/decoupling
> > > guidelines to ensure that my devices don't destroy themselves with a
less
> > > than perfect load?
> > >
> > > Any simple advice would be appreciated.
> > >
> > > 73s
> > > Roger G3XBM
> > >
> > >
> > >
> > >
> > >
> > >
> > > --
> > > -------------------------------------------------------------
> > > http://g3xbm-qrp.blogspot.com/
> > > http://www.g3xbm.co.uk
> > > G3XBM GQRP 1678 ISWL G11088
> > >
> >
> >
>
>
> --
> -------------------------------------------------------------
> http://g3xbm-qrp.blogspot.com/
> http://www.g3xbm.co.uk
> G3XBM GQRP 1678 ISWL G11088
>
|
