Message-ID: <20090516002814.olvw2hgt6g84ss0g@webmail4.kuleuven.be>
Date: Sat, 16 May 2009 00:28:14 +0200
From: Rik Strobbe <Rik.Strobbe@fys.kuleuven.be>
To: rsgb_lf_group@blacksheep.org
References: <4A0D91E3.9040602@btconnect.com>
	<20090515183514.mt168ada0ug4k4c4@webmail4.kuleuven.be>
	<4A0D9BAC.6060808@btconnect.com> <52125C6E9B21456EBEB404A18962150C@JimPC>
In-Reply-To: <52125C6E9B21456EBEB404A18962150C@JimPC>
MIME-Version: 1.0
Content-Disposition: inline
User-Agent: Internet Messaging Program (IMP) H3 (4.1.2)
Subject: Re: LF: Aerial current
Content-Type: text/plain;
	charset=ISO-8859-1;
	DelSp="Yes";
	format="flowed"
Content-Transfer-Encoding: quoted-printable
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org


Hello Jim, LF,

indeed, class D amplifiers will (within reasonable limits) "adapt" to =20
the load impedance.
At least as long as the load is resistive. I have noticed that my PA =20
(G3YXM design clone) reacts "alergic" to a complex load.
My problem is that the antenna feeding point is about 35m away from =20
the house, I need a 40m coax between TX and antenna. This complicates =20
tuning if the antenna (+ loss) impedance is different from 50 Ohm. Not =20
because the TX output is 50 Ohm but because the coax is 50 Ohm and =20
will transform any impedance different from 50 Ohm resistive.
If the antenna is brought to resonance (at 502.5) the impedance is 40 =20
Ohm resistive but this is tranformed by the coax to a complex =20
impedance (at 502.5) at the TX end.
In fact I have to tune the antenna several kHz above 502.5 (measured =20
at the antenna feeding point) to get a resistive load on 502.5 at the =20
TX end of the cable.
As so far I don't have remote tuning of the loading coil, getting the =20
antenna "tuned" is a bit labour intensive ;-)

73, Rik  ON7YD


Quoting James Moritz <james.moritz@btopenworld.com>:

> Dear Richard, LF Group,
>
> It is not generally the case that the output impedance of a PA is equal
> to its design load impedance, and so maximum power delivered to load is
> not the same thing as load impedance being 50 ohms or whatever. This is
> especially the case with switching type amplifiers, which will usually
> deliver more power to a load below the design value.
>
> With a switching type amplifier, assuming nothing in the PA is getting
> hot, most of the DC input power is being delivered to the load as RF
> (80%+ efficiency is normal). So checking DC supply voltage/current is
> similar to what it is when the PA is driving a dummy load will give a
> fairly good indication if the actual antenna load impedance is similar
> to the dummy load, or whether the output power is more or less than
> what is being delivered to the dummy load.
>
> Thermocouple-type RF ammeters are prone to be a bit erratic in their
> old age - a check on calibration is to measure the current into a dummy
> load, and compare with the current in the load calculated from the RF
> voltage measured with an oscilloscope or diode voltmeter, etc.
> Alternatively, DC or (preferably) 50Hz AC current can be used to check
> calibration.
>
> The definitive place to measure RF current is at the feed point of the
> antenna itself - this then allows a reasonable calculation of ERP to be
> made, in conjunction with calculating the radiation resistance from the
> antenna dimensions. As has been pointed out, there is generally some
> impedance transformation occuring between the TX feeder and the antenna
> feed point, so the currents will be different.
>
> The only real way to be sure whether the load impedance is what you
> expect is to have some sort of instrumentation to measure it. A
> suitable LF SWR bridge will tell you if the impedance deviates from the
> design value of the bridge. I use either a simple "scope-match" circuit
> in conjunction with an oscilloscope to indicate voltage/current/phase
> at the TX output, or my LF tuning meter design (see G3YXM's "features"
> page) to do something equivalent. Another possibility is to measure the
> impedance with an RF bridge.
>
> Cheers, Jim Moritz
> 73 de M0BMU