MIME-Version: 1.0
In-Reply-To: <15c73ca5749.marcocadeddu@tin.it>
References: <15c73ca5749.marcocadeddu@tin.it>
From: Andy Talbot <andy.g4jnt@gmail.com>
Date: Sun, 4 Jun 2017 17:42:24 +0100
Message-ID: <CAA8k23R3m4P-ExB3e9rrCM3S_TxxOHYwC3G2qj80mgof-beB5g@mail.gmail.com>
To: rsgb_lf_group@blacksheep.org
Subject: Re: Re: R: Re: LF: Re: I: Fw: For today the FETs survived...
Content-Type: multipart/alternative; boundary="94eb2c1a08f23f8adf05512512c4"
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Status: O

--94eb2c1a08f23f8adf05512512c4
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Yes.
You need to adjust the link so current is driven back into the supply when
RF amps exceed the maximum that you have designed for.   Up to that point
there should be no guard current as the induced voltage in the link, when
rectified , is less than the supply.

If the link coupling were to be perfect, ie it behaved as a transformer,
then at full power into your designed load the voltage across the tank coil
would be Q * Vdc * 4/pi  * SQRT(2)    (since you calculated Rload this
way).    The SQRT terms comes from the DC being the peak

So the turns ratio (for perfect coupling) if made equal equal to 1 : Q * 4
/ pi * SQRT(2) ,will start to generate load current just as maximum output
is reached.

But things aren't perfect; the mutual coupling is never 1 in an air wound
coil (mine is about 0.6), the rectification doesn't   give exactly peak
volts, so you have to experiment, changing the link turns until you get the
limiting you want.   Fortunately it scales with DC volts, so guard circuit
testing can be done at low supply / power out.

'jnt

On 4 June 2017 at 16:47, marcocadeddu@tin.it <marcocadeddu@tin.it> wrote:

> heyheyhey..
> now I'm testing with the new coil still at 50Vdc ;-)
> ops I was forgetting: without dissipator :-)) just with the mounting
> alu block it runs at room temp, maybe it is worth later at full power!
> The
> picture of Andy is not in thema: is the consequence of qro ampli on
> undersize external loading coil :-)
> By the way when this ampli will finally run on the antenna I need to re-
> size the variometer because it warms already with 250W..
>
> Also the new tank coil with 2,5mm=C2=B2 wire after 1 hour has a bit of
> fever.. but is not warm as the former.
>
> Just a quick question Andy: speaking of the guard circuit, if I read
> Iguard whithout mismatch is correct the assumption that the link has
> too many turns?
>
> Alan: I'm vintage.. and I'm sure I have a variac in the basement :-)
> thanks for the hint ;-)
>
> Thanks all for the psycological and praxi help !!
> Marco
>
> ----Messaggio originale----
> Da: andy.g4jnt@gmail.com
> Data: 4-giu-2017 17.49
> A: <rsgb_lf_group@blacksheep.org>
> Ogg: Re: R: Re: LF: Re: I: Fw: For today the FETs survived...
>
> Something like this ...
> http://www.g4jnt.com/QRO_LF_DoesThis.JPG
> 'jnt :-)
>
>
> On 4 June 2017 at 16:25, Chris Wilson <dead.fets@gmail.com> wrote:
>
> > Hello Alan,
> >
> >
> > No, what he needs is a video camera, and the courage to post the
> > result here for scrutiny, we need a bit of fun now and again ;)
> >
> > Good luck Marco, what's the worst that can happen.... :)
> >
> >
>
>
>
>
>

--94eb2c1a08f23f8adf05512512c4
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Yes.<div>You need to adjust the link so current is driven =
back into the supply when RF amps exceed the maximum that you have designed=
 for. =C2=A0 Up to that point there should be no guard current as the induc=
ed voltage in the link, when rectified , is less than the supply.<div><br><=
/div><div>If the link coupling were to be perfect, ie it behaved as a trans=
former, then at full power into your designed load the voltage across the t=
ank coil would be Q * Vdc * 4/pi =C2=A0* SQRT(2) =C2=A0 =C2=A0(since you ca=
lculated Rload this way). =C2=A0 =C2=A0The SQRT terms comes from the DC bei=
ng the peak</div><div><br></div><div>So the turns ratio (for perfect coupli=
ng) if made equal equal to 1 : Q * 4 / pi * SQRT(2) ,will start to generate=
 load current just as maximum output is reached.=C2=A0</div><div><br></div>=
<div>But things aren&#39;t perfect; the mutual coupling is never 1 in an ai=
r wound coil (mine is about 0.6), the rectification doesn&#39;t =C2=A0 give=
 exactly peak volts, so you have to experiment, changing the link turns unt=
il you get the limiting you want. =C2=A0 Fortunately it scales with DC volt=
s, so guard circuit testing can be done at low supply / power out.</div><di=
v><br></div><div>&#39;jnt</div></div></div><div class=3D"gmail_extra"><br><=
div class=3D"gmail_quote">On 4 June 2017 at 16:47, <a href=3D"mailto:marcoc=
adeddu@tin.it">marcocadeddu@tin.it</a> <span dir=3D"ltr">&lt;<a href=3D"mai=
lto:marcocadeddu@tin.it" target=3D"_blank">marcocadeddu@tin.it</a>&gt;</spa=
n> wrote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;b=
order-left:1px #ccc solid;padding-left:1ex">heyheyhey..<br>
now I&#39;m testing with the new coil still at 50Vdc ;-)<br>
ops I was forgetting: without dissipator :-)) just with the mounting<br>
alu block it runs at room temp, maybe it is worth later at full power!<br>
The<br>
picture of Andy is not in thema: is the consequence of qro ampli on<br>
undersize external loading coil :-)<br>
By the way when this ampli will finally run on the antenna I need to re-<br=
>
size the variometer because it warms already with 250W..<br>
<br>
Also the new tank coil with 2,5mm=C2=B2 wire after 1 hour has a bit of<br>
fever.. but is not warm as the former.<br>
<br>
Just a quick question Andy: speaking of the guard circuit, if I read<br>
Iguard whithout mismatch is correct the assumption that the link has<br>
too many turns?<br>
<br>
Alan: I&#39;m vintage.. and I&#39;m sure I have a variac in the basement :-=
)<br>
thanks for the hint ;-)<br>
<br>
Thanks all for the psycological and praxi help !!<br>
Marco<br>
<br>
----Messaggio originale----<br>
Da: <a href=3D"mailto:andy.g4jnt@gmail.com">andy.g4jnt@gmail.com</a><br>
Data: 4-giu-2017 17.49<br>
A: &lt;<a href=3D"mailto:rsgb_lf_group@blacksheep.org">rsgb_lf_group@blacks=
heep.org</a>&gt;<br>
Ogg: Re: R: Re: LF: Re: I: Fw: For today the FETs survived...<br>
<br>
Something like this ...<br>
<a href=3D"http://www.g4jnt.com/QRO_LF_DoesThis.JPG" rel=3D"noreferrer" tar=
get=3D"_blank">http://www.g4jnt.com/QRO_LF_<wbr>DoesThis.JPG</a><br>
&#39;jnt :-)<br>
<br>
<br>
On 4 June 2017 at 16:25, Chris Wilson &lt;<a href=3D"mailto:dead.fets@gmail=
.com">dead.fets@gmail.com</a>&gt; wrote:<br>
<br>
&gt; Hello Alan,<br>
&gt;<br>
&gt;<br>
&gt; No, what he needs is a video camera, and the courage to post the<br>
&gt; result here for scrutiny, we need a bit of fun now and again ;)<br>
&gt;<br>
&gt; Good luck Marco, what&#39;s the worst that can happen.... :)<br>
&gt;<br>
&gt;<br>
<br>
<br>
<br>
<br>
</blockquote></div><br></div>

--94eb2c1a08f23f8adf05512512c4--