Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on lipkowski.org X-Spam-Level: X-Spam-Status: No, score=-2.3 required=5.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, HTML_MESSAGE,RCVD_IN_DNSWL_MED,SPF_PASS autolearn=ham autolearn_force=no version=3.4.0 X-Spam-DCC: EATSERVER: mailn 1166; Body=2 Fuz1=2 Fuz2=2 Received: from post.thorcom.com (post.thorcom.com [195.171.43.25]) by lipkowski.org (8.14.4/8.14.4/Debian-8+deb8u1) with ESMTP id v390H8iJ011360 for ; Sun, 9 Apr 2017 02:17:09 +0200 Received: from majordom by post.thorcom.com with local (Exim 4.14) id 1cx0Tv-0004uI-I7 for rs_out_1@blacksheep.org; Sun, 09 Apr 2017 01:13:27 +0100 Received: from [195.171.43.32] (helo=relay1.thorcom.net) by post.thorcom.com with esmtp (Exim 4.14) id 1cx0Tu-0004u9-5o for rsgb_lf_group@blacksheep.org; Sun, 09 Apr 2017 01:13:26 +0100 Received: from smtpout5.wanadoo.co.uk ([80.12.242.80] helo=smtpout.wanadoo.co.uk) by relay1.thorcom.net with esmtp (Exim 4.89) (envelope-from ) id 1cx0Tp-0001Jk-5o for rsgb_lf_group@blacksheep.org; Sun, 09 Apr 2017 01:13:24 +0100 Received: from AGB ([95.151.119.238]) by mwinf5d63 with ME id 60DL1v00658iHV3030DLhK; Sun, 09 Apr 2017 02:13:20 +0200 X-ME-Helo: AGB X-ME-Date: Sun, 09 Apr 2017 02:13:20 +0200 X-ME-IP: 95.151.119.238 Message-ID: <42119DC3899048C78B8B0C63BEE6579F@AGB> From: "Graham" To: References: <68858815.2799186.1491674334546.ref@mail.yahoo.com> <68858815.2799186.1491674334546@mail.yahoo.com> <5EFDD5BF98D64A9394F25F2E55867EFB@gnat> In-Reply-To: <5EFDD5BF98D64A9394F25F2E55867EFB@gnat> Date: Sun, 9 Apr 2017 01:13:20 +0100 MIME-Version: 1.0 X-Priority: 3 X-MSMail-Priority: Normal Importance: Normal X-Mailer: Microsoft Windows Live Mail 14.0.8117.416 X-MimeOLE: Produced By Microsoft MimeOLE V14.0.8117.416 X-Antivirus: Avast (VPS 170408-0, 08/04/2017), Outbound message X-Antivirus-Status: Clean X-Scan-Signature: 31e464b680e3427073f16da582bb11f1 Subject: Re: LF: Wilkinson combiners Content-Type: multipart/alternative; boundary="----=_NextPart_000_0044_01D2B0CE.79E57C80" X-SA-Exim-Scanned: Yes Sender: owner-rsgb_lf_group@blacksheep.org Precedence: bulk Reply-To: rsgb_lf_group@blacksheep.org X-Listname: rsgb_lf_group X-SA-Exim-Rcpt-To: rs_out_1@blacksheep.org X-SA-Exim-Scanned: No; SAEximRunCond expanded to false X-Scanned-By: MIMEDefang 2.75 Status: RO X-Status: X-Keywords: X-UID: 11170 This is a multi-part message in MIME format. ------=_NextPart_000_0044_01D2B0CE.79E57C80 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable This was the original issue What I am seeing is each amp giving a nice (well, they seem nice to me...) drain and gate waveforms individually, but when combined both waveforms deteriorate and the cold running FET's and output transformers heat up quite dramatically Based round, what appears to be detuning ?? , what's missing , is the = filtered sine wave , then the combiner , the most I have seen for HF is the Racal 10 module 1Kw HF amp and , 1= 6 in a ' tree ' uhf wide band system , using 4:1 combiners 73-G,. -------------------------------------------------- From: "Alan Melia" Sent: Saturday, April 08, 2017 9:30 PM To: Subject: Re: LF: Wilkinson combiners > John this is a common misconception, these PAs produce something close to= a > sinewave not an RF square wave. I am not sure of Chris's design but Class= E > generally has a THD of about 7%. > See Prof. Nat Sokal's original QST article. They are however sensitive to= > load reactance changes. > > There is another way of combining 2 Class E stages that was devised By Jo= han > Bodin, but again I am not sure it applies to this design. > > Alan > G3NYK > > ----- Original Message ----- > From: "John Fisher" > To: > Cc: > Sent: Saturday, April 08, 2017 6:58 PM > Subject: Re: LF: Wilkinson combiners > > > > Hi Chris, > > To me, it doesn't sound like a good idea to combine two switching square = > wave signals together...Might be okay if they were synchronized...Better = to > filter them into sine waves first and then combine them...More expensive = for > double the number of iron powder toroids though... > > Jay is combining them ahead of the filter...That type of combiner must wo= rk > okay that way. > > Cheers, > John VA3VVV > > > -------------------------------------------- > On Sat, 4/8/17, Chris Wilson wrote: > > Subject: Re: LF: Wilkinson combiners > To: "Markus Vester" > Received: Saturday, April 8, 2017, 12:42 PM > > Hello Markus, > > > Thanks for the detailed reply, > let me see if i have this straight > though > please, sorry if I appear a bit dense, it's probably > because I > am.... :) > > Right now I run one voltage mode Class D push > pull amp into a T type > low pass filter. All > works great, nice gate and drain wave forms, and > my Scopematch box shows perfect sinusoidal wave > forms for voltage and > current after the T > type filter. Modest output transformer warming, > each pair of FET's near as damn it cold. > > I add another duplicate amp > and the Wilkinson combiner. The combiner > feeds the T type filter. With either one or > both amps running the gate > and drain wave > forms go to pot, the FET's run mad hot, as does the > output transformer (and the power input choke > gets hot, too, which it > doesn't with a > single amp straight into the filter). > > Are you suggesting I build a duplicate T type > filter bank and put a > filter between each > amp and the combiner, with no filter on the output > of the combiner, just run the antenna co-ax to > the matching > transformer and loading coil > which are outside? I can do this, it > means > finding and buying some more big toroids, but it's > perfectly > possible.... > > Is the W1VD combiner what people call a > "broadband" combiner? And if I > built this instead of the Wilkinson one would I > be able to run it in > front of my single T > type filter? Or do I still need a pair of filters > between it and the two amp's outputs? To > remind, the W1VD one is here: > > http://www.w1vd.com/137-500-500WCombiner.pdf > > Thanks for your patience, > and thanks Graham for your input too, just > don't want to waste time and money > building things that are incorrect > due to my > ignorance! > > > > > Hi Chris, > > > the PI-type combiner per se is not the > problem. All I was trying to > > say is > that you have to prevent it from short-circuiting the > > harmonics directly at the output of a > (voltage-mode) class-D PA. In > > that > case, I would recommend a T-type low-pass filter between > each > > PA and combiner input, starting > with a series inductor rather than a shunt capacitor. > > > Note that the desired > harmonic load impedance depends on the type > > of Class-D PA: Voltage-mode class D with > rectangular voltage and > > sinusoidal > current should see a high impedance termination for > > harmonics, i.e. an inductor or series > resonant circuit. On the other > > hand, > current-mode class D designs would have sinusoidal voltage > and > > rectangular current, and require a > low-impedance shunt for harmonics. > > > Best 73, > > Markus > > > > > > -- > Best regards, > Chris > mailto:dead_fets@gmail.com > > > > > --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ------=_NextPart_000_0044_01D2B0CE.79E57C80 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
This was the original issue

What I am seeing is ea= ch amp giving a nice (well, they seem nice to
me...) drain and gate wavefor= ms individually,

but when combined both

waveforms deteriorate a= nd the cold running FET's and output
transformers heat up quite dramatically

Based round, what appears to be  det= uning ??  , what's   missing , is the  filtered sine wave  , then  the  combiner , 
 
the  most I have  seen  for HF is  the Racal = 10 module  1Kw HF amp and , 16 in a  ' tree '  uhf wide band  system , using  4:1 combiners
 
73-G,.
 
 

--------------------------------------------------
From: "Alan = Melia" <alan.melia@btinternet.com>
Sent: Saturday, April 08, 2017 = 9:30 PM
To: <rsgb_lf_group@blacksheep.org>
Subject: Re: LF: Wilkinso= n combiners

> John this is a common misconception, these PAs produc= e something close to a
> sinewave not an RF square wave. I am not sure= of Chris's design but Class E
> generally has a THD of about 7%.
>= ; See Prof. Nat Sokal's original QST article. They are however sensitive to
&= gt; load reactance changes.
>
> There is another way of combining = 2 Class E stages that was devised By Johan
> Bodin, but again I am not= sure it applies to this design.
>
> Alan
> G3NYK
>
= > ----- Original Message -----
> From: "John Fisher" <johnfish837@yahoo.ca>
> To: <rsgb_lf_group@blacksheep.org>
> Cc: <dead.fets@gmail.com>
> Sent: Saturday, April 08, 2017 6:58 PM
> Subject: Re: LF: Wilkinson combiners
>
>
>
> Hi Chris,
>
> To me, it doesn't sound like a good ide= a to combine two switching square
> wave signals together...Might be okay= if they were synchronized...Better to
> filter them into sine waves fir= st and then combine them...More expensive for
> double the number of ir= on powder toroids though...
>
> Jay is combining them ahead of th= e filter...That type of combiner must work
> okay that way.
>
> Cheers,
> John VA3VVV
>
>
> --------------------------------------------
> On Sat, 4/8/17, Chris = Wilson <dead.fets@gmail.com> wrote:
>
> Subject: Re: LF:= Wilkinson combiners
> To: "Markus Vester" <rsgb_lf_group@blacksheep.org>
> Received: Saturday, April 8, 2= 017, 12:42 PM
>
> Hello Markus,
>
>
> Thanks fo= r the detailed reply,
> let me see if i have this straight
> though
> please, sorry if I appear a bit dense, it's probably
>= because I
> am.... :)
>
> Right now I run one voltage mo= de Class D push
> pull amp into a T type
> low pass filter. All> works great, nice gate and drain wave forms, and
> my Scopematch box = shows perfect sinusoidal wave
> forms for voltage and
> current after= the T
> type filter. Modest output transformer warming,
> each pair= of FET's near as damn it cold.
>
> I add another duplicate amp> and the Wilkinson combiner. The combiner
> feeds the T type filter. W= ith either one or
> both amps running the gate
> and drain wave
= > forms go to pot, the FET's run mad hot, as does the
> output transfor= mer (and the power input choke
> gets hot, too, which it
> doesn't = with a
> single amp straight into the filter).
>
> Are you suggesting I build a duplicate T type
> filter bank and put a
>= filter between each
> amp and the combiner, with no filter on the output
> of the combiner, just run the antenna co-ax to
> the matching
> transformer and loading coil
> which are outside? I = can do this, it
> means
> finding and buying some more big toroids,= but it's
> perfectly
> possible....
>
> Is the W1VD combiner what people call a
> "broadband" combiner? And if I
> = built this instead of the Wilkinson one would I
> be able to run it in
&= gt; front of my single T
> type filter? Or do I still need a pair of filters
> between it and the two amp's outputs? To
> remind, th= e W1VD one is here:
>
> http://www.w1vd.com/137-500-500WCombiner.pdf
>
> Thanks for yo= ur patience,
> and thanks Graham for your input too, just
> don't = want to waste time and money
> building things that are incorrect
> = due to my
> ignorance!
>
>
>
> > Hi Chris,
>
> > the PI-type combiner per se is not the
>= problem. All I was trying to
> > say is
> that you have to prevent it from short-circuiting the
> > harmonics directly at the= output of a
> (voltage-mode) class-D PA. In
> > that
>= case, I would recommend a T-type low-pass filter between
> each
&g= t; > PA and combiner input, starting
> with a series inductor rather = than a shunt capacitor.
>
> > Note that the desired
> harm= onic load impedance depends on the type
> > of Class-D PA: Voltage-mode= class D with
> rectangular voltage and
> > sinusoidal
>= ; current should see a high impedance termination for
> > harmonics,= i.e. an inductor or series
> resonant circuit. On the other
> > hand,
> current-mode class D designs would have sinusoidal voltage> and
> > rectangular current, and require a
> low-impedance s= hunt for harmonics.
>
> > Best 73,
> > Markus
>
>
>
>
>
> --
> Best regards,
>  Chris
>  mailto:dead_fets@gmail.com
&g= t;
>
>
>
>

Virus-= free. www.avast.com
------=_NextPart_000_0044_01D2B0CE.79E57C80--