Message-ID: <001701cd9044$bdf83a70$0501a8c0@xphd97xgq27nyf>
From: "mal hamilton" <g3kevmal@talktalk.net>
To: <rsgb_lf_group@blacksheep.org>
References: <dprrumkv6c5blgnl5dka0tdm.1347356848062@email.android.com> <504F27D5.8040305@iup.uni-heidelberg.de>
Date: Tue, 11 Sep 2012 17:41:56 -0000
MIME-Version: 1.0
Subject: Re: LF: MF PA
Content-Type: text/plain;
	charset="UTF-8"
Content-Transfer-Encoding: 8bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org


Take a look at the cct that the Decca 5501 uses and copy this for robust and
failsafe resualts. A series PA tank cct resonated at the frequency of
interest  is desireable. This is what I am currently using on both 500 es
137 and find it very reliable. The module gets barely warm but of course I
am not using it for BROADCAST mode at qrs 6000000000000000
I use the amp in QSO mode mostly and QRSS no slower than 30 sec dot but
usually QRSS 3 - 10
de mal/g3kev



----- Original Message -----
From: "Stefan Schäfer" <Stefan.Schaefer@iup.uni-heidelberg.de>
To: <rsgb_lf_group@blacksheep.org>
Sent: Tuesday, September 11, 2012 12:00 PM
Subject: Re: LF: MF PA


> Hi Christian,
>
> > -------- Originalnachricht --------
> > Betreff: Re: LF: MF PA
> > Von: "C. Groeger"<aot.aot@gmx.de>
> > An: rsgb_lf_group@blacksheep.org
> > Cc:
> >
> > Hi Stefan,
> > Thanks for these infos!
> > This is always helpful because I'm not a really good technican.
Everything I built was a mixtiure of my ideas and stuff from the net...
> > Output power is 100 w, measured with a 100 w bulb, matched to 50 ohms
via transformer. Thus efficiency is ok, I think.
> > The overall setup is:
> > Vxo/xo 10 mc/10.24 mc, mixer, Fd,
> > PreDriver 2n2219, Driver irf640, pa 2 x irf 540.
> > I have attached a foto...
> > 73, df5qg
> >
>
> There is no photo attached. Note that attachments on this reflector must
> not exceed 400 kB (or so). Otherwiese the email is rejected. Better use
> a dropbox link for large attachments, so people can decide if they want
> to have a look at it before downloading all the stuff.
>
> Well, 100W out at 120W in is 83% efficiency. And 20W losses can't be a
> problem for 2 TO-220 FETs as long as you use a suitable heat sink and
> reasonable thermal coupling.
>
> But i'm not sure how you measure 100W RF power with a 100W bulb. Is
> there no reactive component of the impedance of that bulb when running
> at 475 kHz?
>
> Take a look to the gate voltage, be sure that it is a real rectangular
> wave. Also run the PA in CW-key-down without connecting a load or a low
> pass filter and check if the output transformer becomes warm or hot.
> Maybe the core material is not suitable or the number of primary turns
> is to low?  I don't expect overvoltage problems from a 100V FET at 12V
>   DC supply voltage. Just some ideas...
>
> 73, Stefan/DK7FC
>
>