Message-ID: <50238984.4060106@iup.uni-heidelberg.de>
Date: Thu, 09 Aug 2012 11:57:24 +0200
From: =?ISO-8859-1?Q?Stefan_Sch=E4fer?= <Stefan.Schaefer@iup.uni-heidelberg.de>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; de; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3
MIME-Version: 1.0
To: rsgb_lf_group@blacksheep.org
References: <CAKvcm2cwyEg8XqiD6S4cd9B1eBVdte4vY7xU2O-i5NFrODmgjg@mail.gmail.com>	<5022E802.30404@iup.uni-heidelberg.de>	<CAKvcm2cTgNV25y9do77SLN-o3HE8yxCivteVsLuQq9wQX6+3=w@mail.gmail.com>	<50230604.9040801@iup.uni-heidelberg.de> <CAKvcm2fNyGOH=zZeRJ80KZukrtcDHGLqbwXgBHNGPhBn+BZ8oQ@mail.gmail.com>
In-Reply-To: <CAKvcm2fNyGOH=zZeRJ80KZukrtcDHGLqbwXgBHNGPhBn+BZ8oQ@mail.gmail.com>
Subject: Re: LF: Caps for Class-E amp
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org


Hi Dimitris,

What i forgot to say: Actually i can not recommend a class E for /p 
purposes. A class E is a fine thing and it is fascinating to see that 
one can generate 1 kW with a single FET and > 95% efficiency and without 
any ferrite transformers. But the efficiency and voltages and currents 
are highly dependant on the load resistance and eventual reactive 
components. As long as the SWR is < 1:1.5 or so, it is fine but then it 
becomes critical.
I lost a FET in the field while using a 100m kite vertical antenna. The 
kite dropped and i wasn't watching for a minute and then BAANG! If you 
use the class E at home, in a fixed antenna installation with known and 
stable SWR, then this is fine. But if you are in the field and don't 
know the feed point impedance, i think a voltage- or current-mode class 
D using 2 FETs against ground and a ferrite transformer performs better. 
Here you also have more flexibility because you can use various taps 
(and so power levels and/or impedance matching) on the secondary winding...
For example, my MF class D 
https://dl.dropbox.com/u/19882028/MF/100W%20475kHz%20PA.png works at 12V 
as well and is much less critical regarding mismatch...

73, Stefan/DK7FC


Am 09.08.2012 03:08, schrieb Dimitrios Tsifakis:
>>> Output impedance is about 1.4 ohm. I have designed it to produce 50 W
>>> at 12 V and of course 200 W at 24V.
>>>        
>> Yes, good idea. In /p for moderate power and at home in QRO :-)
>>      
> That's exactly the idea :-) Also, if all components are selectedfor a
> QRO operation, they will probably survive any kind of dodgy condition
> you may encounter in the field.
>
>    
>>>    I work in multiples of 12 V as I
>>> normally use big SLA batteries to power my equipment. The FET is a
>>> monster IXYS 500V/55A FET which should be bombproof.
>>>        
>> But you could have much lower on-resistances when using a 100V FET. 200W at
>> 24V is no problem though. IXYS is mostly a good choice. But quite expensive.
>> Once you lost a pair of them you begin to think to go back to the IRFP
>> types, at least i have done so ;-)
>>      
> I did not buy that FET, I was given it by a friend who bought it for a
> project but ended up not using it. The Rds(on) is 0.09 ohm, which is
> not that bad. Higher Rds(on) will result in a bit more heat on the
> heat sink and slightly less power for the same supply voltage, but no
> other serious side effect. The biggest problem with using a monster
> FET is the Ciss which is about 10 nF. This makes it a bit more
> difficult to drive than other smaller FETs, even at 137 kHz. But I
> found that the TC4420 (6 A peak) FET driver that I had in my junk box,
> seems to be happy with that Ciss/frequency combination.
>
> 73, Dimitris VK1SV
>