Content-Transfer-Encoding: 8bit
Message-ID: <3617AC3245C2D1118A840000F805359C01AB8CAC@pdw-mercury-1.dera.gov.uk>
From: "Talbot Andrew" <ACTALBOT@dera.gov.uk>
To: rsgb_lf_group@blacksheep.org
Subject: LF: RE: LF PA with tube...
Date: Thu, 8 Jun 2000 12:52:44 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>


>Anybody have an idea how to build an power amplifier with single tube ?
>I have some GU81's. They needs about Ua=2000 V and Ia(max)=600mA.
>I calculated Pi match, but values of capacitors are to high...anybody
knows
>how to do it other way ?
>Thank you very much for all info...

>Rich OM2TW

Sell it and get some power Mosfets..........

More seriously though, what is wrong with this pi-network ?  assuming an
Rl for the valve of 2000V / 0.6 A  = 3300 ohms  which will allow over
1kW out, a pi match with Q of 12 to get to 50 ohms needs

Cin = 4nF (rated 4kV plus) 
L = 361uH
Cout 23.3nF	Rated 500V.

The first one is slightly tricky, but high voltage ceramic caps up to
10kV do exist up to a few nF do exist, so connect them in parallel to
tune.  Cout is no problem.  The inductor could be 140 - 150 turns of 2mm
wire on an 80mm diameter former.  Get hold of some Litz wire from old
Decca transmitter loading coils.   The DC choke will be interesting
though.   An LOT of turns on a ferrite core, and remember it has to be
gapped or a dust core as it has DC flowing.  Alternatively feed in the
DC through a more modest choke at the 50 ohm point in which case Cout
has to be rated at 3kV and another blocking cap of say 100n rated 3kV is
needed.

Alternatively find the biggest ferrite core suitable for switch mode PSU
you can lay your hands on and wind an output transformer. Even stack
cores if you have to.   Minimum number of turns for a given ferrite core
area is given by :   N = Vrf(RMS) / (4.44. F . A . B)
Using B = 0.2 Tesla (suitable for most SMPSU ferrite materials), A = 300
x 10^-6 m^2 for a large SMPSU core and V = 3kV, at 137kHz gives minimum
number of turns of just 55.  Transformation ratio = SQRT(3300 / 50) =
8.2 : 1   Remember the winding has to be insulated for 3 - 4kV and DC
flowing through it could present a problem of saturation, so monitor
core temperature carefully. 100deg is usually OK provided the wire can
take it.  You will also need an output low pass filter but component
values here are a bit less restrictive.

It is all a lot easier, even at the 1kW level, with solid state devices.

Andy  G4JNT

> ----------
> From: 	Gasparik Richard[SMTP:Richard.Gasparik@siemens.sk]
> Reply To: 	rsgb_lf_group@blacksheep.org
> Sent: 	2000-06-08 10:18
> To: 	Rsgb_Lf_Group (E-mail)
> Subject: 	LF: LF PA with tube...
> 
> Hi all...
> 


-- 
The Information contained in this E-Mail and any subsequent correspondence
is private and is intended solely for the intended recipient(s).
For those other than the recipient any disclosure, copying, distribution, 
or any action taken or omitted to be taken in reliance on such information is
prohibited and may be unlawful.