To: rsgb_lf_group@blacksheep.org References: <002201c46c3d$060963a0$0201a8c0@private.network> <004b01c46cc0$0d40fce0$07d98351@rogersservices> Date: Mon, 19 Jul 2004 00:05:46 -0700 MIME-Version: 1.0 X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1106 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 X-Spam-Score: 2.6 (++) X-Spam-Report: autolearn=no,FROM_ENDS_IN_NUMS=0.869,HTML_20_30=0.474,HTML_MESSAGE=0.001,MAILTO_TO_SPAM_ADDR=1.052,RCVD_IN_SORBS=0.1,TW_XF=0.077 Subject: LF: Re: Re: clicking matching transformer... Content-Type: text/html; charset=windows-1252 X-Spam-Checker-Version: SpamAssassin 2.63 (2004-01-11) on post.thorcom.com X-Spam-Level: ** X-Spam-Status: No, hits=2.4 required=5.0 tests=FROM_ENDS_IN_NUMS,HTML_20_30, HTML_MESSAGE,MAILTO_TO_SPAM_ADDR autolearn=no version=2.63 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-Spam-Filtered: by PlusNet SpamCORE (v3.00) Content-transfer-encoding: 8bit

Thanks for the thoughts James!

I seem to have the system under control into a 50ohm dumby load. The wave form across the drian/grd is a ringing square wave that seems to settle out about twice the supply voltage and peaks at ~140V with a 24V supply. The problems start when I place a 5 element C in/out LPF in the line(same as G0MRF design...) The square wave is gone and a number of ringing peaks take over. The FETs get warm fast! Any thoughts?

I'm using 'Street Ferrite' for the matching xfmr out of an old VGA monitor...

Scott

----- Original Message -----

From: James Moritz

To: rsgb_lf_group@blacksheep.org

Sent: Sunday, July 18, 2004 5:08 AM

Subject: LF: Re: clicking matching transformer...

Dear Scott, LF Group,

It is fairly normal for ferrite-cored transformers to make funny clicks and squeals due to magnetostriction of the ferrite. This is the magnetic version of the piezo-electric effect; as the flux density changes, the dimensions of the ferrite change, leading to vibration of the core. It doesn't seem to do much harm.

The voltage experienced by the FETs depends on the circuit configuration - assuming we are talking about an "MRF" type of push-pull circuit, in an ideal circuit with a purely resistive load, the peak drain voltage will be twice the supply voltage. It will be slightly higher if the circuit is a "current fed" configuration with a capacitive-input low pass filter. In practice, the presence of HF ringing potentially leads to much higher voltages; the best way to tell is to connect an oscilloscope to the drain, which may reveal frightening results! As was discussed on the reflector some weeks back, this can be minimised by reducing the output transformer leakage inductance to the minimum, keeping the primary inductance low, and connecting series RC damping from each drain to ground. If you are using a low-pass filter on the output with a shunt input capacitor, you will also need a suitable un-bypassed choke feeding the DC into the centre tap of the transformer primary. However, these measures will reduce but not entirely eliminate the ringing, and the peak drain voltage in my amplifier is about 3 - 3.5 times the supply voltage. So 48V supply should be OK with 200V devices, provided suitable attention is paid to the ringing. Certainly, there are several TXs around operating at similar levels. If there is excessive ringing, the MOSFETS behave as big zener diodes, clipping the waveform and dissipating a lot of heat in the process. The TO220 devices in the pictures on your web site will not be able to dissipate much more power than they are already at 300W output, so this is certainly something to look for.

The class E design seems to give cleaner waveforms, but results in higher peak voltage in the first place - the component values are probably more critical too. Also, as Alan points out, you have to think about what can happen when the antenna goes off-tune and gives a highly reactive load. I reckon the push-pull design has the virtue of simlicity for supply voltages less than about 50V. Another possibility suitable for higher supply voltages is the 'Decca' bridge configuration (or the "half-bridge" as used by G4JNT in his off-line design). This circuit topology is inherently self-limiting, a feature that effectively limits the peak MOSFET voltage to a little over the DC supply voltage.

Cheers, Jim Moritz

73 de M0BMU

----- Original Message -----

From: Scott Tilley

To: rsgb_lf_group@blacksheep.org

Cc: lowfer@lwca.org

Sent: Saturday, July 17, 2004 9:31 PM

Subject: LF: clicking matching transformer...

Any thoughts on if a matching transformer should 'click' when the key is released... Obviously it is likely an inductive reaction to the rapid change in current supplied...