Message-ID: <00a001c44d38$ed246830$e901a8c0@bob2l2u6k2n1g3>
From: "Vernall" <vernall@xtra.co.nz>
To: rsgb_lf_group@blacksheep.org
References: <6.1.0.6.2.20040608100555.02801340@POP3.freeler.nl>
Date: Tue, 8 Jun 2004 21:13:52 +1200
MIME-Version: 1.0
Subject: LF: Re: Effect of LP-filter om efficiency
Content-Type: text/html; charset=windows-1252
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Content-transfer-encoding: 8bit

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2800.1400" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><FONT face=Arial size=2>Dick PA0SE and others,</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>An interesting contribution.&nbsp; My pushpull 
bipolar QRO PA can run to near squarewave performance, and I have successfully 
used an inductor input low pass filter for many years.&nbsp; The low pass filter 
uses three large potcore inductors and HV polystyrene capacitors.&nbsp; Texts on 
Class D and E do mention resonant circuits, and most of the commercial 
applications are in narrow bands, so tuning is a reasonable way to implement 
Class D/E for those users.&nbsp; At LF I believe&nbsp;a hard driven final can 
simply be followed by a low pass filter and still deliver very high efficiency 
(certainly a lot better than Class C).&nbsp; In New Zealand our band is about to 
be expanded to become 130 - 190 kHz, which is a relatively wide band, and not 
worth having tuning that needs revisiting for significant frequency 
change.&nbsp; A broadband PA with a selection in low pass filtering is my 
preference.&nbsp; For use only in the 136 kHz band, only one low pass filter is 
needed.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>My preference for an inductor input filter came 
from reading (I hope my recollection is right) a Motorola article on stability 
of HF solid state power amplifiers, and the preference was for an inductor input 
filter.&nbsp; I merely scaled the filter values to LF and found they worked 
well.&nbsp; Possibly the instability bogey is not applicable to LF but I was 
aware that inrush current would be different if a capacitor input filter was 
used.&nbsp; However, there are polystyrene capacitors from collector to ground 
so on the PA side of the output transformer there are "shunt capacitors" and 
these certainly reduce the "high frequency nurgles" as seen on a scope 
probe.&nbsp; </FONT></DIV>
<DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>I have not tried "snubber" series RC networks but I 
have seen commerial high power switchmode inverters that use that, and the 
resistors are surface mount with good power rating.&nbsp; Some of that 16% 
"harmonic power" gets absorbed in the interests of controlling spike 
limits.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>Power devices (bipolar of MOSFET) can have slower 
turn-off than turn-on, and I suspect this has a large impact on transients in a 
pushpull power amplifier.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>Regarding suppression of harmonics, and potential 
radiation, the efficiency of amateur (electrically small) 
antennas</FONT>&nbsp;<FONT face=Arial size=2>rises significantly (more than 
linear) with increasing frequency.&nbsp; So merely&nbsp;a loading coil or "net 
antenna Q" is only part of the consideration of radiation of harmonics.&nbsp; 
The insertion loss of a low pass filter can be kept fairly low and I think it is 
best to include it in a transmitter design, especially a hard driven 
"squarewave" machine.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>73, Bob ZL2CA</FONT>&nbsp;</DIV>
<BLOCKQUOTE 
style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
  <DIV style="FONT: 10pt arial">----- Original Message ----- </DIV>
  <DIV 
  style="BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: black"><B>From:</B> 
  <A title=d.w.rollema@freeler.nl href="mailto:d.w.rollema@freeler.nl">Dick 
  Rollema</A> </DIV>
  <DIV style="FONT: 10pt arial"><B>To:</B> <A title=rsgb_lf_group@blacksheep.org 
  href="mailto:rsgb_lf_group@blacksheep.org">LF-Group</A> </DIV>
  <DIV style="FONT: 10pt arial"><B>Sent:</B> Tuesday, June 08, 2004 8:06 
PM</DIV>
  <DIV style="FONT: 10pt arial"><B>Subject:</B> LF: Effect of LP-filter om 
  efficiency </DIV>
  <DIV><FONT face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT 
  face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT face=Arial 
  size=2></FONT><BR></DIV><FONT size=3>To All from PA0SE<BR><BR>The following 
  subject may have been discussed on the reflector before but I can't remember 
  it.<BR><BR>Class D and E final amplifiers have high efficiency because they 
  produce square waves.&nbsp; When the voltage between source and&nbsp; drain of 
  the FETs is high, current is zero; when current flows voltage is almost 
  zero.<BR>But what happens if the transmitter is followed by a low pass filter 
  with a shunt capacitor at the input, as is often the case?<BR>Every cycle the 
  charge in the capacitor must be reversed and with a current of finite value 
  that takes some time. During that time current flows in the conducting 
  transistor whilst voltage over it is not small. So one can expect dissipation 
  in the transistor to increase and efficiency to decrease.&nbsp; 
  <BR><BR>Looking at it in the frequency domain one could say that the shunt 
  capacitor causes the square wave to start becoming a sine wave.<BR><BR>If the 
  current into or out of the capacitor is so high that charge reversal takes 
  negligible time, so the square wave is preserved, then output resistance of 
  the transmitter is obviously so small it almost puts a short circuit over the 
  capacitor. In that case the capacitor does nearly nothing and could just as 
  well be deleted...&nbsp; <BR><BR>The problem does not arise when the LP-filter 
  starts with a series inductor. The sudden cut-off of current would result 
  in&nbsp; a very high kick-back voltage at the drain of a single-ended 
  amplifier, almost certain killing the transistor.<BR>But in a push pull&nbsp; 
  amplifier this is prevented by the other transistor that starts to conduct at 
  the same moment.<BR>Provided of course coupling between the two halves of the 
  primary winding is very tight.<BR><BR>So it seems to me that class D and E 
  amplifiers should preferably be followed by a low pass filter with a series 
  inductor at the transmitter side. <BR>&nbsp;<BR>Please correct me if my 
  conclusion is not valid.<BR><BR>An interesting question is whether a low pass 
  filter is really necessary when the transmitter feeds the aerial via a series 
  tuning coil.<BR>Harry Grimbergen, PA0LQ, has given this some thought and he 
  says the following (my translation).<BR>Quote: <BR>Not unusual is an aerial 
  capacitance of about 300 pF and a total loss resistance in coil, aerial and 
  earth of some 40 ohms&nbsp; At 136 kHz this results in a Q of about 100 for 
  the aerial system. At the third harmonic&nbsp; this produces an extra 
  attenuation of about 280 times, or 49 dB.&nbsp; The third harmonic of a square 
  wave is 9.5 dB weaker than the basic frequency. On the other hand radiation 
  resistance increases with frequency squared; 9 dB for d3. <BR>Nevertheless 
  harmonics will be suppressed almost 50 dB. <BR><BR>About 16% of the power in a 
  square wave is in the harmonics and this power is reflected back into the 
  final amplifier.<BR>But it is not converted into heat. The DC input is 
  decreased by this 16% as compared to the situation where the amplifier is 
  loaded by a resistive load of 40 ohms for the above case.&nbsp; I have been 
  able to show this also using simulation by MICROCAP. <BR>Unquote.<BR><BR>A 
  problem in&nbsp; The Netherlands would be that the radio inspector does not 
  measure harmonics as field strength but as power in the output of the 
  transmitter (or LP-filter, when present). So selectivity of the aerial system 
  does not help.<BR><BR>Any comments will be very welcome.<BR><BR>73, Dick, 
  PA0SE<BR><BR>&nbsp;<BR></BLOCKQUOTE></FONT></BODY></HTML>