Return-Path: Received: from post.thorcom.com (post.thorcom.com [195.171.43.25]) by klubnl.pl (8.14.4/8.14.4/Debian-8+deb8u2) with ESMTP id w39IDgcn022764 for ; Mon, 9 Apr 2018 20:13:43 +0200 Received: from majordom by post.thorcom.com with local (Exim 4.14) id 1f5bCv-0003kJ-4K for rs_out_1@blacksheep.org; Mon, 09 Apr 2018 19:07:57 +0100 Received: from [195.171.43.32] (helo=relay1.thorcom.net) by post.thorcom.com with esmtp (Exim 4.14) id 1f5bCs-0003k6-Sd for rsgb_lf_group@blacksheep.org; Mon, 09 Apr 2018 19:07:54 +0100 Received: from mail-wm0-f47.google.com ([74.125.82.47]) by relay1.thorcom.net with esmtps (TLSv1.2:ECDHE-RSA-AES256-GCM-SHA384:256) (Exim 4.89) (envelope-from ) id 1f5bCm-0008Dd-E8 for rsgb_lf_group@blacksheep.org; Mon, 09 Apr 2018 19:07:53 +0100 Received: by mail-wm0-f47.google.com with SMTP id x4so18417618wmh.5 for ; Mon, 09 Apr 2018 11:07:48 -0700 (PDT) X-DKIM-Result: Domain=gmail.com Result=Good and Known Domain DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:in-reply-to:references:from:date:message-id:subject:to; bh=MsRYo8v0KhpdFsD6aeCEpeE50icw3mij+ddcsho8+J8=; b=FkOPoN63TUOKJ8zrOfYtIrqXjHdhm9CWPl7r7zJ2iSNUjuJeV5I/Y9nlwd5EP8geP3 dgVmWlTKCJjsCfuUdzuIwfxXs6MzcYzuhwHYFvY6yh+5wSHSwcXEgKPoznMPQ3ximxVz 1M2858roVCqq43q7mMM9bMj7As6WS635uorByAdxhxAfT80RW6fDsLjpj8XHtv40/8Nk Y+OckN0Ta2CVnG+yNYybxawFucXqvwHADCWxQG4v2jba2zUI5bd2g8fNuBx5uWbTasg7 0i3UkXWZczeO6+Q0TqPdtyFmrTmet68unQRW81MfH2H1ZzpggAQ6Ln98ijbGh6tSMhx9 kpuA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to; bh=MsRYo8v0KhpdFsD6aeCEpeE50icw3mij+ddcsho8+J8=; b=eiH4Nt8z2ty0nWX6NPVP9NKbDM0IA/XyRwkc6NGXHXkTXU4+59tOnhs1xejJQ/d/9d M4MVA/prKS4NIWkfeD4Ynwqq9bPufDYg0Mcz+0tRXKICl7ZXhdj8afhleY/QtkaLvHwk Kii680uhNueImo8K/stUyXGGCK/FnFKJuCuV3LRGCbmOFHB101nvUBVgvCZK91u711I0 9J5a7mX1HGfya/9du1JlqYC6ME09jw9BNpXZcrVobSfySlJtlhAxzCdyrdz3Q5aK6jxh bm4GfNTyvYW/1Bb5gf5usaGRAIU2CIfhK27bT8oX1BrHDorLxv9pSaRPj5d3wGXOO8T8 mdPw== X-Gm-Message-State: ALQs6tD34Hr1ZJd4jrnJ0qgY17t7uxazzKfSg6YJyRZHA6K4f22sUXbP rdnOezrlscBCb5Vwj8gsztQ6LZXFIGOgQfnOlm3sIdUt X-Google-Smtp-Source: AIpwx4+efHfQ0qMvP8NYpz8F9Csp/4tZVjDGGXX9qQTLCzWaXmYHkqtw2rMsX3O+zZweVIORgqSvzEykNgYZuFC6+FU= X-Received: by 10.80.244.1 with SMTP id r1mr12481915edm.111.1523297207212; Mon, 09 Apr 2018 11:06:47 -0700 (PDT) MIME-Version: 1.0 Received: by 10.80.154.195 with HTTP; Mon, 9 Apr 2018 11:06:46 -0700 (PDT) In-Reply-To: <1539CCA16B9A4092972CB000BE4C440A@DELL4> References: <162a4797468-1db9-bdf2@webjas-vae168.srv.aolmail.net> <1UQROqZht1.1c1VdzC0S69@optiplex980-pc> <3a0c0669-018b-13a7-b56b-dd3f63fab464@n1bug.com> <197556217.20180409124931@gmail.com> <37574432-b72e-f2d1-0e85-af19dd970817@n1bug.com> <4D19B3707137430CBE04857FBA5C80B1@gnat> <1b882ae9-fede-3ec4-256e-9b068d3acf99@n1bug.com> <1539CCA16B9A4092972CB000BE4C440A@DELL4> From: Andy Talbot Date: Mon, 9 Apr 2018 19:06:46 +0100 Message-ID: To: LineOne X-Spam-Score: 0.8 (/) X-Spam-Report: Spam detection software, running on the system "relay1.thorcom.net", has NOT identified this incoming email as spam. The original message has been attached to this so you can view it or label similar future email. If you have any questions, see the administrator of that system for details. Content preview: I'm a bit late coming to this discussion, but that ringing looks exactly like that you get when there is too much leakage inductance in the transformer. It is common to see such waveforms in SMPSUs and they have to be snubbed with a, well, a snubber network, Usually a C / R / Diode network to catch and dump the spikes. As Jay says, the windings do look a bit 'loose' and they'll certainly be giving some leakage inductance. [...] Content analysis details: (0.8 points, 5.0 required) pts rule name description ---- ---------------------- -------------------------------------------------- 0.0 FREEMAIL_FROM Sender email is commonly abused enduser mail provider (andy.g4jnt[at]gmail.com) 0.0 HTML_MESSAGE BODY: HTML included in message 0.1 URIBL_SBL_A Contains URL's A record listed in the Spamhaus SBL blocklist [URIs: n1bug.com] 0.6 URIBL_SBL Contains an URL's NS IP listed in the Spamhaus SBL blocklist [URIs: n1bug.com] 0.0 T_DKIM_INVALID DKIM-Signature header exists but is not valid X-Scan-Signature: 160666a84ef5d9af0d8f07b52ab8f793 Subject: Re: LF: W1VD amp help - more waveforms Content-Type: multipart/alternative; boundary="089e0825321cec6b0205696e4493" X-Spam-Checker-Version: SpamAssassin 2.63 (2004-01-11) on post.thorcom.com X-Spam-Level: X-Spam-Status: No, hits=0.0 required=5.0 tests=HTML_MESSAGE 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 --089e0825321cec6b0205696e4493 Content-Type: text/plain; charset="UTF-8" I'm a bit late coming to this discussion, but that ringing looks exactly like that you get when there is too much leakage inductance in the transformer. It is common to see such waveforms in SMPSUs and they have to be snubbed with a, well, a snubber network, Usually a C / R / Diode network to catch and dump the spikes. As Jay says, the windings do look a bit 'loose' and they'll certainly be giving some leakage inductance. I've never used toroids in high power LF PAs but have always used more traditional shaped cores like ETD types and pot cores - not least because they are cheaper than torroids, but more importantly, they enclose the winding in ferrite much more , so reducing leakage. But there again, I wouldn't do an LF transmitter by switching into a transformer in the first place. Far better to start off with resonant circuitry such as a tank directly onteh active devices, THEN use a transformer when the waveform is sinusoidal, or nearly so. That way, leakage inductance doesn't give spikes; it just contributes a bit of spurious L that you can absorb into the reactive components and tune out. Hence, I'd only ever go for the series tank or class-E solution. http://www.g4jnt.com/PowerAmps.htm Andy G4JNT On 9 April 2018 at 18:41, wrote: > Paul > > Thanks for the photos and information. I was more interested in the > superimposed gate waveforms than the drains, though. A pix of the LPF as > well would be appreciated ... think there was one last week. Just for the > record ... my doubler was never mounted inside the amplifier ... so you are > covering new ground with that. I would think having it external to the > amplifier would be less problematic. For testing, stick to driving it with > the HP3325A for now ... which is what you're most likely already doing. > > Can't find much fault with the construction ... other than the somewhat > loose primary winding on the transformer. In mine, the primary winding was > nestled inbetween the turns of the secondary. This is important to minimize > leakage reactance ... which also contributes to excessive ringing. > > Think I would continue to concentrate on getting the ringing down before > going to high power. > > One other thing ... have you double checked that you have the correct > capacitance in the LPF? > > Jay W1VD > > > > > > > ----- Original Message ----- From: "N1BUG" > To: > Sent: Monday, April 09, 2018 11:34 AM > > Subject: Re: LF: W1VD amp help - more waveforms > > > Yep, voltage breakdown of the FETs was my worry. However I may be >> making a false assumption. I assumed the voltage spikes in the >> ringing would vary linearly with applied drain voltage? Perhaps that >> isn't necessarily so? >> >> With 13V on the drains the first spike here is 'about' 64V. >> >> Efficiency seems fine. I wouldn't try to put an exact figure to it >> what with measurement accuracy in three places (voltage, current, RF >> power) but it 'appears' to be in the mid 90s. >> >> Paul >> >> >> On 04/09/2018 11:02 AM, jrusgrove@comcast.net wrote: >> >>> Alan >>> >>> The main concern about the ringing is that the positive extreme stays >>> below the drain source >>> breakdown voltage ... with some 'headroom'. Otherwise ... >>> >>> Jay W1VD >>> >>> >>> ----- Original Message ----- From: "Alan Melia" < >>> alan.melia@btinternet.com> >>> To: >>> Sent: Monday, April 09, 2018 10:21 AM >>> Subject: Re: LF: W1VD amp help - more waveforms >>> >>> >>> Paul >>>> I think you need to step back from this a little. One point, >>>> design/cable routing at 137kHz is >>>> normally not that critical and it is unlikely it is having much effect. >>>> Many us have ''haywired'' >>>> PAs at these frequencies without major troubles....even using >>>> croc-leads!! >>>> >>>> You are unhappy about the drain waveforms.......why?? the ringing is on >>>> the ''off'' device and so >>>> doesnt really matter in terms of efficiency. Consider what is >>>> happening. The FET drive goes off >>>> and the drain current starts to drop inductance in the drain feed >>>> causes the voltage to rise to >>>> try to keep the current flowing. Then on top of this the other >>>> transistor switches on and the >>>> transformer connection for the off side sees a rising induced voltage >>>> out of the transformer. >>>> There is bound to be overshoot but it does not absorb much power. It >>>> may be the generous ''dead >>>> zone'' (whilst a good idea) that is producing this picture. >>>> >>>> The critical part is the the flat zero volt line when the FET is on. >>>> The current pulse rings a >>>> bit, but the current drops in the second half of the pulse.....why?? >>>> This, I think, is really the >>>> loss of power. Does Jay's show this? The point I am making is don't get >>>> drawn up blind alleys >>>> because a trace just does not look ''pretty''. >>>> >>>> I think we have mentioned this before but the transient performance of >>>> the power supply may be a >>>> factor. There should not be great changes in the current draw because >>>> you are switching the PSU >>>> load from one transistor to the other but there can be higher speed >>>> transients that can affect >>>> power supply regulators in unexpected ways. >>>> >>>> As Andy said efficiency at 12v is not very meaningful because of the >>>> high rds of these FETs. It is >>>> usefull to run up a PA at low volts say for safety when the antenna >>>> needs to be tuned. >>>> >>>> Finally the scope probes are placing an. albeit small, capacitance >>>> across the drains (I am >>>> assuming they are correctly trimmed) The waveform can often be affected >>>> by where the ground is >>>> picked up from. >>>> >>>> Stick with it because you are close to being ''there'' >>>> >>>> Alan >>>> G3NYK >>>> >>> >> > > --089e0825321cec6b0205696e4493 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
I'm a bit late coming to this discussion, but that rin= ging looks exactly like that you get when there is too much leakage inducta= nce in the transformer.=C2=A0 It is common to see such waveforms in SMPSUs = and they have to be snubbed with a, well, a snubber network,=C2=A0 Usually = a C / R / Diode network to catch and dump the spikes.=C2=A0 =C2=A0As Jay sa= ys, the windings do look a bit 'loose' and they'll certainly be= giving some leakage inductance.=C2=A0 =C2=A0

I've n= ever used toroids in high power LF PAs but have always used more traditiona= l shaped cores like ETD types and pot cores - not least because they are ch= eaper than torroids, but more importantly, they enclose the=C2=A0 winding i= n ferrite much more , so reducing leakage.

But the= re again, I wouldn't do an LF transmitter by switching into a transform= er in the first place.=C2=A0 Far better to start off with resonant circuitr= y such as a tank directly onteh active devices, THEN use a transformer when= the waveform is sinusoidal, or nearly so.=C2=A0 =C2=A0That way, leakage in= ductance doesn't give spikes; it just contributes a bit of spurious L t= hat you can absorb into the reactive components and tune out.=C2=A0 =C2=A0H= ence, I'd only ever go for the series tank or class-E solution.=C2=A0= =C2=A0

Andy=C2=A0 G4JNT



On 9 April 2018 at 18:41, <jrusgrove@comcast.net> wrote:
Paul

Thanks for the photos and information. I was more interested in the superim= posed gate waveforms than the drains, though. A pix of the LPF as well woul= d be appreciated ... think there was one last week. Just for the record ...= my doubler was never mounted inside the amplifier ... so you are covering = new ground with that. I would think having it external to the amplifier wou= ld be less problematic. For testing, stick to driving it with the HP3325A f= or now ... which is what you're most likely already doing.

Can't find much fault with the construction ... other than the somewhat= loose primary winding on the transformer. In mine, the primary winding was= nestled inbetween the turns of the secondary. This is important to minimiz= e leakage reactance ... which also contributes to excessive ringing.

Think I would continue to concentrate on getting the ringing down before go= ing to high power.

One other thing ... have you double checked that you have the correct capac= itance in the LPF?

Jay W1VD






----- Original Message ----- From: "N1BUG" <paul@n1bug.com>
To: <r= sgb_lf_group@blacksheep.org>
Sent: Monday, April 09, 2018 11:34 AM

Subject: Re: LF: W1VD amp help - more waveforms


Yep, voltage breakdown of the FETs was my worry. However I may be
making a false assumption. I assumed the voltage spikes in the
ringing would vary linearly with applied drain voltage? Perhaps that
isn't necessarily so?

With 13V on the drains the first spike here is 'about' 64V.

Efficiency seems fine. I wouldn't try to put an exact figure to it
what with measurement accuracy in three places (voltage, current, RF
power) but it 'appears' to be in the mid 90s.

Paul


On 04/09/2018 11:02 AM, jrusgrove@comcast.net wrote:
Alan

The main concern about the ringing is that the positive extreme stays below= the drain source
breakdown voltage ... with some 'headroom'. Otherwise ...

Jay W1VD


----- Original Message ----- From: "Alan Melia" <alan.melia@btinternet.com= >
To: <r= sgb_lf_group@blacksheep.org>
Sent: Monday, April 09, 2018 10:21 AM
Subject: Re: LF: W1VD amp help - more waveforms


Paul
I think you need to step back from this a little. One point, design/cable r= outing at 137kHz is
normally not that critical and it is unlikely it is having much effect. Man= y us have ''haywired''
PAs at these frequencies without major troubles....even using croc-leads!!<= br>
You are unhappy about the drain waveforms.......why?? the ringing is on the= ''off'' device and so
doesnt really matter in terms of efficiency. Consider what is happening. Th= e FET drive goes off
and the drain current starts to drop inductance in the drain feed causes th= e voltage to rise to
try to keep the current flowing. Then on top of this the other transistor s= witches on and the
transformer connection for the off side sees a rising induced voltage out o= f the transformer.
There is bound to be overshoot but it does not absorb much power. It may be= the generous ''dead
zone'' (whilst a good idea) that is producing this picture.

The critical part is the the flat zero volt line when the FET is on. The cu= rrent pulse rings a
bit, but the current drops in the second half of the pulse.....why?? This, = I think, is really the
loss of power. Does Jay's show this? The point I am making is don't= get drawn up blind alleys
because a trace just does not look ''pretty''.

I think we have mentioned this before but the transient performance of the = power supply may be a
factor. There should not be great changes in the current draw because you a= re switching the PSU
load from one transistor to the other but there can be higher speed transie= nts that can affect
power supply regulators in unexpected ways.

As Andy said efficiency at 12v is not very meaningful because of the high r= ds of these FETs. It is
usefull to run up a PA at low volts say for safety when the antenna needs t= o be tuned.

Finally the scope probes are placing an. albeit small, capacitance across t= he drains (I am
assuming they are correctly trimmed) The waveform can often be affected by = where the ground is
picked up from.

Stick with it because you are close to being ''there''

Alan
G3NYK




--089e0825321cec6b0205696e4493--