From: Chris Wilson <dead.fets@gmail.com>
Date: Fri, 28 Apr 2017 14:52:17 +0100
Message-ID: <23033458.20170428145217@gmail.com>
To: Markus Vester <rsgb_lf_group@blacksheep.org>
In-Reply-To: <15bb4c39bb3-2379-b39e@webprd-m107.mail.aol.com>
References: <1907129228.20170428123626@gmail.com> <15bb4c39bb3-2379-b39e@webprd-m107.mail.aol.com>
MIME-Version: 1.0
Subject: Re: LF: Varistor spike management
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Status: RO

Hello Markus,

Friday, April 28, 2017, 2:32:50 PM

Another mail from me should appear in a minute, with a typo saying I
have 1500pF and 10 ohms low inductance resistor "Drain to Gate". It is
of course Drain to Source.... ;)

The circuit is Jay's, W1VD and built even duplicating, near as damn
it, his mechanical layout. As such the supply to each amp has a 4700uF
electrolytic to ground, right by the choke. I then added a 100V 0.1uF
ceramic as well, across the electrolytic. Supply is an HP 52V @ 57 or
something Amps SMPS from a computer server. Like this one, but un
modified:

http://www.qsl.net/kf8od/50vdcpsu.html


The leads from SMPS to amp(s) is figure of eight speaker wire, with
about 4mm diameter pure copper stranded conductors. The lead is about
3 feet long. I tried winding as many turns around a .4 inch OD 43
material toroid but it didn't seem to make any difference to the spike
energy. You are quite correct the energy takes quite a while to
dissipate...

Amp circuit and info here:

http://www.w1vd.com/137-500-KWTX.html

Thanks Markus!


> Hi Chris,

> the "overshoot" screenshot seems to indicate that ihe peak is not
> only a short RF spike but caused by a longer positive excursion of
> the supply voltage. This is probably caused by the suddenly ending
> current consumption when the carrier goes off. If the supply line
> contains significant inductance, it will try to sustain the current
> and thus increase the voltage. In a switchmode supply this could
> easily happen in the output ripple filter. A larger electrolytic cap
> (e.g. 100 uF) across the DC line near the PA should fix it.   

> Best 73,
> Markus (DF6NM)


-- 
Best regards,
 Chris                            mailto:dead_fets@gmail.com