[R-390] Drop-in, GFCI-friendly line filters for R-390A's?

Thu Oct 13 20:53:49 EDT 2005

Several years ago I took a nasty lightning strike which in addition to
totally wiping out the electrical system (it exploded the glass plug
fuses in the fuse box) my R-390 line filter ended up toast (the only
component in the R-390 to suffer).  I ended up temporarily bolting one
of the computer type line filters in place.  The 3-prong socket is
accessible in the round hole for the original and it was necessary to
use a strap to get the second hole of the filter stationary but it does
work.

I have since opened the original filter box (adventures with a dermal
tool 101) and found the capacitors were just paper dust.  The coils
(toroidal which was a surprise) were in good shape.  I have rebuilt the
box with HV disc ceramics but I have yet to solder the box back together
(hey I like he original looks).  

While I never plugged either into a GFI circuit, there are alternatives.

Good luck and 73

Al - WB8FEQ 

-----Original Message-----
From: r-390-bounces at mailman.qth.net
[mailto:r-390-bounces at mailman.qth.net] On Behalf Of Roy Morgan
Sent: Thursday, October 13, 2005 5:07 PM
To: Tim Shoppa; r-390 at mailman.qth.net
Subject: Re: [R-390] Drop-in, GFCI-friendly line filters for R-390A's?

At 03:47 PM 10/13/2005, Tim  Shoppa wrote:
>Others here have talked about replacing the existing high-leakage-
>current R-390A line filter with a GFCI-friendly IEC-plug style
>power entry module.

Tim,

The line filters normally do not "leak".  The capacitors in side there
pass 
ac current because of the basic nature of capacitors. There happen to be
a 
lot of capacitance (and more in the R-390/URR "non-A" line filter) and
the 
capacitance is larger than other more modern line bypass filters.

>If I wanted to be less ambitious and simply bolt in a regular
>line filter unit, are there any that are a "perfect" fit in terms
>of mounting holes and electrical connections?

I don't think so.  The holes in the rear panel for the two small studs 
exiting the R-390A/URR line filter are small and separate.

The R-390/URR has a single hole in the rear panel to allow the entire
line 
cord connector, which is part of the line filter, to go through the
panel. 
This thing is large enough so that an IEC type line cord connector
*might* 
well mount in that hole with only the flange mounting screw holes being
needed.

>Non-medical units seem to have leakages
>of under 1mA and the medical units seem to have
>leakage currents of just a few microamps.

Medial applications require far less leakage. There is a specified
maximum 
leakage for normal appliances and household electrical equipment (I
don't 
know what it is.)  There also has in the past been a recommended test 
circuit that allows measurement of such leakage - if I remember
correctly, 
it consists of a capacitor and a resistor. You measure the voltage
across 
the resistor when it is placed in the leakage path.  (That would be from

the ungrounded chassis of our R-390 radios to the safety ground.)  This 
little circuit may no longer be recommended or allowed under standard 
testing procedures.

>   The number to be under in terms of GFCI-friendly is 5mA, is that the
>right ballpark?

I can't say. You may find more specifications associated with GFI
devices 
than I did: research GFI outlets and circuit breakers to see if this
number 
is specified in catalog or engineering info available on the web.

Here are some references:

Simpson makes an appliance leakage tester called the 229-2. See:
http://www.simpsonelectric.com/pdf/test/229.pdf

That page mentions the applicable standard:
ANSI C101.1-1986 "Leakage Current for Appliances"

Unfortunately, ANSI has historically gained much or some of it's income 
from the sale of standards in paper form that we use to build and run
our 
society. (What's wrong with this picture?)  If you search for
"C101.1-1986" 
at www.ansi.org you traverse through a very frustrating series of dead
ends 
and never do find the standard.

In the UK these test devices are known as a  "PATs" - Portable Appliance

Testers.  The tester is portable, not necessarily the appliance, I 
think.  This page indicates that 0.1 ma is the test for "sensitive
computers".
http://www.instrotech.com/manual_PATS.pdf

One manufacturer of such devices, the
Jinjiang Zhentai Science & Technology Co., Ltd.

lists the specs I partially quote below at:
<http://zhentai.en.alibaba.com/offerdetail/51139649/Sell_Appliance_Leaka
ge_Circuit_Interrupter.html>

" 4) Leakage protection current: when leakage current is 4mA, does not 
trip; when leakage current>=6mA, do trip
5) Tripping delay: when leakage current is 6mA=<5.6s, 204mA=<36.1ms, 
264mA=<25ms, 550mA=<1.68ms"

This tells us that the thing trips at about 5 milliampere and above.

Note that in the R-390A/URR, the line filter has in each side of the
line a 
PI-section filter with two capacitors from the ends of the inductor to
the 
chassis:

C-104 and C-105 on either side of the inductor  L101 and
C-105 and C-106 on either side of the inductor  L102.

Each of these capacitors has the value listed as "Part of FL101, listed
for 
reference only."  One email some time ago indicates that they are 0.068
uF 
paper caps.  Thus, the capacitance from the hot line to chassis is 0.136

uF.  Let's refer to an capacitance calculator at:
http://www.opamplabs.com/rfc.htm
and get a "resistance" of about 19.5 kilohms.
Then the calculator at:
http://www.opamplabs.com/eirp.htm
gives some six miliamperes of ac current.

This is above the four milliamperes of the ground fault interrupter
above.

Roy

- Roy Morgan, K1LKY since 1959 - Keep 'em Glowing!
7130 Panorama Drive, Derwood MD 20855
Home: 301-330-8828 Cell 301-928-7794
Work: Voice: 301-975-3254,  Fax: 301-948-6213
roy.morgan at nist.gov --

_____________________________________________________________
R-390 mailing list
Home: http://mailman.qth.net/mailman/listinfo/r-390
Help: http://mailman.qth.net/faq.htm
Post: mailto:R-390 at mailman.qth.net
Unsubscribe: http://mailman.qth.net/mailman/options/r-390