Elecraft › [K3]

Clamp-on RF current meters - a question for Ian and others..

Classic

List

Threaded

5 messages Options

VK7JB

Clamp-on RF current meters - a question for Ian and others..

Hi All,

I've been reading the discussions about RF and common mode current and using a clamp-on RF current meter to sniff out common mode current in the shack. I've checked the calibration of my MFJ 854 and am surprised that it's calibrated quite well: I measure 1A of current measured running 50W into a 50ohm dummy load.

So, here's my question: how much common mode current is considered "acceptable"? This meter measures down to a few milliamps. Is it an impossible/futile goal to try to eliminate any stray CMC completely? When should you call it quits and accept what you find?

Just curious, because the only CMC I can measure in my shack is about 10mA measured at the coax connector on the back of one of my rigs while tuning up with 10W output.

Interesting discussion, because I'd never thought of quantifying CMC before.

73,
John
VK7JB

gm3sek

Re: Clamp-on RF current meters - a question for Ian and others..

VK7JB wrote:
>
>I've been reading the discussions about RF and common mode current and
>using a clamp-on RF current meter to sniff out common mode current in
>the shack. I've checked the calibration of my MFJ 854 and am surprised
>that it's calibrated quite well: I measure 1A of current measured
>running 50W into a 50ohm dummy load.
>

The MFJ-854 is quite a competent piece of equipment - see my review for
Radcom, at:
<http://www.ifwtech.co.uk/g3sek/clamp-on/mfj-reviews.pdf>

(In contrast the same review warns against the MFJ-805, a definite
"Don't Buy".)

There is more info about HB clamp-on meters at:
<http://www.ifwtech.co.uk/g3sek/clamp-on/clamp-on.htm>

>So, here's my question: how much common mode current is considered
>"acceptable"? This meter measures down to a few milliamps. Is it an
>impossible/futile goal to try to eliminate any stray CMC completely?
>
There is no general answer, because RFI depends not only on the level of
CM current but also - very much - on the type of "victim" equipment. As
we well know, there is a huge spread in sensitivity to RFI between
different types of equipment, and even between different models of the
same general type.

Also, RFI very often shows a 'threshold level' above which you have a
problem, but below which the problem is either disappears completely or
is of no concern to the user.

All of these things combine to prevent any general answers about
"acceptable" levels of CM current.

>When should you call it quits and accept what you find?

If you don't have an RFI problem, you can usually call it quits and stop
worrying :-) However, some of us like to understand what's really
going on, so I'll at least try to answer your next question...

>Just curious, because the only CMC I can measure in my shack is about
>10mA measured at the coax connector on the back of one of my rigs while
>tuning up with 10W output.
>
At 10W you'd be driving about 450mA of normal differential-mode RF
current into a 50-ohm antenna feedpoint. Meanwhile, down in the shack,
you're measuring 10mA of CM current on the outside of the feedline.

Most victim equipment can tolerate 10mA, and you aren't reporting any
RFI problems... but I'm not at all sure that everything is OK. It only
means that 10W isn't enough power to cause a problem! With 1kW output
the CM current would be 100mA, and at that level many kinds of victim
equipment might start to fall over.

So yes, I think this is worth investigating further. The first test
should be to swap to a dummy load, and also to swap transmitters, to see
if there's a shielding problem on that particular rig. If all the CM
current disappears (as it may appear to do, with a simple diode
detector) then the problem probably lies closer towards the antenna and
its feedline.

The next place to measure the CM current would probably be where the
coax feedline enters the shack. Measure what happens if you insert a
good ferrite common-mode choke at this point (eg several turns on a
giant #31 snap-on bead; see K9YC's papers and web presentations). Also
check for CM current on all the *other* feedlines - not forgetting the
rotator cable - because the problem isn't always where you first
imagined.

>Interesting discussion, because I'd never thought of quantifying CMC
>before.
>
The clamp-on CM current meter is a really valuable tool for finding and
fixing RFI problems - sometimes before they even happen.

And please don't imagine that QRP operators are automatically immune!
Even if you don't cause any RFI problems, CM currents can also affect
the noise level in your receiver so a 'common-mode cleanup' is still
worthwhile.

The clamp-on current meter doesn't have to be complicated - just a split
ferrite bead with a 10-turn secondary windings, a simple Schottky diode
detector and a 100uA meter. My website shows an example in "ugly
construction" that took about 30 minutes to build.

Who knows, one day some enterprising company may offer it as a
'bare-board' kit?

(We'll be away for the weekend, so I look forward to reading the list
mail on Monday.)

--

73 from Ian GM3SEK
http://www.ifwtech.co.uk/g3sek
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

Jim Brown-10

Re: Clamp-on RF current meters - a question for Ian and others..

On 11/4/2011 1:31 AM, Ian White GM3SEK wrote:
> Most victim equipment can tolerate 10mA, and you aren't reporting any
> RFI problems..

I generally agree with everything in Ian's excellent post, but I will
disagree with this statement. Around 2003, I did extensive testing of
pro audio gear for RFI susceptibility to common mode current (the the
Pin One Problem), published as an AES Paper that can be downloaded on my
website. I detected quite a bit of RFI with levels of RF current in the
range of 10mA, over a range of frequencies between 100kHz and 500 MHz.

There's another important answer to your question that Ian did not
address -- receive noise. All passive linear circuits follow
reciprocity -- that is, they work in both directions. One of the major
reasons for using a common mode choke is to prevent RF noise picked on
the feedline from coupling to the antenna, and from there to our
receiver. In this application, the choke generally needs to be at the
antenna feedpoint (that is, where the feedline connects to the antenna)
to be effective.

Several researchers, working independently, have established a choking
impedance on the order of 5,000 ohms as a good design goal. W1HIS and I
did so roughly 6-8 years ago, and a US military research group did so
around 1970 in an internal technical applications note someone from the
CIA engineering group forwarded me a year or so ago. W1HIS was the
first to identify this issue in the ham community, and common mode
chokes as a way to solve it. As soon as I saw his paper (around 2006) I
added a choke to one of my antennas that was hearing noise, and the
noise dropped a couple of S-units.

The choice of 5,000 ohms is only a general design goal. Lower impedances
will provide less reduction of common mode current, but may be
sufficient for many situations, and higher impedances may be required
for others. Successful common mode chokes sold for many years in the US
for RFI to telephone equipment, for example, provide choking impedances
in the range of 20K ohms, and lower values of choking impedance are
insufficient.

My writing is at http://audiosystemsgroup.com/publish.htm

73, Jim Brown K9YC
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

VK7JB

Re: Clamp-on RF current meters - a question for Ian and others..

In reply to this post by VK7JB

Hello Ian, Jim & group,

Thanks to everyone for your advice and thoughts on this matter. It's led to a very interesting weekend on a common mode clean up.

I'd always thought I'd had a "clean shack" from a stray RF/common mode current perspective. I'd had a few "minor" issues with RF getting into a hi-fi speaker, but no other RF problems I'd ever noticed. But after the recent discussions, I got out my MFJ 854 clamp on meter and spent the weekend sniffing out CMC in my setup. The results were revealing.

One rig had about 15-20mA of common mode current on the coax at the connection with the rig, but hunting around, I found about 40mA on the AC cable feeding the power supply connected to that rig. There was no CMC on the coax leading from the ATU attached to that rig to the DX engineering 1:1 balun what I use to transition to open wire line to feed my horizontal loop. 10 turns of the power supply's AC cable through a 240-43 toroid removed all detectable CMC from that set up.

The second rig set up ( switchable to the same loop antenna but operating from a different power supply) was more of a puzzle. Here, all the coax cables and inter-connections had no detectable CMC but I found about 30mA of current on the AC cabling supplying my Ameritron AL 811 amplifier. This has an itegrated linear power supply, which I plug into a power distribution board. This power board also has the powersupply running my 2nd rig plugged in to it. With the amp turned off, but plugged in to the board, there was 35mA of CMC on its AC line and about 15mA on the rest of the AC cabling supplying that rig. If I unplugged the Amp from the distribution board, the CMC fell to undetectable. The fix was another large toroid wound around the AC cord of the amp and I put one on the distribution board too. It seemed that the choke on the amplifier AC cord was what did the trick.

None of my audio or DC power supply cables had any CMC detectable. But clearly, CMC was present on the AC power cabling supplying the power supplies. As predicted, simplw clip on ferrites were useless. To reduce the measured CMC to undetectable, I needed 8-10 turns of the cabling through a large diameter 43 mix toroid.

I'm still not sure why the amplifier cable was so important in carrying CMC into my set up, but it was.

This has been a very interesting exercise. Thanks to all for your comments on and off the list.

73,
John
VK7JB

VK7JB

Re: Clamp-on RF current meters - a question for Ian and others..

In reply to this post by VK7JB