From owner-qrp-l@Lehigh.EDU Wed Nov 20 07:56:38 1996 Received: from fidoii.CC.Lehigh.EDU (fidoii.CC.Lehigh.EDU [128.180.1.4]) by oucsace.cs.ohiou.edu (8.7.6/8.7.1) with ESMTP id HAA01023 for ; Wed, 20 Nov 1996 07:56:37 -0500 (EST) X-Received-x: from fidoii.CC.Lehigh.EDU (fidoii.CC.Lehigh.EDU [128.180.1.4]) by oucsace.cs.ohiou.edu (8.7.6/8.7.1) with ESMTP id HAA01023 for ; Wed, 20 Nov 1996 07:56:37 -0500 (EST) Received: from Lehigh.EDU ([127.0.0.1]) by fidoii.cc.lehigh.edu with SMTP id <35240-18305>; Wed, 20 Nov 1996 07:55:44 -0500 Received: from nss2.CC.Lehigh.EDU ([128.180.1.26]) by fidoii.cc.lehigh.edu with ESMTP id <34939-18305>; Wed, 20 Nov 1996 07:54:24 -0500 Received: from utkux4.utcc.utk.edu (UTKUX4.UTCC.UTK.EDU [128.169.76.11]) by nss2.CC.Lehigh.EDU (8.8.3/8.8.3) with SMTP id HAA216811 for ; Wed, 20 Nov 1996 07:54:13 -0500 Received: from localhost by utkux4.utcc.utk.edu with SMTP (SMI-8.6/2.7c-UTK) id MAA22024; Wed, 20 Nov 1996 12:53:57 GMT Message-Id: Date: Wed, 20 Nov 1996 07:53:56 -0500 (EST) Reply-To: cebik@utkux.utcc.utk.edu Sender: owner-qrp-l@Lehigh.EDU Precedence: bulk From: "L. B. Cebik" To: "Low Power Amateur Radio Discussion" Subject: Re: ARRL info on Computer generated RFI/EMI: cores In-Reply-To: <9610208484.AA848480282@smtplink.dscc.dk> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-To: LSCHAUMB X-Cc: Low Power Amateur Radio Discussion X-Sender: cebik@utkux4.utcc.utk.edu X-Listprocessor-Version: 8.0 -- ListProcessor(tm) by CREN Status: RO I read the file on computer RFI, updated to January, 1994, by our own Daniel Wee. It has some excellent information. However, some doubts were expressed about the effectiveness of split ferrite cores in suppressing RFI due to the need for more inductive reactance along the outside of lines than single cores might provide. First, let me note that for QRP, the best way to get rid of computer RFI is working from inside the computer (if feasible) to the outside. Keeping the genie inside the bottle is always the best way to prevent him from causing unexpected problems. However, I have used split cores extensively and with good success at all amateur power levels to prevent RF interactions among the station gear. Rather than the RS cores with plastic shells, I have been using some heavier cores from Amidon. They sell (or used to sell: I bought big cards of them several years back) split cores with 1/4" and 1/2" holes. 1. A Must: always be sure you can press the halfs tightly together for maximum effectiveness. A gap reduces effectiveness. Tape tightly shut. 2. Size the core to your needs: a. 1/2" fits many printer cables, RG-8, RG-213, etc. b. 1/4" fits RG-58/59, serial cables, etc. c. For thin cables (RG-174, telephone, etc.), wrap as many turns of the cable as will fit the hole (usually 1/2") around one side of the core and close with the other half. This increases the inductive reactance to sizeable levels. 3. Mounting: the cores will slide, so I usually slit a grommet, press it around the cable at the limit point, and add a drop of glue to the slit to reseal it. Unlike tape wrappings, this leaves no residue on the cable. Be sure to leave room for the connector shell, if relevant. 4. Where: I place cores at the cable ends of the potential source of RFI. For example, I use 2 1/2" cores at the rig-end of the cable and another pair at the power amplifier end of the same cable. The output cable goes to a bulkhead of feed-through connectors, so pairs go at either end of that cable. When I have an ATU in the line, pairs go at each end of that cable. The object is to a. confine errant RF to its own box so far as possible, and b. to keep any RF picked up by the cable shield out of the boxes. Obviously, this hypothetical case used RG-213. If RG-58 was in use, then 2 or three of the smaller cores would be used. Do not forget to give a wrap-around treatment to non-RF cables, for example, to the 9-volt line that light my ATU meter lamp. 5. Another set of wheres: a. AC cords get a wrap-around treatment at the case-entry point. b. DC cords from wall transformer/power supply units or from independent DC supplies get a wrap-around treatment at both ends of the line. c. audio cables get a wrap-around treatment, especially when entering a box with RF in it. d. All computer cables get paired cores at each end if fat or a wrap-around treatment if thin. 6. When: I do not wait until I encounter RFI: instead, I plan for coring all relevant cables during installation. That way, I plan for enough cable and route it with coring in mind. 7. How many cores: I lost count long ago. 8. How much RFI is left? Actually, I have no inter-unit RFI that I can detect. Since I placed the cores during installation, I cannot even estimate how much or little I suppressed. I simply did not want to be bothered by the problem. For the last several years, through several rig and computer changes, I have not been bothered. 9. Limits: if your computer or rig is radiating from the case, your antenna can pick up the signal--cores do nothing to suppress this source. I wish I could place a giant core around one of the pole pigs in my area, but the connecting wires would likely radiate and nullify the measure. Cores can be effective in breaking up cable-case paths, and they do not have to be heavily lumped to do the job if you break up the paths often. They are best placed on the cable ends of RFI and bothersome RF sources. They are unlikely to do much good placed on the lines to and from QRP rigs--although if you use a separate DC supply, you may want to do a wrap-around treatment of the DC cable and the AC cord, especially on some of those bargain DC supplies. Even batteries can benefit on occasion from a bypass capacitor across the terminals and a wrap-around choke. Yes, a whole bunch of cores is a pretty good investment--maybe a whole kit's worth of dollars if you have lots of computer and rig cables. But they never wear out, are reusable forever, and allow you to use your antenna cable as a backjack on unsuspecting burglars (or on yourself during installation--I know from lumpy experience). Nope, cores are not a cure-all, but I find them to be good insurance against the problems they can alleviate. -73- LB, W4RNL