Posts Tagged ‘RFI’
Front USB ports on your PC.
November 18th, 2024 | 
Author: 
The finished product.
Once a year I take some time to open up my PC and give it a nice cleaning out. I am shocked at how much dust collects just after one year. I built the system and spent some money on it so I want to do my best to look after the upkeep. My computer is tucked off to the side of my desk on a shelf built for it. Taking it out is a chore and I have to make sure each USB cable goes back exactly where it was removed. If not when I start my virtual com port program I get an error message that certain com ports are not working.
I have been doing some reading on RFI these past few months and along with that comes the subject of grounding. One topic that came up over and over again was the front USB ports that some PCs have. Mine has such ports and it has been mentioned that most of the time these ports are not grounded to the chassis of the PC. They in most cases are grounded to the motherboard via the control cable that makes its way out to the USB front board.
Sure enough, my PC was one that did not have a chassis ground for the front USB ports. It was easy to do and was a jumper wire from the USB board to the PC chassis. At the time I had no green wire and had to use red but I will know what it is for. While I had the PC open I did check the onboard USB ports and I have 11 of them and they are chassis grounded.
What does RFI have to do with Pins…….
I have been doing some RFI investigating, which has sparked my interest because I have been having some RFI issues of my own. In my travels, I have read of something called Pin 1 problem. I went to a few websites about this and had a glance here and there but never looked at it closely. This past week I did a bit more of a deep dive regarding the Pin 1 problem. The name came from Neil Muncy in 1994 he wrote a paper entitled Noise Susceptibility in Analog and Digital Signal Processing Systems. In it, he describes how he found the common XLR connectors used in the audio industry pin 1 was always connected to the cable shield connection. The audio gear would connect equipment pin 1 to the equipment metal enclosure, over time Neil noticed that this practice stopped and pin 1 shield ground was being connected to the circuit board common trace and somewhere on the circuit board the common trace had a jumper wire to the chassis ground. This allowed the RFI that was picked up on the shield of the audio cables to travel. As Jim Brown K9YC puts it:
"If the shield goes to the shielding enclosure, current stays outside the box. If the shield goes to the circuit board then the current gets inside the box!"
In the ham radio world, we use computers, external sound cards, USB hubs and ham rigs. All these could have a Pin 1 problem that may be our thorn in the flesh. Now when I say "pin 1" problem with our gear there may not be a pin 1 used for shield grounding like the XLR audio connector the issue is the shielding or ground pin on the connector does not go to the case ground but to the circuit ground and then to case ground. The label of Pin 1 came from as said earlier the audio XLR connectors pin 1 with was always the shield connection pin. When dealing with radio gear it's not Pin 1 per say but the connector ground connection does it go to the chassis but to the circuit board to possibly cause issues.
If you find your radio gear has a Pin 1 issue such as your headphone plug metal housing does not go directly to the chassis ground but to the circuit board most times you don't want to get the soldering iron out to fix it. If stray RF is let's say getting into your USB cable from your rig to PC for rig control and you find you are having issues. I would suggest you first look at the grounding a bonding of your gear, it is a quality USB cable as all are not made alike. Finally, try a quality ferrite on the USB cable if it is stray RF is going places it should not. I mention ground and bonding of your gear because if your gear has proper chassis grounding but you either do not have it grounded or it done incorrectly you could be compounding your RFI issue.
I did look at my Icom 7610 shop repair manual and most if not all critical RF potential grounding is chassis grounding. On the other hand, in my personal PC, well many items are routed to the circuit board ground and then a jumper to the chassis ground. As K9YC put it regarding my PC "current is getting inside the box".
So the Pin 1 problem is something for us to be aware of and to consider when going after RFI. Now when I say RFI I am including both reception issues and equipment acting up.
Below are some great reads regarding the Pin 1 problem.
Time to bring out the black book
In the past, I have like others had issues with RFI. When RFI's ugly head pops up I find myself wondering what I have tried in the past. If you are like me you have a hard time remembering if you had tried something and if so what happened. I decided it was time to roll out the black book. I will use this book to record what I have done, the results, and the steps I want to take. Recently I felt I had my RFI issue resolved but on Monday while taking part in the Medium speed CW contest I had N1MM macros freezing. The encouraging thing is this issue happened in the past and was terrible, now it hardly happens and gives me the impression I am slowly solving it. I have a few more ideas to try and the black book will help map a path forward in the journey.
CW contesting cut short.
This weekend was the WAG (Worked All Germany) contest which I entered CW low power and unassisted. About 10 minutes into the contest things started to unravel, I was on 10m calling CQ TEST when my contest software (N1MM+) had the contact log window turn black. It came back to normal within an instant but that is never a good sign. Next, I pressed my F1 key to send CQ TEST again and it did not respond, I pressed it again and nothing. Then all of a sudden the rig keyed and the macro started to send...twice. Then the issues got worse as when I had a station come back to me and I responded with my exchange the macro would not send right away. It was very confusing to the answering station and as he sent his call again, it was then my exchange macro decided to send.
It was time to shut down and get things sorted out. I have had this issue in the past and thought I figured it out. In the past, my wired Lanovo mechanical keyboard was picking up RFI and causing me problems. I was looking for a fast fix and added some mix 31 toroids to where the cable enters the keyboard. I now have toroids at either end of the keyboard cable.
I then fired up the contest software and called CQ TEST again on 10m. For over an hour, I had no issues and hoped I had fixed the issue. This was the first time this issue had ever shown up on 10m in the past it has always been just 20m but of course now that I am contesting "MR RFI" changes things up. Heading up to 20m and again while calling CQ TEST I had the N1MM+ contest log window go black again. It was time once again to shut things down as I couldn't contest in this situation.
As I sat there and went over things I knew in the past I had had issues with my wired mouse, I went wireless and the problem was solved. Maybe it was time to go with a wireless keyboard? I have a Logitech K830 hanging around and thought I would put that into service and see what happens. Of course, the keyboard's rechargeable battery was dead and needed about an hour of charge time before I could use it. That gave me time to label the keyboard and look up the instructions on how to pair it. 
Logi wireless keyboard.
I was up and running again with the wireless keyboard and things were working just fine! This is a much smaller keyboard and I found myself hitting macro F keys when I was supposed to be hitting numbers. This also is not a mechanical key-type keyboard. I tend to be heavy-handed while my hands are resting on the keyboard waiting for the CQ TEST macro to finish. The keyboard was so sensitive by mistake that I would type letters into N1MM+ call box. When this happened my CQ macro stopped (as it should) I had to lighten up my heavy hands while I was using this keyboard.
While waiting for the keyboard to charge I made my way to Amazon and ordered a full-size mechanical keyboard. I was sure this was the solution to the issue and so far I was right. It was getting more toward evening and 40m looked like it was hopping. When I pressed my F1 CQ key my N1MM+ , Win4icom radio control software and the web page I had open all disappeared! With everything restarted I tried again with the same results, it was time to pull the plug and think about things over the evening. The wireless keyboard was not the answer. It was time to get back onto Amazon and cancel the keyboard order.
Winkeyer with ground lead to main ground bus
As the wheels in my head started to turn when using the wired keyboard the only program that was ever affected was N1MM+ and not Win4icom radio control software or the webpages I had open, it was always just N1MM+. When I started to use Win4icom radio control software along with N1MM+ I needed to purchase a Winkeyer USB unit for the CW macros in N1MM+ to work properly. The Winkeyer is directly involved with the sending of code and it's part of the chain I have not paid any attention to. All outgoing cables had 31mix toroids on them BUT I realized the unit was not connected to the station ground. On Sunday morning I opened the Winkeyer and connected a station ground wire to one of the PC board mounting screws.
I began contesting and all was good on all bands, my fingers are crossed BUT I have had this false hope in the past. Today (Monday) I have my weekly Medium Speed Contest (MST) in the morning and afternoon. It will be a good test to see if the issue has been fixed.
Also, I posted my issue on the Groups.io N1MM+ site and had some excellent feedback. One common thread was the USB outer metal shell on the PC. I was told most of the time they are floating and not grounded and this can cause issues. I have a lot of USB cables connected to my PC for radio-related things. I was given a very good idea on how to ground them but more on that in another post.
UPDATE: This morning and this afternoon I took part in the MST contest and called CQ on all bands without any issues. Maybe I have the RFI issue solved. BUT in the past I have done the happy dance to only be greeted with RFI....so we are keeping this celebration very low key.
What say you?
Here at VE9KK's household, my XYL has returned to part-time work from home and involves a computer and the internet to communicate regarding work. Last week while she was working and I was on the radio she told me her LAN connection kept dropping out. I did some experimenting and sure enough, it was my CW signal on 15m that was doing it. I quickly grabbed my last FT240-43 toroid and wrapped the CAT6 cable around it which did the trick!
I then did some reading and learned that the CAT cable should be loose around the toroid and not tight as this affects the small wire inside the CAT6. I also read that 7 turns were the magic number of turns. Finally and this is the question to my readers I read two thoughts on wrapping CAT cable around a toroid. One school of thought was it is perfectly fine and the other was it slowed down the internet speed? I never did a speed check on my wife's connection with and without the toroid. The reason for this is she never complained the speed dropped so why poke the bear! Has anyone out there had an issue with your HF signal dropping internet connections at home and what did you do about it?
Time to make a choke balun
In your ham radio adventures, you may have heard the words..choke balun, feedline isolation, common mode choke or a 1:1 balun. In my humble opinion different names for the same device. These devices for the most part are present in a system to control common mode currents. I am not an engineer just a simple ham guy trying to sometimes understand some very complicated and technical explanations.
I hope to keep the next bunch of lines understandable. Here we go...what are common mode currents! Most ham use some form of 52 ohm coax as it's very easy to route, secure and can within reason be placed anywhere. Ideally, RF will flow out on the OUTER SURFACE of the coax INNER CONDUCTOR and flow back on the INSIDE SURFACE of the coax shield. This is what happens in a perfect world but if a system is unbalanced RF current will return on the OUTSIDE of the outer coax braid. It does not return but it radiates.
This is when we hams can experience RFI issues and these issues vary with the amount of common mode currents flowing on the outside of the shield of the coax. How does one control this common mode current? One very effective way is to introduce a common mode choke, 1:1 balun, choke, feed line isolation or whatever else it could be called.
In a previous post, I did go over how I installed a 1:1 balun at the antenna feed point of my Hustler 4BTV and that seemed to satisfy my RFI issues but further reading informed me that a choke balun at the radio feed point would be very advantageous. I ordered 2 FT240-31 mix toroids as I decided to make a choke balun instead of buying one. The two common mixes for HF toroids are 31 and 43. The difference being the 31 mix is effective from 1-300 Mhz and 43 is effective from 25-300 Mhz. I chose to go with the 31 mix and purchased 2 from a reputable dealer as not all toroids are created equal. I would be very wary if you find toroids at a very low price....just remember you get what you pay for. I ordered mine from Mouser Electronics Canada, part number 623-2631803802 made by Fair-Rite for 13.00 each.
I used RG8X coax to wind around the toroid but the issue with using a ring toroid compared to a split bead is I had to remove the PL-259 to wind the coax around the toroid and then solder the PL-259 back on. I intended to make 10 loops and for some reason, I ended up with 11 turns through the toroid. I then soldered the PL-259 connector back on but not before placing the barrel and coax spacer on....how many of us have soldered a PL-259 on and then realized we forgot the spacer and barrel!
Another advantage to having a choke balun at each end of the coax is that it helps remove some RF noise that is picked up on the shield of the coax.
 |
| Dont forget these BEFORE soldering the PL-259 |
RFI noise antenna issue
I have decided to give my MFJ 1026 noise canceller a call to action and combat this RFI I have been having. In the past, I did own a 1026 and used it with great success to remove RFI our plasma TV was giving me. With the MFJ 1026, you need to have what is called a noise antenna and it has to hear the same RFI your main antenna is hearing. When I used this device to remove the plasma TV issue I had no problem getting the noise antenna to receive the RFI loud and clear.
This brings me to my head-scratching moment with the RFI issue I am having now. The first noise antenna I tried was my H-800 active antenna. I picked up nothing, I then placed it from vertical to horizontal without success. I then moved it around the property and still I was not able to hear the RFI. I then tried a long wire in the horizontal plane fairly close to my Endfed antenna and I picked up nothing. Again I moved this long wire around the property without success again.
I do have a spare Endfed that I used for portable use and I ran some coax outside to set up the Endfed close to the main Endfed and low and behold I picked up the RFI BUT it was way too close to my main antenna. The MFJ unit (via an internal light/fuse) was indicating far too much RF was making its way down the noise antenna.
The other thing I tried was the change the length of coax that feeds my main Endfed antenna but this made no change to the RFI issue. In July we are getting a new aluminum gutter installed and I may try using that as the noise antenna and see if that works. Up to this point I have not been able to pick up the RFI that the main antenna is picking up.....well other than almost putting a noise antenna ontop of the main antenna that as I said did not work out very well.
