Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: Transmitting trips CGFI's in house...  (Read 2616 times)
K2RLH
Member

Posts: 23




Ignore
« on: October 09, 2011, 08:03:20 PM »

We just tuned up 15, 20, and 80 meters on our 5BTV so my son could start working those bands. When we installed it 10 and 40 were tuned enough to work, and he's been using those bands for a while. We finally got back up there to tune the last three (and fine tune the first two) and he's been working them all weekend, reaching places like Finland and Sweden with excellent reports. This is from North Idaho with 100W out of an FT-857D.

However, we've noticed something interesting: When he uses 15 and 20 meters, two GFCI's in our house have a tendency to trip. They trip at the same time, though they are on opposite sides of the house and on completely different circuit breakers. They are identical makes and models.

Only 15 and 20 seem to do it, so it appears to be wavelength related. I presume we are coupling into the house ground and when the GFCI's see the unexpected voltage, they are fooled. There is no question the RFI is causing it - I happened to be near one after resetting it and when he keyed down it instantly tripped.

Our house has a metal roof and the 5BTV is mounted right in the center at the peak. Some, but not all, of the roof is bonded with heavy tinned braid. At present I do not have a decoupling choke at the antenna. Coax is LMR-400. The GFCI's are integrated into application-specific thermostats so I cannot just switch to a "different brand".

Anyone else ever experience this? Will a decoupling choke at the antenna solve it? Bonding the rest of the roof? Any other ideas?

Thanks!
« Last Edit: October 09, 2011, 08:05:48 PM by K2RLH » Logged
K2DC
Member

Posts: 1385


WWW

Ignore
« Reply #1 on: October 10, 2011, 02:17:38 AM »

A choke balun at the feedpoint certainly wouldn't hurt.  Bonding the rest of the roof might also be helpful.  You should also have tuned 1/4 wave radials for each band, which the metal roof may not be providing whether it's bonded or not.  But it's also possible that none of those will solve the problem.  Some makes and models of GFI breakers are just extremely sensitive and susceptible to RFI.  And some GFI breakers just get hinky with age.  You might try the choke balun frst, because if there are antenna currents on the feedline shield, there's no telling where it could be coupling into the power line.

73,

Don, K2DC
Logged
W8VZM
Member

Posts: 67




Ignore
« Reply #2 on: October 10, 2011, 07:08:31 AM »

Don't forget to verify the quality of the ground connections on the GFCI's, the service panel and the ground rod. If any of these are poor the the ground will act as an antenna resonant at some frequency. One of the first things I do with RFI of this type is inspect and tighten all ground busses in the panel and double check all ground connections in the affected circuit. It might be in another outlet or in the fixture itself.

Ron W8VZM
www.w8vzm.com
Logged
K1ZJH
Member

Posts: 1185




Ignore
« Reply #3 on: October 10, 2011, 07:11:00 AM »

Make sure the balun is below the "radial" system when you install it.

Pete
Logged
N7SMI
Member

Posts: 373




Ignore
« Reply #4 on: October 10, 2011, 07:15:42 AM »

I had this issue with 2 separate GFCI's. I simply replaced them with newer models and there have been no issues since. I guess the older models had no requirements for RFI filtering.
Logged
WX7G
Member

Posts: 6321




Ignore
« Reply #5 on: October 10, 2011, 07:31:46 AM »

I had a similar problem with GFIs tripping. Replacing them with RFI resistance GFI's cured the problem.
Logged
K2RLH
Member

Posts: 23




Ignore
« Reply #6 on: October 10, 2011, 09:07:26 AM »

Thanks for the feedback. A couple of responses:

1) I cannot just replace the GFCI's with newer/different models. These are integrated into the themostats that control the heating system in our tile floors. They are specialized devices, not just standard GCFI outlets or breakers.

2) These GFCI's are new (within the last 12 months) and I installed them myself, running them on brand new romex runs straight from the breaker box. Each has its own, dedicated breaker. I can assure you their connections (all of them, not just ground) are solid. They are on opposite ends of the house so the romex runs are in different directions and radically different lengths. Yet when they trip, they trip together every time. This is why I believe it's EMI conducted into the overall wiring system of the house.

3) One of the GFCI's directly controls the heating wire in the floor, so potentially that wire could act as an antenna. But the other one just controls an immediately adjacent AC relay which then controls the heating wire, so that GFCI has no connection to the "antenna in the floor". The one thing they have in common is the house wiring - all the way back at the breaker box. But from the breaker box there is a nice spidery network of romex going everywhere which could act as an antenna.

My current ideas are to install the "coiled coax" choke, install a delta configuration of low-value caps on the power side of the affected devices, and possibly insert some sort of EMI suppression into the line cord of the DC power supply.
« Last Edit: October 10, 2011, 09:10:21 AM by K2RLH » Logged
KC9RCG
Member

Posts: 142




Ignore
« Reply #7 on: October 10, 2011, 09:25:28 AM »

Now you got me paranoid.....

I'm getting ready to buy my first HF rig and run a dipole up a tree and run the cord in the house. It is likely to happen with me? Or, is this likely due to being attached to the house and metal roof?
Logged
N2JDQ
Member

Posts: 60


WWW

Ignore
« Reply #8 on: October 10, 2011, 10:03:08 AM »

Make sure the balun is below the "radial" system when you install it.

Pete

Different subject sort of but wanted tro ask a question abt this, On my 80m Inv L I have a coaxial choke on a 6.75" form.. its about 2" above ground and my radials, would it be better to have it lower than my radials? If so whats the reasoning behind this? I could easily do it, by just having the coaxial choke in basement below ground level about 3' from where it is now.

-Steve Raas
Logged
K2RLH
Member

Posts: 23




Ignore
« Reply #9 on: October 10, 2011, 03:56:46 PM »

I'm getting ready to buy my first HF rig and run a dipole up a tree and run the cord in the house. It is likely to happen with me? Or, is this likely due to being attached to the house and metal roof?

My searches today have revealed lots of people with this problem, with all sorts of antennas on all sorts of support structures including towers, trees, etc.

In many cases, they fixed it with newer GFCI's. Apparently there are later standards that intentionally reject RFI to some degree. In other cases they fixed it by solving a grounding issue.

In one writeup I read, the guy disconnected a romex ground lead at both ends and then used a TDR meter to determine its length - which turned out to be almost exactly a resonant wavelength in the troublesome band. Thus it acted like an antenna tuned for that band, and did its job very well! {grin}

After much reading and discussion, I don't think it's specifically due to the antenna being on the roof, having a metal roof, etc. My next steps are going to be building a coax choke for installation just below the feedpoint of the antenna, and perhaps some ferrite beads on the power leads going into the affected GFCI's. The former will cut emissions coming back down the shield, and the latter will help suppress any that still make it onto the house's power grid.

I've already got the male N connectors, and I just picked up some female N's so I can place the choke inline with the existing feedline. This will permit me to easily swap the choke in and out for experimentation. I haven't found a local source for series 7x ferrite beads yet.
Logged
KC9RCG
Member

Posts: 142




Ignore
« Reply #10 on: October 10, 2011, 04:03:26 PM »

Is this caused by emission from the antenna or it reflecting back in the radio and messing with the power source?
Logged
K2RLH
Member

Posts: 23




Ignore
« Reply #11 on: October 10, 2011, 08:12:36 PM »

Is this caused by emission from the antenna or it reflecting back in the radio and messing with the power source?

I don't know, yet. My suspicion at the moment is RFI coupling into the house wiring network. Think about it - there's hundreds of feet of romex power lines running all over the place in a typical house. If one or more of those lengths is resonant at the frequency in question, it can act as an antenna and then conduct that signal into the GFCI's.

However, the feedline acting as part of the antenna is a possibility too, which could conduct RF frequencies back through the radio to the power supply and then the wall.

I'm hoping to attack both this weekend.
Logged
K1ZJH
Member

Posts: 1185




Ignore
« Reply #12 on: October 11, 2011, 10:06:59 AM »

If it was my house, I'd be adding .01-uF  UL rated XY caps across the
those GFIs to reduce the RF problem. But, being a public forum, that
advice may conflict with the UL rating for the GFIs, electrical boxes and
national or local codes.

Pete
Logged
N5RMS
Member

Posts: 37




Ignore
« Reply #13 on: October 13, 2011, 06:09:34 AM »

I also had this same problem on a GFCI outlet.  I used a ferrite bead on the romex cable, close as possible to the outlet.  No more problem. You could possibly do the same at the circuit breaker.  Ferrite bead was about 3/4 inch long, 1/2 inch round or so.
Logged
K2RLH
Member

Posts: 23




Ignore
« Reply #14 on: October 15, 2011, 05:29:11 PM »

Fixed! (At least, we believe so.)

I obtained some Fair-Rite Type 77 ferrite beads and toroids. This morning, experimentation showed that transmitting on 14.230 MHz would trip both GFCI's repeatably.

Next, we installed ferrite on the power leads going into and out of the first GFCI thermostat, along with the leads from its temperature sensor (this is a tile floor heating application). We got two passes through the ferrites on the AC in and out, and four passes on the sensor wire. We reinstalled everything and fired up the radio.

Transmitting tripped the second GFCI, but not the first. So we duplicated the ferrite installation on the second GFCI, and it too has been reliable ever since. This is after an entire day of my son using 15 and 20 meters to talk all over the world, so it's looking pretty solid.

Thanks to everyone for their input and advice!
Logged
Pages: [1] 2 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!