My computer, was an old Dell server, which was transplanted planted, into a very large case, not all MB & case, aligned, to secure mb to case. This one machine emits RFI strong enough, to cause interference with up signal to GDO (garage door opener), from 30 feet away. What got it moving was a small 'Ferrite' on the GDO's antenna wire. 3 other Pc's in the garage, ready built, by Dell etc. NO MODS, these machines, do not cause GDO problems. I read about bad grounding can cause in a PC, can emit RFI: How do I ground it? _________________ Outside the box, attach wire to, to a screw on PSU, other end, perhaps to face plate on wall outlet? Or do I need to attach the wire, inside the box somewhere? I imagine a proper ground the RFI noise, follows the ground, eliminating RFI to GDO
The PSU screw should be fine - assuming the motherboard is properly mounted and all PSU power connectors are securely fastened. But you should be getting good ground through the PSU's power cord to the wall outlet - assuming the wall outlet is 3 prong and the outlet is properly wired. Every home and every computer user should have access to a AC Outlet Tester to ensure your outlet is properly wired and grounded. I recommend one with a GFCI (ground fault circuit interrupt) indicator as it can be used to test bathroom and kitchen outlets (outlets near water) too. These testers can be found at most home improvement stores, or even the electrical department at Wal-Mart. Use it to test all the outlets in the home and if a fault is shown, have it fixed by a qualified electrician. If your outlet is the old 2-prong type, I recommend having it replaced with a 3-prong that is properly wired. In the meantime, the wall outlet face plate screw might provide an Earth ground, but it might not. You would have to inspect the inside to see if the house wiring to the outlet is 2 or 3 conductor, and if 3, if wired properly to the outlet. Again, something that might better be left to a qualified electrician. Another alternative is to run a ground wire from that PSU screw to a copper cold water pipe. But this is not ideal either as it assumes there is a direct run of copper from that point that sinks several feet into the Earth. And note it must be the cold water pipe and it must be copper pipes. Many newer homes or homes with updated plumbing use PVC pipes. BTW, such strong RFI suggests a faulty component for which no amount of grounding may help. You should also make sure your GDO is properly grounded too. If it is, many openers let you change channels to avoid interference with other devices and openers (or their remotes) in crowded neighborhoods.
Thanks Bill grounding did not help, out side of case, house is all 3prong outlets. Perhaps bad grounding case to mb, or bad part. Good idea about adjusting receiving frequency, I'll see if it's possible with my GDO. Is there someway to see the interference, Kill things one by one, to see if additional RFI is problematic? Note with 'Ferrite' on "antenna" test many times, UP down worked FINE, this morning worked one time, then no up, unplugged, the suspect ground machine, no up. Perhaps other sources of RFI. Thanks
What does "out side of case" mean? Not likely motherboard to case because the motherboard is grounded to the case at every motherboard mounting point and most motherboards mounted at at least 6 points. Then they are grounded again through the motherboard power connectors, through each expansion card when it is secured to the case, and through each port through the rear panel I/O shield. So more likely a bad part. Probably not. What are you going to kill? You can disconnect drives, and maybe run with one RAM stick at once. Beyond that, everything is integrated into the motherboard. You can try different cables (notably the network, graphics, and sound).
"What does "out side of case" mean?" Not removing exposing internal parts, you see the case power switch etc. "Not likely motherboard to case because the motherboard is grounded to the case at every motherboard mounting point and most motherboards mounted at at least 6 points. Then they are grounded again through the motherboard power connectors, through each expansion card when it is secured to the case, and through each port through the rear panel I/O shield." That Dell server was transplanted to a huge case, the person doing the transplant, said quite a few of holes (mating mb to case did not align). Was told by case makers support, (he had pics of MB, & recommended) which model case use What are those things called to attach mb to case. That's why I thought grounding, (remove one PSU screw install wire then other end to 3 prong center screw of wall outlet. Poor ground to case perhaps, my ground no sufficient enough to allo operation of up. FYI - found by Mr. Google 1. FCC sets specific frequency, that are built into GDO logic, and cannot be changed. 2. That forum said Sears garage door mechanics, carry with them a device that shows RFI emissions, coming from individual sources. Also when less frustrated, I'll, turn off one circuit breaker at a time, then test up, if no up, turn off garage breaker & plug in elsewhere. Narrow the circuit down, if, still no up, then perhaps, neighbors interference. Just thought try long extension cord plug, GDO other circuits house, then test. Would you like to see those comments from the other forum? Other wise it will be gone from FF's history soon
Cases are designed to support 1000s of different motherboards of different sizes. So there typically are more mounting points in the case than there are in the motherboard. But if ATX compliant (as they should be), any mounting holes in the motherboard will correspond and align properly with a mounting point in the case. So unless something was proprietary, they should have matched - even though there could have been several unused case mounting points. Note there should have been a standoff in the case for each mounting point in the motherboard, and no extra standoffs in the case. A common newbie and distracted expert mistake is to insert more stand-offs in the case than needed. But that typically results in the motherboard being grounded in the wrong place, shorting out the board. I didn't say "frequency". I said channel. Each frequency is divided into several channels. This is the same with your wifi network too. Not all openers allow changing the channel, but some do. Note you said your computer is causing your garage door to open. That suggests the wayward emissions causing the interference is coming from the computer, and not the garage door opener. There are frequency signal (or field) strength meters, but you might be able to accomplish the same thing with a "packet sniffer" program running on a notebook. I use XIRRUS WiFi Inspector to see what wireless channels are in use and available. NirSoft's WifiInfoView is another good one. Also popular is inSSIDer.
Got it change the channel the frequency will use. That would be done, by accessing, & changing channels from, online for Router, on 2.4 ghz band. I do have inSSider. I have a Netgear WiFI booster plugged into an outlet, it allows you to see from the device 2.4 or 2.4X, likewise, two for 5ghz band choices from device 5 or 5X. That solved wifi radio problems on patio. Can the router be made to see only the more powerful 2.4X (boosted signal) signal, so the GDO would then have only access to the 2.4X. I'm assuming I'm correct about channel changing
I was talking about changing the channels the door opener uses, but if you want to try it on the router, you can try that too. IDK. RTFM.
See page 8 of this: Code: https://transition.fcc.gov/Bureaus/Engineering_Technology/Documents/bulletins/oet62/oet62rev.pdf I suspect that the server was a Class A device. Only Class B devices are certified for residential environments.
@mirimir On page 8, talks about "A" & "B" devices one deemed for biz, the other for consumers. My GDO was/is consumer, installed by builder of home(S), & the consumer model installed was "Lift Master", Chamberlin is the corporate owner of LiftMaster. Model 41A5021--4M-315 IC 1021499A, & just above that number, says IC: 2666A so far cannot confirm, that, that "A" refers to FCC type A or B, all refer to Chamberlin LiftMaster. ____________________________________ Here's my latest path to fix, (does this make sense?). Ferrite still on GDO antenna, RFI computer, unplugged. Yesterday UP worked one time, moments later UP failed, then with same conditions my home, approx 7pm UP works. Previously just turning off, the noisy RFI computer was enough, to get UP to work, then later it stopped working. I believe, noisy RFI machine off, allowed UP to work, until neighbors, started using WIFI, which caused UP to fail, then when not using WIFI, up would work. Trying diff router channels 2.4ghs, 1 then 11 or 13, may allow operation, inSSider, can help with choice. Monitor each channel during the course of the day, to see which provides most consistent, UP/Down. I do not believe a new GDO would work, question to support GDO makers, NO shielding by mfg from RFI, mine & new use 'rolling code or frequency within FCC designated 'bandwidth', so my fail likely, will happen with new GDO. Also the car being outside, when trying UP, is a straight shot, for RFI (RFI out the neighbors wall, straight shot to outdoor, car, less obstacles for RFI to outdoor car. With RFI noisy PC on or off, par, in the garage UP worked. I might be wrong about on or off of noisy PC. So being in the garage, somewhat shielded RFI from neighbors, by car being inside garage. I used to get periods where my patio wifi radio would or would not work, that solution, was NetGear wifi extender. I'm curious, if the 2.4ghz band can be tuned off, to eliminate, outside interference. Then perhaps antenna ferrite, fingers crossed, can tolerate the noisy RFI machine. Note I have two wifi radios one 2.4 other 5.0 ghz & two, one BBQ smoker which uses wifi, I think I can 'not use wifi' & jut use ethernet (thinking out loud here), no wifi no RFI from neighbors. Damn forgot about the printer. Last resort here for no 'Router' Thanks
That's not right. You need to check the grounding on your opener and how it is configured. Sure there is. Every RF device has some shielding.
"Sure there is. Every RF device has some shielding." that is not factually correct from Ryobi GDO support. Also your comment does not add to, FIX. Perhaps all GDO's have some level of RFI protection, but not improved, to warrant Tech support, to offer as a solution. Previously you spoke about channels. When I spoke, about Grounding, you talked about channels, which indicates/alludes to Router Channels. Seems were a little out of synch! Evidence seems to say, neighbors RFI is a problem as well (see above). Okay! I will check that outlet GDO plugged into for proper ground. So as to eliminate the culprit. Thanks Bill
@Bill_Bright, The Tripp-lite surge protector, which GDO is plugged into has a 'red light' indicating ground, shining. Is that sufficient proof, for proper GDO grounding Naive question from literally, 'retarded in electricity', me. Trying to learn here!
I was talking about your server, not your GDO. It seems pretty clear that you have a Class A server, which was not certified for residential use. And so it emits more RF. Enough to mess with your GDO. And additionally, I gather that the motherboard is in a nonstandard case. So your best bet is shielding the server better. Maybe build a Faraday cage, using a wooden frame with aluminum window screen stretched over it. But even if you did that, you might still have too much transmission via the power wiring.
Well that guy is just wrong! Sorry, but he clearly is not a "radio" technician. But I am. That's my background (look at the link in my sig). I maintained radio communications systems for the USAF for over 20 years. It is just a fact that RF devices, both receivers and transmitters MUST have RFI/EMI shielding to both prevent causing interference with other devices, and to prevent being affected by interference from other devices. If nothing else, the FCC requires it. Even the most basic antenna cable has shielding. Of course it does. I specifically said you need to check your grounding. And you did via your surge and spike protector (S&SP) and that was good. And while a good indication you have a good ground, that is not proof everything is okay. Your opener could have physical damage. Or the protector could be bad. Even they must use shielding. Have you tried running your opener without the S&SP to see what happens? No it doesn't. You just jumped to that conclusion without taking a moment to understand what I said. Just because WAPs (wireless access points, typically integrated into wireless routers) use channels, that does not automatically suggest, indicate, imply or allude that any comment about channels must refer to routers. That's just silly. If I say "tires" are you automatically going assume I mean car tires? I may be talking about the tires on my bicycle or truck. Read it again above again and see that I specifically said (my bold underline added this time), "Openers" clearly indicates I was specifically talking about your garage door opener and was NOT suggesting or alluding to any wifi router.
Oh my! I was barking at the wrong tree, that server did not interfere, previous car (2014), & never interfered, with GDO's handheld openers. Even Noisy machine off & unplugged, the car's GDO's opener, has times during the day (it will work, & not work) on the driveway. When car has difficulty with UP, parking the car in the garage, UP works. Doesn't that indicate, outside interference? That 'server' runs, full blast 24/7, how do I handle the heat, it will most likely get very hot in the box. Okay! That machine has always used "ethernet". Would ethernet long cable, be transmitter of RFI shorter ethernet better than long cable? So would, the source be AC side With ethernet acting as antenna. The noisy machine, parts were replaced with consumer products, NEW PSU, 2 new 1070 video cards, new case. Would distance Noisy to GDO, weaken RFI from noisy to GDO. I can move it, 60 or 70 feed from GDO, but still, would get (noisy off, no wires connected) times during the day, driveway car fails UP, in garage it works. If I move it that far away, I would need WIFI instead of ethernet, would it using WIFI, just make it worse & using wifi extender's choice 2x (stronger wifi, make RFI worse)? A neighbor said, his new GDO uses two antennas, of different lengths wire. Is that something to look at? The most luck so far is a small snap on "Ferrite" on GDO's single wire antenna. Also perhaps 25' from GDO removed, screw from PSU attached lamp cord, other end to middle screw outlet AC face plate Grounded, GDO in driveway some hours UP fail other hours ok. Thanks mirimir I really appreciate your help Take care
You cannot just add another antenna. You are talking about a new opener. Two antennas are nice in some areas where you may approach your home from different directions. The fact they are of different lengths could indicate it is a dual frequency opener. These are often necessary in heavily crowded RF areas - like around an airport. In fact, there are even converters to make single frequency openers dual channel for those problem areas. It might, but might not. The opener is located inside a building and surrounded by walls - typically thick fire resistant walls, and perhaps behind a metal garage door too. Those "barriers" greatly affect RF propagation. Range and obstacles in the path are always factors with radio communications. It could also mean a weak battery in your remote reduces the range too much when outside your garage and when on the other side of those barriers - unless your remote is integrated with the car itself, as some are. Of course - if that really is the source of the interference. Distance is always a factor.
@mirimir @Bill_Bright Learning allot here, many thanks. Update!!! at router url > wireless > 2.4 band > changed "wireless mode" to "N-only", "Channel Bandwidth" 20mhz, "Control Channel" auto, shows using 6 at time of screenshot Configured this way, worked all day (hope this isn't 'fools gold'). PRIOR TO MAKING ABOVE CHANGES! Fired up old Laptop, wirh InSSIder, saw my network channel 11, & many users on channel 11, my network had highest peak on on InSSider, graph, I then went out & the door opened AMAZED Also "wireless mode changed to "Legacy" killed UP & DOWN. Back to N-only Now I'll test a few days, see if UP works consistently, then try to add, the noisy computer back, via ethernet. The computer could have been a FALSE POSITIVE, as perhaps, when up found not working, unplug PC, & caught a momentary lull from other sources, allowing door to open. If this server is an A, does changing to consumer parts (case, psu, (2) 1070's) a B machine?? Also i did see special thick Tin foil at Amazon for this purpose. Can I leave front back & vents open (no foil) or does that defeat the purpose. If necessary a 2nd layer of foil, inside the box, leave front, back vents no tin foil Outside PC shiny tin foil next black case, next shiny tin foil in side. Like bread meat bread
Reading your post leave me confused Is it the server that's messing with your GDO? Or your router? Or is it some neighbor's device? And I suppose that it could be all of them As I understand it, motherboard design is at least one issue re RF emissions from computers. Also case design, of course. I don't recommend playing with aluminum foil. Shielding microwave radiation is a nontrivial thing. It's not just about blocking holes. There's reflection and standing waves and all sorts of subtle craziness. Maybe Bill can say more. It's mostly over my head.
Lots of devices can emit RF noise. Basically anything electronic that switches rapidly from one state to another can emit RF noise. This includes ICs and switching power supplies. This is why cases are designed to shield as much RF as possible. Look at those rear panel I/O shields. They are called "shields" for purpose. Nte all those little prongs in the connectors. Those are there specifically for RF "shielding". And it is not just to suppress emissions from the components inside, but to [hopefully] block any RFI/EMI from affecting sensitive devices from outside interference sources. I think messing with aluminum foil is just Mickey Mouse. Don't forget for any shielding (including a screen room/Faraday cage) to be effective, it has to dump any absorbed extraneous RF energy to Earth ground somewhere - typically by a direct, solid copper grounding wire to a properly set grounding rod. Another common source or RFI/EMI is the network card, network cable connectors, or the network cable itself.
So what are the main kinds of differences between Class A servers and Class B desktops? I have an old Dell R710. It's a Class A device. And yet it has very impressive RF shielding. Far better, at least superficially, than any desktop that I've owned.
Thank You mirimir! 1. At one point I strongly suspected, server was, interfering with GDO. As UNPLUGGED SERVER seemed to solve, only to fail hours or days later. 2. Unpluges server, & GDO still fails, at random times 3, Try small ferrite on GDO antenna, this worked for awhile,then random fail of GDO. It worked longer with antenna ferrite attached, but fail randomly. step 3 was a little, better than, step to, for GDO UP 4. Server still unplugged, I suspect RF from neighbors, causing RF for me, that during "peak usage" causes UP to fail 5. I reread, about router channels, watching with 'InSSIDer' 7. I changed router url setting for wireless > 2.4Ghz. From AUTO > " N-only 8. (N-only), observe, & see it it fails UP over time, like previous steps. Should N-only solve the problem! 9. Introduce start server, if server on, up works consistently. Finished Done So'ved. Step 10 was asked so as to prep for fail of step #9. If necessary #10, would be plan, gather. so as to cure 10 Sould server stop UP from working. Then shield server from GDO. Suggested here & elsewhere try Faraday box, this would cause heat issues (distributed computing for server, runs the machine at max throttle 24/7). Or how about tin-foil, on the outside of the case, curious leaving, "back, front, vents, no tinfoilk, would leaving, no foil on (front, back vents) defeat the purpose of foil. If still fail, would additional foil help. Sorry I caused confusion. Thanks
