Emi/rfi - q & a

e.Man

Member
i have a hunch EMI/RFI problems are often overlooked in the RC industry. i have some experience in solving these issues, but i'm no expert.

for those looking for solutions to interference problems, there are a few products i'm aware of. you can look on MM-Carr - they have some stuff including paints, tapes and foils. another product is MuMetal but i think it is pretty expensive. can find that here.

the cheapest way to deal with the issue may simply be aluminum foil. how effective household aluminum foil is compared to the same thickness of specialty products i have no idea.

i also remember something about grounding but i don't remember the particulars. let's take a long servo cable for example - if you wrap it with a foil, i think 1 end, and ONLY one end, of the foil tube should be grounded to the (-) batt terminal. if anyone can expand on this i would be grateful.

another possible source of EMI is any wire/cable because wires can act very much like antennas if the length is right. i'm really foggy on this, but i seem to remember having to protect servo extensions in excess of 24 in. or so when you're flying a 72 MHz system. what i am told works, or helps, is simply to braid the cable (or buy braided cable). can do this easy enough by slipping out the connectors on one end, braiding the cable and re-inserting the pins.

i'd like to hear onions and known solutions to EMI/RFI problems regarding the above, as well as your own solutions.
 

CaptnKliegle

Well-Known Member
As for unshielded cable lengths, I would stay away from 1/4& 1/2 wavelengths of any nearby transmitter. Otherwise you would end up with a really resonant antenna to introduce RF energy into servos and RXR's. Not a problem if no video transmitter is onboard. When an analog video signal (wide bandwidth due to AM) is transmitted from the aircraft, especially with any OSD/FPV gear, proper frequency planning of RC link, video downlink, and GPS RX is in order to keep RFI to a minimum. Near field RF will also swamp sensitive receiver gear as well. Not to mention harmonics not filtered out by careful engineering practices. :eek:

Length of 1/2 Dipole antenna (ft) = 468/F(mhz)
Length of 1/4 Dipole antenna (ft) = 468/F(mhz) x 1/2

So,
1/4 wavelength at 72MHz = 468/72x 1 /2= 3.25 feet.
1/2 wavelength at 72Mhz = 468/72= 6.5 feet (That's a BIG RC Airplane!):D



Shielding for RFI is done as you mentioned, the outer shield of multiconductor wire is grounded only on ONE end. Leads exposed to RF fields should be kept short as possible too. Aluminum foil (thinner the better for weight savings) does work and is used commonly in shielded wire. Carbon fiber is conductive as well, CF tape can strengthen as well as help prevent RFI on cable runs.

Best advise to to separate as much as possible any transmitter antenna (video downlink) from onboard receive antenna (RC link & GPS). Also antenna placement is very critical. Video downlink RXR's typically use small vertical antennas, so transmitter antenna should be vertically polarized as well. Otherwise there's a 20dB loss (Huge!) at the receiver end when TX signal is horizontally polarized. That's just one of the reasons why one has video drop outs using this setup when the aircraft is in full view, but performing steep banks or pitch attitudes. This is where a higher gain antenna in the form of bigger 'rubber duck' high gain antenna, a Yagi beam type, or a patch type antenna helps with the drop outs. There's some very good videos on the web demonstrating differences in using different gain antennas for the video receiver.

The little 'rubber duck' antenna used for video downlink transmitter emits energy 360 degrees around it, typically the maximum amount is close to 90 degrees to the antenna (perpendicular). Pattern may be distorted slightly due to metal objects in the antenna's near field. Imagine a bagle or doghnut placed over the antenna and the anenna sticking up through the hole. Note that the thickest part of bagle or doghnut is perpendicular to the antenna at about 90 degrees. As the angle increases or decreases from 90, there is less bagle/doghnut. As you approach the top or bottom of antenna, no bagle, no signal. Placing other receiver's antennas directly above or below is optimum, but may not be physically possible. Avoid placement of receive antennas directly broadside to the transmitter antenna if at all possible.

RFI is also generated from ESC's and oscillator(s) in OSD gear and GPS equipment, but usually in VHF/UHF spectrum.

Braiding of cable works as well.... ever look at Cat5 wire? It's four pairs of wires, each pair is twisted upon itself and then upon the other pairs.

RFI hardening can be expensive for manufacturing, but 90% (?) of all RC flying is done 'normally'... ie no OSD, GPS, video cameras, video TXR, etc etc.
Not cost effective to spend all the money in R&D and production costs to make the remaining 10% happy, when the other 90% RC users are satisfied with equipment as is.

JMHO:cool:

Kriss... Still windy, cold, snowy in NH....
 

e.Man

Member
thanks for sharing that Kriss! i'll be saving that formula - good to have.

regarding video Tx ant., when i was doing aerial photography with a heli i was using 2.4. it was highly suggested to use at least an 8" coax extension to hang the dipole below the airframe. another way i rigged it was to cut a 4" or so aluminum circle and mount the antenna base to that with the antenna pointing down of course. the latter may not be best for planks, but i'm wondering why i'm not seeing extensions used to hang the antenna? i almost shudder to see a 2.4 dipole on the same plane as any other electronics or wiring. Db loss should be minimal if no bends and a somewhat stiff cable would keep it pointed down i would think.

this time around i'm going with a lower freq. i think - likely 900. to much multi-path with 2.4 unless you spend allot of clams (diversity) and i'm hoping 900 will run a little smoother. i'm also assuming 900 would be better suited to a moving craft. will be implementing antenna tracking too :D

also thinking about a high gain dipole for the video Tx if i can figure out how to keep it from swinging.

oh, another ? regarding shielding and grounding -- if you wanted to foil wrap something like a receiver or servo for instance, where you have a box instead of a cable more or less, how would you ground it? i suppose the best way would be to wrap the servo AND cable with one piece and ground one end as usual?
 
Last edited:

CaptnKliegle

Well-Known Member
Copper sheeting.... draw out on paper/light wt cardboard the enclosure (exploded diagram with cut and fold lines), transfer to copper sheeting, then you can spot solder the tabs/folds together to have a Faraday Cage around RXR or what ever. One little wire from GND and soldered to anywhere on the copper sheet and good to go. I think the January issue of QST has a large article on building equipment chasis this way for RF proof enclosures. I had a Home Dumpo light touch lamp dimmer for under cabinent lighting... thing emitted so much RF on the HF bands, even after hiding it in copper sheeting. The plate to touch happened to be the emitter as well as the wire going to it. Yanked it out an installed one of those wall mounted SPST light switch with a little slider next to the toggle. Quiet as can be! My HF gear is happy as is I to be on HF when lights on or off!

I picked up at a ham radio fleamarket, RG-174 coax w/SMA's installed. Some were male-male, male-female, etc. Some even had a bulkhead SMA male with very ridgid coax. Could bend into position and would stay... probably in flight up to cruise speeds.. pending upon drag of antenna. I took the flexible coax, cut one end off and soldered my 1/2 wave dipole (3.2" each leg) to the center conductor and shield.

My EasyStar project, used the extension and installed a semi ridgid 900Mhz dipole just FWD of vertical stab. Made from RG-174 coax to flex, where it exits the fuselage, I placed heat shrink tubing to stiffen it up to somewhat keep it's dipole shape. It's mounted vertically, so landings sweep the bottom one back... but bends back into shape before tossing into the air.

If I only had a 900Mhz slug for my Bird43 wattmeter, I could really nail down the dimensions to a mm. There's videos on some site where the guy was flying all sorts of antennas... rubber ducks of various gains and the homemade dipole.... dipole won! Best when the RX station was using patch antenna as well. He flew a racetrack pattern in a Yardstick (?) or similar testing various antennas on both TX and RX, so good data could be derived.

My RC link is Spektrum (DX6i) for now... still staying in the neighborhood.
Been to bloody cold and windy to test the various antennas I got to extend the range. 2dB duck, 7dB duck, 15dB patch. Got the jumper w/appropriate connectors to mount a reverse polarity SMA male where the stock antenna was on the DX6i.
At some point... may have to go to one of the 440Mhz RC link systems for real distance flying. For now, 2.4Ghz is fine.

As for protecting servos, not much inside to pickup RFI. I took one apart to reverse the action for the flaps on my Cularis Motor Glider. A Y cable is used for both flaps and another for the ailerons. I think most if the RFI is picked up from the wires to/from servos.

I had the 900Mhz TXR w/stock antenna mounted atop my Radian glider (while waiting 2 +1/2 months for EasyStar to ship), between the wings, overhead the Spektrum 2.4Ghz RXR and tail control servos. Both servos twiched with video TX on. I placed a layer of tinfoil between the TXR and smoothed it out over the contours of the wing/fuselage junction and out onto the wings a few inches. No more twitching! Later I ran the stiff coax with the bulkhead connector down the top of the fuselage to locate the video TX just forward of the vertical stab, no twitching there with foil or w/out foil. Distance from servos/wires does the trick!

Before you go crazy adding weight to the aircraft, if you can test all equipment before glueing sealed areas of wings/fuselage/tail/etc, try it.
I hate to see wires running down the side of aircraft, too much drag! So... I attempt to close in all wiring if I'm able. The coax and antenna are now permanently sealed into the Easystar Fueslage, as is the remote RXR for the AR6200 RXR. In dark lighting, I can see the LED glow, so I can troubleshoot with the flashing LED when done flying.

http://www.rc-cam.com/forum/index.php?/topic/2821-low-cost-diy-900mhz-omni-dipole-antenna/

Which will lead you to this....
http://www.rcgroups.com/forums/showthread.php?t=927590&page=2

Some interesting ideas, backed up with $4500 spectrum analyzer.

More good stuff to read! A notch filter to help with spurs/harmonics created by 900Mhz TXR....
http://www.rcgroups.com/forums/showthread.php?t=937068#post10721267

That should keep everyone busy for a while... until Christmas...LOL

73 de Kriss
KA1GJU




thanks for sharing that Kriss! i'll be saving that formula - good to have.

regarding video Tx ant., when i was doing aerial photography with a heli i was using 2.4. it was highly suggested to use at least an 8" coax extension to hang the dipole below the airframe. another way i rigged it was to cut a 4" or so aluminum circle and mount the antenna base to that with the antenna pointing down of course. the latter may not be best for planks, but i'm wondering why i'm not seeing extensions used to hang the antenna? i almost shudder to see a 2.4 dipole on the same plane as any other electronics or wiring. Db loss should be minimal if no bends and a somewhat stiff cable would keep it pointed down i would think.

this time around i'm going with a lower freq. i think - likely 900. to much multi-path with 2.4 unless you spend allot of clams (diversity) and i'm hoping 900 will run a little smoother. i'm also assuming 900 would be better suited to a moving craft. will be implementing antenna tracking too :D

also thinking about a high gain dipole for the video Tx if i can figure out how to keep it from swinging.

oh, another ? regarding shielding and grounding -- if you wanted to foil wrap something like a receiver or servo for instance, where you have a box instead of a cable more or less, how would you ground it? i suppose the best way would be to wrap the servo AND cable with one piece and ground one end as usual?
 

e.Man

Member
good stuff Kriss

i've built antennas for 2.4 like in your first link, but never had the equipment to accurately test them.

also duplicated Bill's 2.4 ground plane. i used to crack ping-pong balls in half and use them to cover the GP elements. i don't recall the results, but i think they were marginally better than a typical omni, but difficult to solder (for me) without burning the inner coax insulation.

the antenna in your 3rd link (910 MHz) looks like it got you good results. i bookmarked that baby!

i'm gonna stay from copper cages, but i'll be using foil for sure. thanks for the tips there.

regarding servo shielding, i wasn't so much thinking of blocking incoming, but rather outgoing. on the flip side, i recall a particular brand being more resistant to RFI and i think it was Futaba (verses JR). don't recall how well HiTec/MPX did. at least on a plane you have room to disperse stuff, where on a heli everything is in a big knot :)

i have a TwinStar on the way. don't know if that was any smarter of a choice than an EZ but i liked what appears to be a bulkier fuse (more room) and the 2nd engine (good for generating more RF:rolleyes:). never once flew a plank before except on a sim -- heli boy -- should be interesting! damn near did FPV with a heli many years ago with but had an equip. failure and never tried it again till now.
 
Top