|
|
Spockie-Tech
Site Admin
Joined: 31 May 2004
Posts: 3160
Location: Melbourne, Australia
|
[Warning, mega-tech ramble ahead - maybe I drank too much "V" earlier.. )
What is an "ElectroMagnetic Interference Dampener" ?
Well, actually there are lots of such devices out there, but I was wondering if you had any particular one in mind when you asked your question ?
Capacitors, Coils, Shielding, Tuned cavity's etc etc.
Supressing (and making) ElectroMagnetic waves is quite an artform, and its not really my field of specialty, but I have had to learn a bit about it along the way, and I can tell you its not a simple subject.
Anytime the electrical current flowing through a wire *changes* in strength, a "ripple" is created in the surrounding magnetic field. (much like throwing a rock into a pond). That ripple spreads out, and whenever it encounters another wire in the way, the magnetic ripple will convert back into a current flow in the second wire. Assuming you didnt intend for that to happen, then thats interference.
You dont really want the magnetic ripples created by your motors to be inducing unintentional currents in the tracks of your radio receiver (which is busy trying to listen for the deliberate but much weaker ripples created by the transmitter), so you have two choices..
Either try and stop those ripples from being made in the first place (supression), or contain them (shielding) in an area away from your interference-sensistive parts like your receiver.
The harder and faster the current level flowing through your motors and wires changes, the bigger the ripples they make are. Like throwing a bigger rock into the pond, or a smaller rock in faster, you get a bigger splash.
In electronics terms its called dI/dT - delta-I/Delta-T, or "change in current (I) over change in time (T). If you increase the current change (delta-I), or decrease the time that change happens in (delta-T) (making it happen faster), then you get a stronger electromagnetic ripple which has to be contained to prevent it from interfereing with things.
If you can slow-down the time over which the current changes, then you can reduce the ripple strength and consequently the radiated intereference gets less.
This is what the motor brush capacitors do.
Say you have your typical low-tech drill motor spinning at 20,000rpm (guesstimate figures). It has 3 segments on its commutator (where the brushes touch), so the current is being re-routed to a different motor coil 60,000 times a minute, or about 1,000 times per second (which =1Khz).
We cant do much about this, since its this field switching that makes the motor run, but its all happening at down at 1 Kilohertz, which is a long-away from the 36Megahertz (36 million cycles per second) that our radio receiver is trying listen out for.
So whats the problem ? Motors dont *just* put out waves at their switching frequency, because they have sparky brushes that are driving big coils of wire (inductors) that are belting about inside magnetic fields and currents are flying this way and that, so they actually radiate pretty darn well on a whole wide range of frequencies - right up into the High Frequency bands we're trying to use to talk to our 'bot with, which makes it like trying to talk quietly to someone in the middle of a rock concert.
If we could someone how stop (or at least slow down - increase delta-T) those high frequency current pulses, then we could quiten the ripples coming from the motor (and associated wires etc) down at the frequencies our radio system is trying to use.
Capacitors can be viewed as a sort of frequency-sensitive-resistor (with storage capabilities), In general they will "block" (show a high resistance to) a low frequency signal, while allow a higher frequency to pass through them (by showing a lower resistance to that frequency).
By mounting some capacitors (of the correct value to suit the frequencies you want to modify) across your motor brushes, you dont affect the low-frequency switching the motor needs to work (because the small capacitance value used looks like a high value resistor to the low frequencies, so it has little to no effect on the low frequency main motor currents)
But once the noise being generated by the brushes gets up into the high-frequency radio level, the capacitor starts to look like a pretty easy (low resistance) path to follow for that signal, so it passes through the capacitor and is given a way through which it can slowly fade away (rather than just being chopped off short as the commutator moves to the next position).
This "gradual discharge path" the high frequency signal earths through prevents the fast dI/dT change that creates ripples and hence prevents the motor from radiating (as much) interference at that frequency as it did without the caps.
Twisting the wires is done in the hope that the fields created by each wire will be opposite in polarity and hence will cancel each other out (like two waves going in opposite directions out of phase), since they are intertwined around each other.
Ferrite beads on the motor wires *might* help as well, since they add the idea that an inductor (an coil of wire usually with a ferrous core) will resist rapid changes in current and help to slow down any high frequency surges from making waves as well. To get the optimum effect, you would need to calculate what frequency you wanted to supress most strongly, and how many turns of wire, and what the size and permeability of the ferrite bead used should consequently be.
In practice however, unless you are an RF engineer who knows all this stuff of the top of your head, most of us just grab whatever sized ferrite bead looks about right and wrap a couple of turns of wire through it and hope for the best - and in most cases, anything is probably better than nothing, so it cant hurt given that they are cheap, light and easy to add.
Theres even more too it than that, you can get into building and tuning your radio antenna (Andrew-Totaly Recycled knows more about this than I do) in such a way that it is more inclined to pickup only the frequencies that you want it to, or is wave-polarised so its more sensitive to some direction waves than others and lots of other things, but I'm hovering on the edge of my radio knoweldge here now, and if everyone hasnt already run away in boredom or confusion by now, I'll be amazed..
I wish there was a simple "attach this interference dampenener and everything will be ok" device, but nothings ever quite that simple or easy..
You either get someone else who knows all about this sort of stuff to fix it for you (which is tricky, because bots are always being rebuilt which changes their interference charactersistics each time), or learn a bit about this guff to understand how to keep the radio working yourself..
Or if thats all too hard, do what I did and just buy a more expensive radio (PCM) that has digital error correction built into the radio link and makes 99% of your radio problems go away.. (at the cost of a few hundred $ for the radio)
Sorry, there isnt a cheap AND easy AND good answer to it (as usual).. you pay or you learn how to do it yourself, or you have to put up with it. Learning to fix it yourself is the most interesting if you have the time and inclination to do so
In summary..
The simplest and most effective thing you can do is to put small (10 - 100nanoFarad) ceramic capacitors across your motor brush connections, twist the motor wires together in a spiral, and maybe add a ferrite bead or two). That help in 80% of situations, beyond that it gets extremely complicated. _________________ Great minds discuss ideas. Average minds discuss events. Small minds discuss people
|
Sat May 28, 2005 4:25 pm |
|
|
|
|
|
|
|
Spockie-Tech
Site Admin
Joined: 31 May 2004
Posts: 3160
Location: Melbourne, Australia
|
actually you're half right there tim..
our Radio's *are* "Crystal-locked" sets where the crystals control the frequency they run at, rather than say a AM/FM music radio where the receiver is tuned to whichever station you want to listen to..
but, Crystal-lock sets still need to be "tuned", although its not tuning in the normal sense of altering the operation frequency..
The radio's contain a lot of filter networks, IF-Frequency oscillators and other clever radio-thingies (tm) that do need to be tuned to match the frequency the crystal wants the set to run at. They are there to help get rid of the "other" frequencies you dont want your radio to listen to and improve its sensitivity to the ones you *do* want it to listen to.
So if you open up your radio, you will find plenty of twiddly-bits that can be adjusted and if mis-adjusted will totally screw up your reception, despite the main oscillator being crystal-locked.
Thats why you arent supposed to just pop a 36Mhz crystal into a 29mhz radio. it might work OK at short range, but all the filters and things wont be tuned correctly for that frequency, so the radio Tx might still be leaking signal on frequencies its not supposed to, and the Rx might still be sensitive to signals on the 29Mhz frequency, even although the crystal is at 36mhz.
In fact, you're not even supposed to change from a "low-side" frequency crystal to a "high side" frequency (within the same band) without getting the radio re-tuned to suit. So changing 36.200 Mhz to 36.390 mhz is ok, but going from 36.150mhz to say 39.950mhz is a supposedly no-no without retuning of the filters etc.
I'll bet not many peope worry about it though unless they're aero-modellers who need to operate in close proximity to lots of other aero-modellers at the same time without unintended interference making someone elses plane crash.
If your radio is working properly or is relatively new then it probably doesnt need any tuning, and getting it retuned wont help, interference wise. The only time it would need re-tuning is if you alter the crystal frequency significantly, or it has required repair or some other alteration that changed the setting requirements for the twiddly-bits (or someone unqualified has been in there fiddling with them ) _________________ Great minds discuss ideas. Average minds discuss events. Small minds discuss people
Last edited by Spockie-Tech on Sun May 29, 2005 1:16 pm; edited 1 time in total
|
Sun May 29, 2005 1:08 pm |
|
|
|
Spockie-Tech
Site Admin
Joined: 31 May 2004
Posts: 3160
Location: Melbourne, Australia
|
When you buy a radio, it comes tuned for the frequency you order. if you want to change that frequency from one end of the band to another afterwards, then you should get it retuned to avoid interefering with other people on nearby frequencies.
Its not *as* important for bot-use, since we rarely run lots of radios in close proximity at the same time (except in a rumble maybe). However, its still a good thing to do, since one day coincidence will make sure that you will be standing next to someone who just happens to only be a channel or two away from your frequency..
Its easier (and a lot less $, tuning services arent cheap) to just buy crystals that are on the same "side" as the original ones though.
While we're on that note, I've just updated the frequencies table to show the new radio frequency we bought for Scoopy so if you're planning on getting a new radio, please check with the www.robowars.org/frequencies.html table first to avoid clashes. and if anyone else has any frequencies not listed as in-use that they are using, please let me know..
p.s. - If you want to order something from Tower Hobbies and can put the money into my bank account, I'll do the credit-card thing for you.. _________________ Great minds discuss ideas. Average minds discuss events. Small minds discuss people
|
Sun May 29, 2005 1:21 pm |
|
|
|
|
|
|
|
|