Connect with us

PWM shielding

Discussion in 'General Electronics Discussion' started by darren adcock, Apr 18, 2019.

  1. darren adcock

    darren adcock

    457
    33
    Sep 26, 2016
    Hi all, if I have a motor being controlled by PWM I need to shield the cable, so I can use a shielded twisted pair cable.

    My question is, this cable runs close to a JFET pre-amp for piezo's, terminating to a panel mount stereo jack socket (but possibly this isn't the best solution) what other type of panel mount socket is more suitable? I was looking first at BNC sockets and then Twinax (Twinax I couldn't find a panel mount solution), or maybe the stereo socket solution I am currently using is fine. In my mind I imagine a panel mount socket that has a ground shield that encompasses the two signal lines?

    Thanks in advance

    Darren
     
  2. Bluejets

    Bluejets

    3,374
    678
    Oct 5, 2014
    Why would you need to shield the motor cable, it's not an input.
     
  3. darren adcock

    darren adcock

    457
    33
    Sep 26, 2016
    I thought that the PWM signal being carried along the cable to the motor creates EMI, have i misunderstood?
     
  4. darren adcock

    darren adcock

    457
    33
    Sep 26, 2016
    I should explain further, in that the PWM signal is being carried from a module to a mechanical sculpture which is remote to the module, at the minute this is two meters distance.
     
  5. darren adcock

    darren adcock

    457
    33
    Sep 26, 2016
    [​IMG]

    I should have started with this, apologies. Do the +12v supply to the motor and the wire exiting the motor which connects to the Collector of the TIP120 need shielding if I'm running this externally from the circuit to connect to the motor via cable? At the minute I am running this via a Jack socket and using 1.5a two core cable with quarter inch jacks.
     
  6. Harald Kapp

    Harald Kapp Moderator Moderator

    9,043
    1,811
    Nov 17, 2011
    Depends on the sensitivity of the electronics around this circuit. Shielding is never a bad idea, but expensive. If everything works fine without shielding, go for it. If not, use shielded cables.

    2 m is a long antenna. Personally I'd use shielding in any case - if only to avoid interference with the neighbour's radio...
     
  7. darren adcock

    darren adcock

    457
    33
    Sep 26, 2016
    Ok I'll shield it as i bulk bought some cable a while ago when Maplin was closing down so it's no big hit in terms of cost. Thanks Harald.
     
  8. hevans1944

    hevans1944 Hop - AC8NS

    4,095
    1,944
    Jun 21, 2012
    As for connecting the motor shielded cable to your chassis... perhaps the KISS principle applies here. I would use solder lug type barrier terminal strips and attach crimp ring terminals to the external wiring, including the shield and its "drain" wire (if present). That makes it easy to move things around in the future if you need to use a longer cable or a different cable.
     
  9. darren adcock

    darren adcock

    457
    33
    Sep 26, 2016
    Ahhh ok, I hadn't thought about this thank you. I was going to use stereo jacks and use the ground from there, attached to the chassis, is the lug 'kiss' principle a better option? At the minute my housings are wooden, I am waiting on a delivery of copper tape to apply to the inside of the housings to then ground to. I know this probably isn't ideal, but have seen this being done to the insides of guitars and assume I can use this technique also?

    I haven't come across a 'drain' wire in my online reading, I'll do some reading.

    I realise also I don't know exactly what constitutes an 'antenna' so I'll do some reading about this.
     
    Last edited: Apr 20, 2019
  10. darren adcock

    darren adcock

    457
    33
    Sep 26, 2016
    I drew up the pre amp, filter and filter mix circuit so to make things clearer here also, the sections are from different sources on the internet that I have working together. In case there is any issues here also.
    [​IMG]
     
  11. hevans1944

    hevans1944 Hop - AC8NS

    4,095
    1,944
    Jun 21, 2012
    Back in days of old, shielding consisted of braided copper wires woven in a cylinder around the insulated conductors being shielded, quite often just one conductor in which case it was called coaxial cable. Then for low-level instrumentation signals, Belden invented Beldfoil shielding, which consisted of aluminum film bonded to a thin plastic film and the two bonded layers wrapped with interlocked edges around the insulated conductors being shielded. This worked really well, and it provided a 100% shield coverage without any gaps in the shield. For added strength, this configuration was sometimes overwrapped with a conventional woven copper shield.

    However, it was difficult to make any sort of reliable electrical connection to just the aluminum part of the shield although many crimping techniques quickly evolved. To solve the electrical connection problem without resort to crimping and special tooling, Belden laid a so-called, stranded copper, "drain" wire inside the aluminum foil shield along its entire length in the shielded cable. Contact area was sufficient that no further bonding between aluminum foil and copper drain wire was necessary or required. The end user then just crimped or soldered to the drain wire to complete the shield wiring.

    Any exposed, un-shielded conductor, can serve as an antenna at some unspecified frequency. Add some math and electromagnetic field theory to make a deliberate antenna design suitable for some specific frequency or range of frequencies. Or use the math and theory to predict how your "accidental" antenna will perform in receiving real-world interfering radiation from its environment or transmitting real-world interfering radiation to its environment.

    In electronics, antennas are the real-world connection between electromagnetic signals and electromagnetic radiation.
     
    darren adcock likes this.
  12. darren adcock

    darren adcock

    457
    33
    Sep 26, 2016
    Thanks Hevans!

    I see on google there is various circuits on making an EMF detector and/or measuring but they are inter-dispersed among wacky 'ghost detectors' etc. Are any of these circuits and thus tools useful?
     
  13. hevans1944

    hevans1944 Hop - AC8NS

    4,095
    1,944
    Jun 21, 2012
    I am sure they are useful and profitable to the purveyors of such things. For what purpose do you reckon you would purchase and use one?

    Years ago, when microwave ovens for home use were new on the market, some folks promulgated the theory that some of the ovens leaked "dangerous microwave radiation" into the nearby environment. To safeguard against exposure to this "dangerous" radiation, instruments were sold that could allegedly measure any "leakage" that occurred and warn the end user. I don't know what remedy, if any, was offered to protect against these dangerous emissions... perhaps a specially designed tin-foil hat or chain-mail vest or lead underwear was offered with the usual "money back" guarantee? Cash or money-orders only! And please enclose a quarter to cover the weight of the money order. ...with apologies to Brother Dave Gardner (b. June 11, 1926 - d. September 22, 1983) may he RIP.
     
  14. darren adcock

    darren adcock

    457
    33
    Sep 26, 2016
    hahahahha!

    I didn't want to purchase one, and mentioning the ghost detectors was tongue in cheek!

    Well maybe my naivety is embarrassing me here, oh well, I simply thought if i could take a measurement then I could use this to mitigate and inform whether I need to shield in the future. But then also being able to take supposed measurement is just ignoring some theory and good practice techniques which isn't really gonna help me learn. And as Harald has pointed out 'shielding is never a bad idea' just is a bit more costly.
     
  15. hevans1944

    hevans1944 Hop - AC8NS

    4,095
    1,944
    Jun 21, 2012
    Shielding is a complicated subject, much more complicated than just a cursory examination will reveal. All shielding is based on a simple principle: no electric field can exist inside a closed, conducting surface. The Faraday Cage is the most basic example. A Van de Graaf electrostatic generator (used for the first atomic particle accelerator) depends on it. All shielding is frequency dependent, some more so than others. It is impossible, because of the energies involved, to shield against the penetration of cosmic rays. It is fairly easy to shield at very low frequencies approaching DC.

    At my first real job, after my term of enlistment in the U.S. Air Force, was as an entry-level electronics technician at the University of Dayton Research Institute. When I was hired, the University was in the process of rapid expansion (which continues to this day!), having started construction on a new five-story engineering and research building. The Electronics Laboratory, where I was soon to work, was located on the top floor. We shared about one third of the available floor space with a materials engineering laboratory. Administrators and project managers occupied the remaining third with windowed offices surrounding the two laboratories. It was a pretty cozy arrangement, and I was destined to work there full-time while attending school part-time until a year after I graduated with a Bachelor of Electrical Engineering (BEE) degree, awarded by UD in 1978.

    When we moved into the new building, one of the first things I noticed was a large "room" built inside one corner of the laboratory. It had walls with metal laminations on the outside and a large metal door with "finger contacts" that electrical sealed the door to the room when the door was closed. There was an air vent with a honey-comb metal structure inside that made it impossible to see directly through it because of internal reflections. I was told the vent stopped all radiation from DC to microwaves. There were also line filters that brought mains power into this shielded room, which I was told was a Faraday Cage.

    I knew what a Faraday Cage was. We had one in our USAF Armament and Electronics maintenance shop, a simple wood construct with copper screening tacked to the wood and the seams soldered together. Nothing as fancy as the UDRI shielded room, but apparently "gud enuf" for guv'mint work. The UDRI Faraday Cage was intended for precision metrology work involving pico-volt and pico-ampere signals. Problem was, none of knew anything about metrology. So the Faraday Cage sat mostly idle while I was there. I think I used it for a project just once, and then only briefly because I soon discovered how to adequately shield my sensitive circuits.

    Years later, at another employer, we had a contract to develop something for the Navy that required EMF (electromagnetic field) testing. Their requirements were very specific and we didn't have the facilities or instrumentation to perform the tests the Navy wanted to do. It took awhile, but eventually the Navy said "come on down" and use their facility. The Navy was mainly interested in microwave susceptibility over a very specific range of wave lengths (frequencies), but they also insisted on very precise measurements of the fields inside our equipment. This is where the rubber meets the road: we were trying to shield the inside of our equipment from the very thing we were trying to measure. The better we did our job, the harder it was to prove to the Navy that we had done our job. On the plus side, the Navy could crank up the output of their EMF generator until our probes inside the equipment could actually "see" something. We never did find out what the Navy really used their test stands for (they had several) but it made the UDRI Faraday Cage look like a tinker toy in comparison. Who would have thought that someone would want to put a high-power microwave transmitter and antenna inside a Faraday Cage?

    You are on the right track in wanting to take measurements, but you need to temper that with the why and the how of it. There is a saying, possibly true, that "if it ain't broke, don't try to fix it." A good engineering design, which you learn how to do with practice, routinely employs good shielding principles, sometimes even when they aren't needed. Another saying, possibly true, "better to have it and not need it, than to need it and not have it." You can shield with adhesive copper tape. That's exactly what it's made for, among other uses. Whether it is necessary or not depends on your specific application.
     
    darren adcock likes this.
  16. darren adcock

    darren adcock

    457
    33
    Sep 26, 2016
    Thanks Hevans, that's a generous reply! Lot's to read about and learn.
     
  17. FuZZ1L0G1C

    FuZZ1L0G1C

    363
    112
    Mar 25, 2014
    Velocity of light & electricity c=300 000 000 m/sec.
    A 2 meter length of wire has a full-wave resonance of 300E6/2 = 150E6 Hz.
    A single length of wire resonates strongest at quarter wavelength, so peak resonance radiates at 150/4 = 37.5E6 or 37.5 MHz.
    Twisted pair effectively shortens each "segment" of radiating element, raising the resonance, and nulling the radiated (or received) waveform by 'inverting' the polarity every twist of cable.

    Sig==X====X====X====X====X====X====X====X====X====X==
    Gnd Source o-----------------------------------------------------------------------------> (shield)

    Add:
    The 37.5 MHz doesn't necessarily refer only to the maximum controlled PWM frequency, as the square-wave transition from "hi" to "lo" can create a resonant pulse if 1/f or 1/37.5E6 = 26.66 ns duration.
     
    Last edited: Jun 10, 2019
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-