power line communication

Discussion in 'General Electronics Chat' started by vineethbs, Feb 3, 2005.

Thread Status:
Not open for further replies.
  1. vineethbs

    Thread Starter Well-Known Member

    Nov 14, 2004
    56
    0
    i am doing a power line communication proj in my college , right now i am working on the link layer for that . am using FSK to send a bitstream from one point to another , can somebody enlighten me on these things

    1 . Coupling transformers etc
    2 . lm565 pll spice model
    3 . clock recovery from FSK

    and plz tell me almost every other thing that u can tell me abt this

    Thanx in advance
     
  2. vineetcec

    New Member

    Feb 12, 2005
    5
    0
    hey even i'm doing a project on powerline communication.but i've just started doing it.started with the FSK modulator.....got the same problem coupling circuit.i'll let u know if i get any good circuit for that.........expecting the same from u.need help for the rest too.by the way which college are you from?wer?me a 3rd year engg student from kerala.....

     
  3. torpedopudding

    Member

    Feb 6, 2005
    28
    0
    I have never tried this although it has been done for many years, including wireless intercoms maybe back in the '60's.

    I have seen such coupling circuits although it has been a long time.

    High confidence:

    1) Your line driver should have its impedance more-or-less matched to the high frequency impedance of the power line. Unfortunately I do not know what that impedance is and it might vary according to where you are. So for example if you have an AC coupled output amplifier that runs from 12V and can drive 1A, that will be +/- 6V AC coupled and about 6 ohm impedance. Remember that transformers match impedance according to the square of the turns ratio. But the number of turns, a reasonable core selection, etc. are other items that need to be worked out. I suggest using EE-cores for ease of assembly and low cost.

    2) You will need to block AC line from the amplifier I think that a combination of series AC coupling capacitor (choose a suitable value) and an impedance matching transformer will both help in that regard. The transformer will also provide DC isolation between the amplifier output and the power line.

    Medium confidence:

    3) I guess that frequencies in the 100-200 KHz range sound reasonable. For that, a ferrite core impedance matching transformer would work well. It is not physically hard to make such a transformer, but I think you had better bone up on that as it is too much of a lesson for me here. Some websites such as ferroxcube might help.

    4) I would try driver power levels of a few watts.

    I am open to other questions.....
     
  4. vineethbs

    Thread Starter Well-Known Member

    Nov 14, 2004
    56
    0
    is this somebody pulling my leg ? :angry:

    too many coincidences
    same name,doing the same thing hmmm

    sorry if not

    neways me having some exams now

    got one question
    what kind of line coupling transformers were used in telephones ?
     
  5. vineetcec

    New Member

    Feb 12, 2005
    5
    0
    thanx a lot for the reply......hopefully it will help.powerline communication involves alot of noise,rite?is it actually really difficult to work?i'm planning to do a model of this plc by drawing a wire from a plug point to act as a power line.my project is is basically a digital energy meter with automatic bill reading using powerline communication.


     
  6. dragan733

    Senior Member

    Dec 12, 2004
    152
    0
    :D
     
  7. torpedopudding

    Member

    Feb 6, 2005
    28
    0

    I do not understand why you think I am pulling your leg. I do like to kid people but I like to think I make it pretty obvious!

    I also do not know what is the coincidence.

    By the way power lines are now being used for internet connections in some locations, the same basic idea.

    Telephone line transformers are designed for audio frequencies 100 Hz to 4 KHz while we have been discussing frequencies over 100 KHz. I do not think telephone line transformers would work well for this purpose; also don't forget you wanna get the right impedance match. BTW, mechanical impedance matching is what an automobile transmission is all about.......
     
  8. n9xv

    Senior Member

    Jan 18, 2005
    329
    1
    Telephone line transformers are basically a 600-Ohm : 600-Ohm isolation transformer. Transformer coupling should be alot les critical and more forgiving than capacitive coupling. The transformer can deal with a wider range of impedances.
     
  9. dragan733

    Senior Member

    Dec 12, 2004
    152
    0
    torpedopudding, you missunderstood the member 'vineethbs'. His replica was sent to the 'twin' member 'vineetcec', and not to you. A name 'vineetcec' who said to have also same project and same problem about asking that. Time for a little grin.
     
  10. vineethbs

    Thread Starter Well-Known Member

    Nov 14, 2004
    56
    0
    hey am sorry guys it was a little misunderstanding , the message was meant for my almost namesake,not to torpedopudding.
    thanx to dragan733 for clearing it up.
    i thot it was somebody from my college ,
    sorry vineetcec !

    now the important thing
    what abt switching transformers (torroids) used in SMPS and all? can they be used for coupling etc,i hear they r expensive , tell me more abt them

    thanx in advance
     
  11. torpedopudding

    Member

    Feb 6, 2005
    28
    0
    I think of myself as experienced with regard to circuit design. I also think of myself as inexperienced with regard to the workings of forums. Could we call it a fivum and then I make up the rules? :)

    For someone designing or experimenting, the magic is to get your own cores (and bobbins if you are doing core shapes other than toroids) and wind your own.

    There is not much available "standard" for ferrite core transformers, so whether you wind it yourself or someone winds it for you it is probably custom. In home/school projects, no business will want to do it for you at prices you will want to pay, so you will end up doing it yourself.

    With some basic math, patience, and some basic manual dexterity it is not hard, unless you are trying to make something competitive. Again, Ferroxcube used to have some very practical info regarding design of ferrite transformers, and something ought to be available on the web from them or Magnetics, Inc. or some others.

    The most important formula for transformers is N*phi=N*B*A=Vdt
    N=# turns (toins if you are from New Jersey)
    Phi = Flux in Webers
    B= Flux density in Tesla = Webers per square meter
    A = Core area in square meters (will be a small number but what the heck.)

    Typical numbers might be:
    a 0.1 square cm cross section core would have an area of 1*10e-5 m^2

    a 12v-0v square wave at 200 Khz (mid point is 6V) will give you a V*dt of 15e-6

    For 100 - 200 KHz power applications, target a maximum flux density of 0.1 Tesla, depending upon your core material.

    In the example above, you would need 15 turns on this input winding.

    And what do you get? Another day older and deeper in debt............. :)

    There are no other fudge factors to remember if you keep everything in MKS system.

    Regarding impedance matching, the transformer will not always care so much but believe me the driver does! It can make all the difference between getting the desired power into the line, or a small fraction of it, or smoking your driver....
     
  12. torpedopudding

    Member

    Feb 6, 2005
    28
    0
    Actually I am thinking that if you do FM or FSK demod of a suitable frequency such as 200 KHz you might avoid a lot of noise.

    If you live in USA, I can tell you that powerline communication may work well within your house on anything plugged into a given side of the 220 VAC power line. Normally you have 110 VAC, Neutral and 110 VAC opposite phase. An electric stove or clothesdryer will plug into the two 110s whose differential is 220 VAC. Your 110 VAC outlets connect to neutral and one 110 VAC phase.

    The problem in getting far beyond your house is the distribution transformer. The power company does not transmit 110/220 VAC long distances; it would be too lossy. Somewhere near your house is a distribution transformer with 440 VAC (??) going into it. I do not think that transformer will pass along your 200 KHz information unless it is specifically designed to do so.......

    The 440 VAC comes from a bigger transformer upstream with higher voltage feeding it, and so on. The huge power transmission line towers support lines with voltage of (very roughly) 100 KV. I do remember about 30 years ago they were building a 750 KV DC line across MN.
     
  13. vineetcec

    New Member

    Feb 12, 2005
    5
    0
    thanx to dragon733 and torpedopudding............i havent actually started workiing on the couping ckt....didnt even get the fsk mod correctly.i'm using XR2206
     
  14. vineethbs

    Thread Starter Well-Known Member

    Nov 14, 2004
    56
    0
    thanx for the replies ,one more question , if u can monitor the noise level on the powerline before u send a symbol and then change the powerlevel of transmission according to that , the noise performance shud increase , r8 ? but is something like that really necessary .

    and vineetcec can u tell me from where u got xr2206, i was looking arnd for that and no shops seem to hav that in stock

    thanx in advance
     
  15. vineetcec

    New Member

    Feb 12, 2005
    5
    0
    i bought the xr2206 from kochi,kerala....rony electronics thats the shops name
     
  16. vineethbs

    Thread Starter Well-Known Member

    Nov 14, 2004
    56
    0
    hmm seems like they hav everything , me from the south end man :)
    also can someone plz tell me abt clock recovery from fsk

    thanx in advance
     
  17. torpedopudding

    Member

    Feb 6, 2005
    28
    0
    Wow! I wanna remain anonymous. But I cannot help but comment that I was working with the XR2206 in the mid '70s.
     
  18. vineetcec

    New Member

    Feb 12, 2005
    5
    0
    hey ...i dont why....but the circuit with xr2206 didnt work......i tried everything no use.i tried the exact circuit in the datasheet .can you help me on this ?
    then i got 1 more question.....if i use an IFT in the coupling ckt will it work properly?
     
  19. eoinf

    New Member

    Mar 16, 2005
    1
    0
    ive tried to build the circuit in the xr2206 datasheet too and its not working for me
    did you get yours working
     
  20. Ahmed

    Member

    Mar 16, 2005
    17
    0
    Hi vineethbs

    me too, power line carrier communication is a main part of my project which is "automated home system using phone line and power line", i have finished the phone line part but still working in power line.

    about FSK:

    i used in modulation another circuit with XR2206 and in Demodulation another IC from the same family XR2211, the result until now is not accurate.

    so iam still working .... B)
     
Loading...
Thread Status:
Not open for further replies.