Connect with us

Transmit and receive parallel data through radio link

Discussion in 'General Electronics Discussion' started by Anon_LG, Aug 12, 2014.

  1. Anon_LG

    Anon_LG

    453
    117
    Jun 24, 2014
    I need an encoder and a decoder with over 4 inputs/outputs respectively that feeds out of 1 output/input respectively (for a transmitter and receiver) from a website that either has free delivery on orders over £15 or charges under £5 for delivery to the UK. I have found this: http://www.digikey.co.uk/product-detail/en/LICAL-ENC-MS001/LICAL-ENC-MS001-ND/701458 and its decoder counterpart, these would be great (having 8 inputs/outputs) however Digikey charges £15 order charge unless I spend over £50, so that's a no and Mouser I can not find the ordering charges for.
     
  2. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,267
    Nov 28, 2011
  3. Anon_LG

    Anon_LG

    453
    117
    Jun 24, 2014
    Yeah, the charges, I thought Farnell had a £50 charge on orders for non-business account holders. I am sure I read this on the site once. If I where to the CD4051B how would you recommend I get the 3 control bits in sync, I could use a 3-bit binary ripple counter, but how will I synchronise the pulses?
     
  4. BobK

    BobK

    7,645
    1,663
    Jan 5, 2010
    It is not clear to me what you are trying to do. If what you want is a xmit/rcvr pair that can encode 4 bits, i.e. have 4 channels that you can control on / off separately, there are such units available on Ebay really cheap with the encoder and decoder built in.

    Bob
     
  5. BobK

    BobK

    7,645
    1,663
    Jan 5, 2010
    What is it with you Brits. Here in the US, Mouser or Digikey have no minimum and the shipping charges are about $5.

    Don't you guys have illegal immigrants to process the orders cheaply? (At least I think that is why Mouser is located in El Paso.)

    Bob
     
  6. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,267
    Nov 28, 2011
    What do you want to do? You need to describe the project.
     
  7. Anon_LG

    Anon_LG

    453
    117
    Jun 24, 2014
    The multiplexer/demultiplexer uses 3 control bits, counting in binary to select which channel bit is outputed/inputed to the output/input, so c=0, b=0, a=0 it will select C0, c=0, b=0, a=1 it will select C1, c=0, b=1, a=0 it will select C2... and so on, I will be inputting 8 (or maybe a few less) separate bits of information into the multiplexer, these bits then need to output, one after the other (this could be achieved by counting pulses from an astable multivibrator using a mod-8 binary ripple counter), the output from the multiplexer will be fed into a transmitter (the transmitter will simply transmit 1 or 0 depending on the input), the transmitter output will feed into the input of the demultiplexer, as with the multiplexer 3 control bits are used, I could therefore use an astable multivibrator pulses to count on a mod-8 binary ripple counter to cycle through these bits, the channels will then output into bistable multivibrators to store the information. The only problem is the pulses need to be synchronised, is there any way I could use the first channel, C0, which requires c=0, b=0, a=0 to re-align the pulses? The astable multivibrator could easily keep the pulses synchronised for 8 or so pulses but I can not rely on any more because of tolerances in the capacitors. I hope I have described it in enough detail.
     
  8. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,267
    Nov 28, 2011
    What kind of transmitter and receiver are you using?

    There are various ways of approaching that requirement. One is to use a "dumb UART" at each end. such as the ancient Harris/Intersil 6402. But I just checked availability for these, and Digikey doesn't have them and they're USD 30 each from Mouser! So scratch that idea.

    Other ways would involve self-clocking encoding schemes like MFM (https://en.wikipedia.org/wiki/Modified_Frequency_Modulation) and NRZI (https://en.wikipedia.org/wiki/NRZI#Non-return-to-zero_inverted) but these need an encoder and a decoder, and some way to synchronise the bytes.

    Edit: Also check out https://en.wikipedia.org/wiki/Self-clocking_signal and https://en.wikipedia.org/wiki/Manchester_code

    How about describing what you want to do, not necessarily how you want to do it, and post links to the transmitter and receiver.
     
  9. Anon_LG

    Anon_LG

    453
    117
    Jun 24, 2014
    The listing on Ebay, where I bought it from, has ended and there is something wrong with seeedstudio where the information on the transmitter/receiver is. Basically I want to be able to input 8 inputs, these are then sent over a transmitter, received by a receiver and then decoded to give the original 8 bits. (and as I said above, the 8 output bits would have to be fed into a bistable to store the data as the output bits would be pulsed every 8th interval)
     
  10. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,267
    Nov 28, 2011
    And the transmitter just takes a digital data stream in, and that stream is reproduced at the receiver?

    Do you know the maximum data rate? It's not important; I'm just interested. Also is there any kind of clocking at either the transmitter or the receiver?

    And you want to broadcast continuously, or periodically, so any change at the transmitter's inputs will be reflected at the receiver within a certain period of time? What update rate do you want?

    Do you have any experience with microcontrollers? Or do you want a discrete solution?

    Do you need eight signals or would 7 or 6 be enough?

    What power supply voltages do you have at the transmitter and at the receiver?
     
  11. Anon_LG

    Anon_LG

    453
    117
    Jun 24, 2014
    1st point: Yes, that is exactly what I want to achieve

    2nd point: errr... I can not find the max baud rate

    3rd point: the data rate can be really slow as in as slow as in 1 update for each channel every second. Any update speed faster than or roughly equal to this would be fine, its not for a controller at the moment, just to receive information from various sensors outside (e.g. thermistors, LDR and humidity sensor, which I will be converting the electronic levels to digital first, eventually I want to provide more accurate information from these sensors* and so preferably I need a known time each of the channels is broadcasting. Continuously or periodically? which do you think is simpler to achieve?

    4th point: I would like a discrete solution, I do not have access to, nor have the experience with microcontrollers.

    5th point: 6 or 7 would be fine

    The site is back up: http://www.seeedstudio.com/wiki/433Mhz_RF_link_kit
    • Transmitter Input Voltage: 3-12V (high voltage = more transmitting power)
    • Receiver Input Voltage : 3.3-6V (high voltage = more receiving power)
    I will probably be powering the transmitter off 4.8 Volts (4*AA rechargables) and the receiver off 5 Volts.


    * by re-using the initial method of encoding within a channel to give a binary number which I will than convert to analogue and use in a LED driver to light LEDs above a certain level of voltage to give the temperature etc
     
  12. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,267
    Nov 28, 2011
    OK cool. Something based on a 74HC165 and a 74HC595 transmitting a start bit, some data bits, then an idle gap, would probably be pretty good. I'll draw something up in the next few days, OK?
     
  13. Anon_LG

    Anon_LG

    453
    117
    Jun 24, 2014
    OK, Thanks Kris!
     
  14. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,267
    Nov 28, 2011
    Don't worry, I haven't forgotten you! I'm still working on the design.
     
    dashy1981 likes this.
  15. Anon_LG

    Anon_LG

    453
    117
    Jun 24, 2014
    Thanks, me too, I have decided I need to use a quartz oscillator (then dividers, counters) at each end in my design as I do not think the capacitors are made with accurate values to give a reliable synchronised 8 pulses in the astable. Also in my design the main channels will be used to cycle through 8 encoders (I will most likely be using the RF803 E/D chipset: http://uk.farnell.com/rf-solutions/rf803e/ic-encoder-rf-8dip/dp/1655508), as the channels are not encoded and so would be susceptible to interference. I would really like to see what you come up with.
     
    Last edited: Aug 19, 2014
  16. Anon_LG

    Anon_LG

    453
    117
    Jun 24, 2014
    This is the basic layout of the transmitter circuit (no power connections or detail). The receiver will look similar to this except it will not need the input labelled scan as that signal will be received through channel 0, the output on the multiplexer (which at the receiver end will be a demultiplexer) will be an input and on the receiving end there will be bistables to store the outputs. Can the CMOS IC's handle (reliably) the frequencies involved? U3 will have to take varying frequencies (on the l.s.b leg about 30 kilohertz if the oscillator is around 1 Megahertz), U1 will be fine if U3 is fine. But U2 is going to be taking the frequency of the oscillator, if this is a problem then I could use several other ways to replace this section. U3 was originally a 3 input NOR gate taking inputs solely from the control bits but when I thought through the circuit in order this would cause the circuit to instantly reset, I did not want to use a 3 input AND gate as channel 7 would be unusable, so I I decided to use the 4th JK to give a higher M.s.b causing the circuit to go through 2 scans before resetting.
    Transmitter 8 channel multiplexer (23 sub-channel).jpg
    If someone experienced with RF circuits could look at this and tell me whether they think it would work or whether there are any bugs that would be appreciated.
     
  17. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,267
    Nov 28, 2011
    Yes, likewise.
    The information you want to transmit is digital, right? Seven digital signals?
     
  18. BobK

    BobK

    7,645
    1,663
    Jan 5, 2010
    If I were doing this, I would use a micro on each end and a protocol with enough redundancy that corrupted messages could be rejected. I.e. you transmit a multi-bit message that includes a code, the data and a checksum. Unless the code and checksum are correct, the data is rejected. The status is continually transmitted so that it will eventually (hopefully quickly) will catch up. You could also then build in a default action if no transmission succeeded within a time limit. At that point you would set the output into safety mode.

    Anything less than that is likely to have frequent errors due to noise and interference.

    What type of transmitter are you using, 433 MHz?

    Bob
     
  19. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,267
    Nov 28, 2011
    Me too. But he doesn't want to use micros. And I think the information is non-critical and brief disturbances are acceptable. Is that right Lavaguava?
     
  20. BobK

    BobK

    7,645
    1,663
    Jan 5, 2010
    Well, then I would suggest the PT2262 and PT2272 which basically do what I said without using a micro. Only problem is that they can only do 6 bits of data. You could use 2 with different address bits to extend the number of bits and then alternate the transmitter between the two.

    http://www.adafruit.com/datasheets/PT2262.pdf

    Bob
     
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

-