Multiplexer IC Question

Discussion in 'General Electronics Chat' started by NoSkill, Sep 2, 2011.

  1. NoSkill

    Thread Starter Member

    Jan 18, 2007
    13
    0
    Hi All,


    I am having some trouble working with the following 24V 8chan MUX from Intersil. The datasheet is here

    Let me explain the situation.

    For the above IC I have connected the A0 A1 A2 (signal select) lines and the EN and GND lines from my micro to the IC. The A0-A2 signals are pulsed periodically by the micro while the EN is held high at 3.3V.

    I have one 24 volt power supply sharing the SAME ground as the micro (connected the negative terminal on the PSU to the gnd of the micro) acting as a 24V input to the mux.

    Now this is probably where I am doing something stupid: I have connected the mux's V- line to ground and the V+ line to the SAME 24V power supply.

    So what I am trying to do is have my power supply, power the IC through the V-,V+ lines AND create the signal that should appear on the D output of the MUX.

    Now this doesn't work. In fact it fails horribly when I power the whole thing up by shorting my power supply. Thankfully the current clamp kicks in. Could someone explain to me what I am doing wrong?

    Thank you very much for your time.
     
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Sounds like you're tying the IC's GND and V- together. Not good. Hope you haven't fried the IC.

    Instead of grounding the V-, you need to supply V+ with +12v and V- with -12v. GND should be (roughly) halfway between +12v and -12v. Your logic needs to go from GND (0v) to at least +3v.

    You haven't told us much about the 24v supply except that it outputs 24v.
    Is it a "floating" supply, or is it's 24v return (V-) connected to earth ground internally via the power cords' earth ground?
    Better yet, do you have a manual or a manufacturer and part number for the supply? If not, what are its' available connections?

    If you don't have documentation on the supply, the easiest/cheapest way I can think of offhand to power your setup is to use an ATX or ATXPLUS12 computer power supply converted to a bench supply. It will have +12 and -12 to power your MUX, with GND halfway between; +3.3v and +5v available to power your logic/PIC/other components. Googling "ATX bench supply" will give you lots of pages on converting one.
    If you don't have a spare ATX supply, MPJA has 150W units on sale for ~$6+shipping:
    http://www.mpja.com/prodinfo.asp?number=18034+PS

    Otherwise, you haven't explained how you're powering your PIC, or if the 24v and 3v-5v supply grounds can be isolated from each other. If your 24v supply is "floating", there is a way to create a "virtual ground" using an opamp, resistors and caps.

    Do you have any spare opamps on hand? If so, list what they are.
    Do you happen to have some spare NPN and PNP transistors on hand? If so, say what they are as well.
     
  3. NoSkill

    Thread Starter Member

    Jan 18, 2007
    13
    0
    Thanks SgtWookie, as always a wealth of information, thank you for your response.

    For the micro I am powering it with an LM317 that is currently converting the supply from the 24V PSU down to 3.6V.

    The PSU that I am currently using is a GWInstek GPS-3030 with a negative, green gnd and positive terminal.

    I would not like to use an ATX supply since I need to find a way to power the IC from the same supply that powers the LM317. As such I think the virtual ground that you had in mind would work best.

    I do have an op amp on hand. an RC4558P from TI datasheet here. The only transistor I currently have on hand is a measly STS-9013 datasheet here . Please feel free to recommend some more appropriate parts.

    Otherwise I have a myriad of resistors caps and diodes.

    I have been reading the wikipedia entry on how to do a virtual ground and it seems like an inverting amplified is all I really need but I am not sure how to power the opamp since it also has V- and V+ rails! From -18V to + 18V.

    Any ideas?
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Ugh, that RC4558P is electrically equivalent to the 741. Might as well be a 1458. :rolleyes: I don't know why TI bothered giving it another part number. It's a tad faster than the 741, but then has all of the other drawbacks.

    Now that I'm done ranting... ;) Are you in the States? If so, do you have a Radio Shack store nearby? They usually carry TL082 opamps, which are much better than the 741 - with the exception that they can't "see" within about 3v of the negative rail. They also carry 2N2222 and 2N2907 complementary transistors, so that we can basically make a "power opamp" out of a standard one. This will enable you to have a stable virtual ground. Then you can connect the LM317 from the +in to the virtual ground.
     
  5. NoSkill

    Thread Starter Member

    Jan 18, 2007
    13
    0
    Unfortunately I am not located in the States, but I have ordered the parts.
    Quick question about the virtual ground. How does it affect the LM317 calculations for the thermal coefficient?

    I am pulling quite a bit of current from the LM317, estimated at about 500mA max. My calculations for heat sink necessity were based off a 24V input to the LM317. If I now have a virtual ground at +12V will I need to re-calculate the hit on the LM317 since that is now its voltage input?

    Thanks again.
     
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    OK, NOW I need to see a schematic of how you have things connected currently.

    You said a +24v supply previously; I didn't assume that it was a home-made supply.

    I need to know more about the power requirements. If you don't post an accurate and complete schematic, you may not get accurate and complete information, which can lead to smoke.
     
  7. NoSkill

    Thread Starter Member

    Jan 18, 2007
    13
    0
    The 24V constant supply is just the input to the LM317, I can draw as much current as I want out of it. The LM317 brings it down to 3.3V which supplies my micro and other peripherals where I draw a maximum of 500mA.

    Since I didn't have a virtual ground before, the thermal calculations for the LM317 at a 24V input, 3.3V output at 500mA were acceptable and did not need me to use a heatsink.

    What I was wondering before, is that if I create a virtual ground using an op-amp, does the LM317 now see 12V as its input which will, in effect change my thermal calculations.
     
Loading...