Latching On with Transistors: possible?

Discussion in 'General Electronics Chat' started by rhyno, Apr 20, 2008.

  1. rhyno

    Thread Starter New Member

    Apr 20, 2008
    5
    0
    Hi All, I'm new to these boards, but i plan on sticking around and helping where possible, so please go easy on me to begin with. I have what i think is a simple problem to solve. I have a background as an electrician and have designed many control circuits for electrical projects which have a 24v control circuit and 240+ power circuit. The standard system i use to turn it on is a momentary NO switch through a relay, holding the coil in, and power on, one the momentary switch has been pressed. Turning off via a momentary NC switch cutting power to the relay coil. This works fine and hasn't caused any problems. A schematic is shown below. SW1 turns on the circut, SW2 off. when coil is energised, contacts 1 and 3 are joined applying constant power to coil.
    [​IMG]
    What i want to do is make this same circuit using transistors as they are considerably cheaper and smaller than relays. I had a theory on a simple circuit but upon breadboarding it i didn't work; my understanding of transistors is limited.
    [​IMG]
    Obviously this circuit can only be used in the control circuits and relays will be needed to switch any 240v power circuits. Any help would be greatly appreciated.

    Thanks,
    Ryan
     
  2. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,699
    907
    When you close the switch, you will be applying the full battery voltage across the LED. It will probably burn up. Also, since the transistor base is grounded to the emitter, it will never get high enough to turn on the transistor.

    Your circuit might be somewhat fixable with some resistors, but before we do that, let's consider what I think you are proposing to do. I am familiar with the relay-based circuit you want to replace. In that circuit, the coil stays energized while the "tool" is operating. However, the heavy current for the tool is being carried by the relay contacts. With the type of transistor circuit you are suggesting, that current would have to be handled by the transistor. Of course, transistors can handle many amps, and solid state relays do just that, but be aware it will not be a simple one transistor circuit.

    Are you ready to take that on?

    John
     
  3. rhyno

    Thread Starter New Member

    Apr 20, 2008
    5
    0
    Thanks for your reply,

    i should have mentioned that the circuit is a small part of a larger circuit and what i have drawn is to show where the load, etc is going; its not actually powering up an LED, it will be kicking off a timing circuit, amongst other things, all low current draw and low power. At the points of the circuits that do need to switch on high powered devices (heating elements, pumps, etc) i will be using proper relays.
    The actual circuit is to automate my home biodiesel plant. At the moment i am flicking on/off switches, opening valves at particular times and i'm trying to eliminate as much of this as possible.
     
  4. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    How are you dealing with the dread glycerine problem?

    If the conversion process has measurable points where this process happens, and then that one, a big but worthwhile step is to consider automating it with a microcontroller. It's a steep learning curve, but you'll probably get there later if not sooner.
     
  5. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,699
    907
    Let me re-phrase to be sure I understand. You want a circuit that will keep itself turned on when triggered with a momentary switch (just like the motor control circuits do) and will not turn off until press an "off" switch (a momentary N.C. switch). You want to keep it simple and use just a few transistors at most (there are logic devices that effectively do that called flip-flops). Finally, the maximum current will be??? I'm guessing about 200 mA. Is that OK? If you can get by with even lower current, that is better.

    Do you know what an SCR is? It is like a transistor that you can trigger on, and it stays on as long as the current flows. There is quite a simple circuit that does what you want to do with an SCR and two resistors. It's an old Radio Shack (Forrest Mims) project. I'll re-draw and post it here, if you think that will do what you want.

    John
     
  6. rhyno

    Thread Starter New Member

    Apr 20, 2008
    5
    0
    Which Glycerine problem is that:

    A) Getting rid of it? We are distilling it to remove the methanol from it which is reused, and then the pure (97%) glyc is made into soap and sold at a local market.

    B) Having the process automatically know when the layer changes from Glyc to Bio? As we are opening and closing valves by hand at the moment i haven't yet experimented with controlling it. I thought of using some sort of optical device as the bio is quite translucent whereas the glyc is deep brown. Otherwise i can use a float switch as the volume of glyc coming out is often quite similar per batch. If too much glyc comes out, taking some bio too, then the bio is recovered when we distill it, if too little comes out, taking some glyc to the wash tank, then it settles to the bottom and comes out in the wash process, so the flost idea doesn't have to be too accurate.
     
  7. rhyno

    Thread Starter New Member

    Apr 20, 2008
    5
    0
    I've just looked up an SCR and it looks like it could be my answer, if you wouldn't mind posting that up it would be great.
    As far as your uderstanding of the question, your assumptions are correct. It should be quite low current consumption (exactly how much i'm not sure as the plant keeps expanding and changing so there are often parts added and removed) but yes, the control circuit would be in the mA range. The off switch would be by two methods, emergeny stop, or override, which is a momentary NC, and also an electronic stop which i will add later, basically both in series, if one opens, then the plant stops.

    Thanks again
     
  8. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,699
    907
    I will draw it up tonight and post.

    As for the bio diesel, I am a bit surprised there are not electronic sensors already available to detect the change from glycerol to bio diesel. Can you point me in the right direction to read more about it?

    John
     
  9. rhyno

    Thread Starter New Member

    Apr 20, 2008
    5
    0
    Theres not really much on the net in terms of sophisticated plants, but one of the major places i get my info from is biofuelsforum.com.
    Most people are struggling with the chemical side of it, and are far from considering automating their setups.
     
  10. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,699
    907
    Here's the schematic drawn from Mims book. I have not actually built that one, but I have a little experience with SCRs and think it should work.

    I happen to be a chemist, so let me read up on the issue. My initial thought was that you could make a sensor that would tell you when the glycerol reached a certain level or could even measure the level and monitor the rise. Methyl esters and glycerol are quite different animals. With all of the money being spent on development, I will be surprised if it has not been done already.

    John
     
  11. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    There are optical devices that can look for the color change and indicate when the batch has gone from one state to the biodiesel.
     
  12. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,699
    907
    I just PM'd rhyno because I didn't want to hijack his thread. I need to find a place that is doing the conversion and go visit. The chemistry is simple; it's the practice that may not be. John
     
  13. mightywombat

    New Member

    Apr 22, 2008
    1
    0
    Here's a link to an article that will probably answer your needs pretty accurately. There's a link to a PDF with a circuit diagram at the bottom. It recommends scaling up the resistors in the circuit to accommodate higher loads, but you could just as easily put another transistor in line with the load to switch to a larger draw.
     
Loading...