Triac Control

Discussion in 'General Electronics Chat' started by Zenock, Jun 1, 2009.

  1. Zenock

    Thread Starter Active Member

    Jun 1, 2009
    36
    1
    I apologize if this has been asked before or if I'm asking in the wrong place.

    I am looking to build a controller for a sprinkler valve. I want to use a 5V microprocessor to run the valve. I have pretty much decided to use a Triac instead of a relay.

    1. Any reason I shouldn't do this?

    The specifications for the sprinkler valve are as follows...

    24 VAC
    Inrush current .35 Amps
    Holding current .23 Amps

    I'm considering using the following Triac...

    http://www.littelfuse.com/data/en/Data_Sheets/Littelfuse_Triac_LX8.pdf
    Model LX807

    2. Will this work for this application?

    3. Do I need an Optocoupler (The data sheet seems to imply that it can be driven directly from the microprocessor.)?

    4. Can you explain the quadrants to me?

    I thought that if you wanted the full wave, you needed to have the gate driven with a matching wave. However, this seems to indicate that you must have a matching current but not a matching voltage. The voltage can be a constant 1.3 Volts DC. Is this correct?

    Any help understanding this would be greatly appreciated.
     
  2. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    No. It can work. The TRIAC model you have pick is fine too.

    What you will need is an interface between the microprocessor I/O pin and the 24V system. This can be safely and easily achieved using an opto-coupled TRIAC driver to provide the gate drive to you load TRIAC.

    The opto-coupled TRIAC MOC302x series, which differs in LED drive milliamp sensitivity could be driven directly by the I/O pin of processor(if it is small enough to be within the I/O drive capability) or via a NPN or PNP transistor. So select which one to get carefully.

    You may have to reduce the values of those two resistors if you are using 24V instead of 220V.

    [​IMG]



    It would be better for your understanding if you search for "SCR quadrants" instead in Google and read the material that turns up.

    Posts back if you have questions about the information in them.
     
  3. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    8,754
    760
    When powering inductive loads like Induction motor with traics is the only main reason you should be concerned about TRAICS firing Angle and quadrants, a simple load in your case is nothing to worry about. The circuit above is more than enuf for your needs. The traic opto coupler / driver is well suited for firing traics in full conduction. it has zero crossing detection which minimizes switching noise and instantaneous load currents.
    Still if you like to know I can post it for you.

    Rifaa
     
    Last edited: Jun 2, 2009
  4. Zenock

    Thread Starter Active Member

    Jun 1, 2009
    36
    1
    Riffaa -- Thank you for your reply. I will be powering a sprinkler valve. This is a silonoid. Is this not an Inductive load?

    eblc138 (L.Chung) -- Thank you for taking the time to respond. I am familiar wiith the schematic provided and feel perfectly capable of building the circuit with the optocoupler. However, my question was of whether I really needed the optocoupler or if it would be possible to drive the triac without it. From the datasheet, "offering direct interface to microprossesor drivers" seems to indicate it might not be needed. Also it shows a trigger current of 7 mA and a trigger voltage of 1.3 volts. Both of which can easily be provided by the Microcontroller I'm using.

    The Datasheet shows a "DC Gate trigger current" and "DC gate voltage" this lead to my question about quandrents and why I'm confused. I thought the voltage and current on the gate would have to reverse for the triac to continue working during the bottom cycle. Thank you for suggesting SCR quadrants as a google term. I had tried many other google terms to try to get a better understanding of this. But had not tried this yet. Unfortunately, I'm not sure if you googled "SCR quadrants" however none of the results I got seems to shine any further light on the question. I will continue to try to find information. But if someone could point me in the right direction, I would appreciate the help.

    Once again, thanks for your responses I truly appreciate them.

    Z
     
    Last edited: Jun 2, 2009
  5. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    If isolation between valve supply system and microprocessor +5V is not required, then you can of course connect the drive signal from the uP directly to the TRIAC gate.

    It is usually connected as shown, i.e. pulldown on the uP output pin to turn ON the TRIAC gate to use the TRIAC in quadrant 2 & 3. You don't have to provide a "matching" polarity signal to the TRIAC as the TRIAC will conduct fine between quadrants 2 and 3 with a constant 0V gate signal.

    [​IMG]

    I'm sorry to have misled you. The search term should be "TRIAC quardants" instead of SCR.
     
  6. Zenock

    Thread Starter Active Member

    Jun 1, 2009
    36
    1
    eblc1388 --

    Thank your so much for you help. I truly appreciate it.

    Let me go over with you my understanding on the Triac. Hopefully, you will tell me where I'm right and where I've got it wrong.

    1. If the gate is allowed to float the triac will be off.

    2. If there is a negative current on the gate and mt2 is at high voltage the triac is on opperating in quadrant 1. If there is a possitive current on the gate and mt2 is at high voltage the triac is on and said to be opperating in quadrant 2. Negative current and mt2 is at low voltage, on and opperating in quadrant 3. Possitive current and mt2 at low voltage, on and opperating in quadrant 4. (This is all assuming gate current meets minimum thresholds.

    3. If the voltage on the gate meets a certain thresh hold the triac will be on in all quadrants (Here polarity must match?)

    4. If the change in voltage on the gate exceeds a certain threshhold the triac will trigger on.

    In all cases, if a threshhold current is reached the triac will remain on until it drops below the threshhold.

    If this is all correct then it would seem it doesn't matter if the gate goes to a high voltage or a low voltage, I'm going to have current through the gate and thus the triac will be on. So the only way to shut it off is to let the gate float. This can't be right?

    Help

    Z
     
    Last edited: Jun 3, 2009
  7. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    Maybe. But letting the gate of a TRIAC float is nothing but bad practice. In the above post#5 circuit, one can output a logic HIGH on the uP pin to stop the TRIAC from triggering, instead of letting the gate float.

    Not quite. You have the quadrants mixed up.

    For positive gate current, the TRIAC operates in either quadrant 1 & 4, depending on the moment if MT2 is more positive/negative than MT1. With negative gate current, it operates in quadrant 2 & 3. See image:
    [​IMG]

    Datasheet usually specify gate current threshold instead of gate voltage.

    This threshold current is the current via MT2 & MT1, i.e. the TRIAC current.

    Commutation is always a problem with TRIAC, especially on non unity power factor loads. Therefore, for some TRIACs, the manufacturer had changed the construction of the TRIAC gate structure so that the TRIAC will now NOT triggerable in quadrant 4 for these "new, improved" TRIAC.

    If these type of TRIACs is being used, the only configuration that will allow it to maintain triggering with a unidirectional gate current and AC voltage between MT2 & MT1 is to use the connection method as shown in post#5, i.e. triggering the TRIAC in quadrant 2 & 3.
     
  8. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    8,754
    760
    It's not a safety practice to use traic in 220VAC without isolation but in your case a transformer with 24VAC is quite safe but I suggest you use the opto since zero crossing feature is an advantage plus a guarantee that traic will switch without a flaw

    Rifaa
     
  9. Zenock

    Thread Starter Active Member

    Jun 1, 2009
    36
    1
    Rifaa --

    Thank you for your comment. To be honest I most likely will use an opto (depending on the cost, 8 valves = 8 optos). This really is more about understanding something than about how I'm actually going to end up doing it.

    eblc138 (L.Chung) --

    Thank you so much, I'm starting to understand better. A few things still confuse me...

    1. The data sheet for the triac I'm looking at listed in my first post appears to specify a "gate voltage" am I understanding the "gate voltage" incorrectly?"

    2. You said, "In the above post#5 circuit, one can output a logic HIGH on the uP pin to stop the TRIAC from triggering, instead of letting the gate float." I wasn't suggesting letting the gate float I was confused at how to turn it off. I still am. If you drive the output pin high on the uP, won't that push the Triac into quadrants 1 and 4 causing it to continue to be "on"? This is what is confusing me the most.

    3. In the circuit in post 2, isn't the gate essentially floating when the optocoupler is off? Thus causing the triac to shut off on the 0 crossing? If this is "bad practice" is there a better way to use an optocoupler?

    4. How do you use the multi quote option? I tried to quote you but I must of been doing it wrong.


    Thank you so much for all your help.
    Z
     
  10. bertus

    Administrator

    Apr 5, 2008
    15,649
    2,348
    Hello,

    @ alberto,
    The specs of the sprinkler are 350 mA inrush and 230 mA holding current.
    There is a need for an extra triac, a MOC alone will not manage.

    Greetings,
    Bertus
     
  11. Zenock

    Thread Starter Active Member

    Jun 1, 2009
    36
    1
    Alberto:

    Thanks for replying, I always appreciate everyones help and comments. However, please correct me if I'm wrong but...

    .35 Amps = 350 mA > 100 mA
    likewise
    .23 Amps = 250 mA > 100 mA

    All the best...
    Z
     
  12. Zenock

    Thread Starter Active Member

    Jun 1, 2009
    36
    1
    bertus --

    I looked up the datasheet for the moc here...

    http://www.fairchildsemi.com/ds/MO/MOC3020M.pdf

    It shows ...

    I
    TSM Peak Repetitive Surge Current(PW = 1ms, 120pps) 1 A​


    I'm assuming ITSM is the current through the Triac? Perhaps Alberto merely miss-typed and the moc can handle it? ​



    In which case maybe I'm putting the cart ahead of the horse and this is the better way to do it? I still hope to get all my questions regarding TRIAC control answered as that I'm sure they will be pertinent to me in the future.​



    Thanks
    Z​
     
  13. bertus

    Administrator

    Apr 5, 2008
    15,649
    2,348
    Hello,

    Here are the driverspecs from the MOC3020.

    [​IMG]

    The 1.2 A for 10 mS in non repetitive.
    The 1 A from your datasheet is 1 mS pulses 120 pulses per second.
    The 100 mA is the continues current allowed.

    Greetings.
    Bertus
     
  14. Zenock

    Thread Starter Active Member

    Jun 1, 2009
    36
    1
    Bertus --

    Yeah I wasn't sure what the "TSM" was for. When I looked further down in my datasheet I found the same info you gave.

    So it's back to my original questions. Here's hoping they get answered.

    Thanks,
    Z
     
  15. bertus

    Administrator

    Apr 5, 2008
    15,649
    2,348
  16. Zenock

    Thread Starter Active Member

    Jun 1, 2009
    36
    1
    Bertus --

    Thanks tons. That page is great!

    A few things still confuse me...

    1. The data sheet for the triac I'm looking at listed in my first post appears to specify a "gate voltage" am I understanding the "gate voltage" incorrectly?" I'm just not getting it. Gate current I understand better. But what is the gate voltage in relation to MT1 or MT2?

    2. L. CHung indicated, "In the above post#5 circuit, one can output a logic HIGH on the uP pin to stop the TRIAC from triggering, instead of letting the gate float." I am confused at how to turn it off in this circuit. If you drive the output pin high on the uP, won't that push the Triac into quadrants 1 and 4 causing it to continue to be "on"? This is what is confusing me the most.

    3. In the circuit in post 2, isn't the gate essentially floating when the optocoupler is off? Thus causing the triac to shut off on the 0 crossing? If this is "bad practice" is there a better way to use an optocoupler?

    If I can get some insight into these questions I would sincerely appreciate it. Especially question 2.

    You guys are awesome!
    Z
     
  17. bertus

    Administrator

    Apr 5, 2008
    15,649
    2,348
  18. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    I might be wrong on this explanation.

    That is a case where there is no current limiting between gate drive and gate pin. If you put DC voltage higher than that limit to the gate pin without some forms of current limiting, the TRIAC might suffer damage.

    Perhaps this parameter is useful in calculating the resistor for current limiting.

    You have missed that all voltages are actually "voltage difference" and thus relative to some point of reference. Also the MT1 & MT2 pin designation of a TRIAC and how they are connected up plays an important role too.

    Instead of MT2 being connected to +5V, in the circuit of post#5, TRIAC MT1 is connected to +5V(High). If one sets the "OUT"(Gate) pin to high also, then there will be no(or minimal) voltage difference between Gate pin and MT1 pin regardless of the AC cycle voltage between MT1 and MT2. Without a voltage difference, no gate current can flow thus no triggering.

    Yes, but one can always connect a resistor between gate pin and MT1 so that the gate is now not floating. Letting the gate pin float would invite unwanted triggering in noisy electrical environment.
     
  19. Zenock

    Thread Starter Active Member

    Jun 1, 2009
    36
    1
    L. CHung --

    Thank you!!

    Everything just slammed into place for me. I was thinking that 0V and nuetral were tied together! Thus my confusion. But looking at the schematic again, they are obviously NOT (I also somehow missed the fact that +5V and hot were tied together)

    Let me see if I've got this right... you can correct me if I'm wrong...

    When the triac is off the out line is high. The voltage on the gate is the same as the "hot voltage". The hot voltage referenced to nuetral in my case would be 24 VAC. Referenced to gnd it would be 5V. Since the reference on the gate matches MT1 the voltage difference = 0 and there is no current. So it's off.

    When the line goes low, the voltage on the gate goes to gnd so there is a -5v difference and current flows out the gate turning on the triac. Thus MT2 nominally get's the same potential as MT1 and the voltage difference across the load is (nominally) 24VAC.

    Hope I got that right I think this is other than I might have gotten MT1 and MT2 swaped around (Never mind went and checked which one was MT1 and which was MT2 I think I have it right).

    Thanks again,
    Z
     
    Last edited: Jun 4, 2009
  20. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    Yes. That would be pretty much about it really. :)

    Now, at those level of current and voltage, a small sealed relay can also be a good alternative.
     
Loading...