Thyristors not latching?

Discussion in 'The Projects Forum' started by ImWolf, May 26, 2013.

  1. ImWolf

    Thread Starter Member

    May 26, 2013
    63
    2
    I recently became interested in electronics which has been a fair mix of triumphs and frustrations... I'm a complete novice in this area.

    My current project is to build a sequencing light circuit for use in a scale model kit. Due to space constraints inside the model kit, I'll be using a 6V DC power supply.

    I've managed to build several working "sequencing LED" circuits on my bread board in various configurations, but none that perform as I expect. What I want is for each LED to remain lit once triggered until the entire sequence is complete.

    The only example of a schematic that I could find on the web is this:

    http://scdiagramwiring.blogspot.com/2013/03/sequential-led-bar-graph-circuit-using.html

    While waiting for parts to arrive, I D/L the Proteus 8 Demo software (which is pretty cool stuff) and played with that, but could not get the above referenced diagram to function until I took out some resistors and tweaked other resistor/capacitor values. So I'm not sure if that schematic is functional.

    When my components all showed up however, I couldn't repeat the working software behavior with the actual hardware.

    I was able to build the AMV successfully on one bread board. I used a version that included LED's so that I could monitor the "flip-flop" rate and ended up using 91K resistors with 4.7u capacitors for the desired pulse rate to send to the 4017 clock pin.

    While trying to build the remainder of the circuit on a separate board though, I have only been able to attain blinking LED's....

    The data sheets I've read on the BT169D SCR's are conflicting information to me so I'm not sure if my 6V power supply is adequate to trigger the gates.

    I've also tried removing the PNP transistor from the power flow completely to test if I could get a solid bar of lights for at least one sequence but that did not work either.... so again, I suspect the SCR's are not latching.

    The last piece of info that might be note worthy is that I have determined 150 ohm resisters were ample to protect my LED's so I am not using the 1K resistors as in the example schematic above.

    OK..... long winded enough? I hope this is not too much info, but still enough for any who wishes to suggest a fix.

    Thank You,
    Wolf
     
    Last edited: May 26, 2013
  2. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    You said you tried removing the PNP. Did you replace it with a wire between emitter and collector?
     
  3. ImWolf

    Thread Starter Member

    May 26, 2013
    63
    2
    Hi Ron,

    No, I did not modify the PNP transistor... I just by passed it by sending the +6V supply directly to the anode ends of the LED's.

    As I understand it, a PNP is normally in the closed state and remains so until it's shorted by sending current to it's base. But just to make sure I hadn't wired that portion of the circuit wrong I by passed it to see if I could get my string of LED's to remain lit for at least one sequence.

    They just blinked on and off though in exact time as the AMV clock attached to the 4017 IC.
     
  4. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    What is the forward voltage of your LEDs? Can you post a part number, or a link to the datasheet?
    Do you have two LEDs in series, as shown in the schematic?
    When the LEDs flash, are they bright, or dim, or??
     
  5. ImWolf

    Thread Starter Member

    May 26, 2013
    63
    2
    Hey Ron,

    I'm using your garden variety green 3mm LED's rated at 3.0-3.2V / .02uA. I don't have a data sheet on them.

    The SCR's are BT169D TO-92 rated at 400V / 0.5A. I've looked up data sheets on these and the various sheets don't even agree on the pin designations, let alone the min. trigger voltage.

    My power supply is two 3V coin cells that I bridged to give me the 6V input.

    I'm still playing with different configurations while completely by passing the PNP (off) switch, but I can't get the desired result...... :*(

    But thanks for your interest..... two heads are better than one!
     
  6. MrChips

    Moderator

    Oct 2, 2009
    12,410
    3,353
    Have you tried testing the SCR by itself with just a load resistor and triggering it manually?
    What pinout are you using?
     
  7. ImWolf

    Thread Starter Member

    May 26, 2013
    63
    2
    BTW Ron, I do believe that the LED's in the (first post) schematic are wired in parallel?

    Ahhh..... if you're referring to the LED's in the AMV then yes..... they are in series.... but I put my LED's between the outer resisters and the NPN transistors..... which doesn't effect what I'm testing now as I by passed that entire part of the circuit just to test if I can obtain one sequence of "stay lit" LED's.
     
    Last edited: May 27, 2013
  8. ImWolf

    Thread Starter Member

    May 26, 2013
    63
    2
    Hi MrChips,

    When I became interested in electronics my g/f bought me a multi-meter, but I barely know how to use it, and I don't know how to test the SCR's or how to trigger them manually. I started out on this forum by studying the SCR data available but didn't find what I was looking for there....

    The pinout I'm using on the SCR is actually both because I keep flipping them 180 on the board to test both possible current flows (if they even exist).

    The pinouts I'm using on the 4017 are the correct sequence of outputs because my LED's do consistently blink as expected.

    On that subject however, I am still wondering if the pinouts of the 4017 IC are bi-polar, or if they are best used as either the Pos. or Neg. leg of the circuit?

    If that Q sounds stupid...... well..... there's a reason for that! LOL
     
  9. MrChips

    Moderator

    Oct 2, 2009
    12,410
    3,353
    Test the SCR as shown. With the gate connected to GND (-ve side of the supply) measure the voltage across the load resistor R2. Voltage should read zero volts.

    Trigger the SCR by connecting the gate to the +ve side of the supply via the gate series resistor R1 as shown in the second schematic. Measure the voltage across the load resistor R2. Voltage should be supply voltage minus 1V.

    Disconnect the gate. SCR should remain latched. The only way to unlatch the SCR is to turn off the supply or disconnect the load resistor R2.

    You can also add an LED in series with the load resistor as an quick test indicator.

    The pinout of the SCR shown is for BT169D.

    [​IMG]
     
  10. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    If you have two 3V LEDs in series with each SCR, and you're using a 6V supply, you will probably never be able to get enough current to flow to keep the SCRs latched.

    Try this on one of the SCRs: Use only one LED and a 220Ω resistor in series with the SCR. You might want to start by bypassing the PNP. That should give you about 10mA through the LED. If this works, try it on another one.

    If you really want 2 LEDs in series with each SCR, you will need at least 8V on your supply.

    Tell us with your LED requirements, including current. BTW, you said the LEDs are rated at .02 uA. That should be .02 mA.

    EDIT: I meant to say 20mA.
     
    Last edited: May 27, 2013
  11. MrChips

    Moderator

    Oct 2, 2009
    12,410
    3,353
    You mean .02A ?
     
  12. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    I made the same mistake as the OP.:(
    Yes, I did mean .02A, i.e., 20mA.
     
  13. sheldons

    Active Member

    Oct 26, 2011
    616
    101
    you need to increase the supply to 12v -if you have put it together as per the schematic it might start working,the supply you are using is probably not good enough current wise-also the circuit as shown doesnt look to be too impressive there are a few changes that i would make if i were building it
     
  14. ImWolf

    Thread Starter Member

    May 26, 2013
    63
    2
    @MrChips: Thank you for that input. Other tests demonstrated that the pin designations on my SCR's are as the diagram you supplied. I wasn't able to build either of the test circuits though..... I just couldn't "reason out" what's happening with them to a point were I could replicate them on a board.

    @Ron: I stand corrected on a couple flaws you picked up..... although the original schematic does show 10 parallel series of 2 LED's, I have only been using one attached to each SCR. Also, you are correct that the LED's are rated at 20mA.

    @Sheldons: You may be correct about the inadequate power supply. I've made a bit of progress however and time will tell if the 6V supply is a major issue.

    Using another SCR testing schematic I found out there, I was able to finally get a single SCR to latch properly using the 6V supply. This leaves me a bit puzzled as I'm sure I used this sort of configuration already when attempting to light up the string of LED's. I'm wondering if I have too much power loss after current has went through the AMV, the 4017, etc?

    I had also performed the "crude" multi-meter test on one of my SCR's. With the meter set for Ohms I used the pos. probe at the Gate and the neg. probe at the cathode.... the display read 10.23 M (o?). With the probes in all other configurations the display remained at O.L.

    Then with the meter set to "diode" testing, and the probes as above, the display read .703 V. Again, all other probe configurations left the display at O.L.

    Next I need to get back to work on the string of LED's and make sure I follow exactly what the test schematic showed me works.

    I'll attach a photo of that schematic, along with a screen capture of the actual full PCB design I've been trying to imitate. (even though it worked in the software but not with the hardware so far). :b
     
  15. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    Your new schematic has no LED current limiting resistors! You will destroy the LEDs and/or the SCRs and/or the PNP if you get the PNP biased correctly. The PNP base does not have enough current to drive even one LED, let alone 10. If you run the LEDs at 20ma each, the PNP has to pass 200mA. It should have 20mA base drive to guarantee saturation, although, as a one-off, you could probably get away with 10mA. In any case, the 4017 can't drive the PNP.
    I would change the PNP to a Darlington. Jameco has TIP115 for US $0.25.
    Then you can use 10k for R5.
    I would also change all the SCR gate resistors to 10k. The SCRs need at least 200uA to ensure triggering. 10k will give them 500uA.
    Your LED current limiting resistors need to be around 82 to 100 ohms each if you want 20mA.
     
  16. ImWolf

    Thread Starter Member

    May 26, 2013
    63
    2
    Of course you are right about the lack of resistors Ron..... and even though I'm very new at this stuff I never actually tried to build this circuit without adding at least the 150ohm resistors at the anode end of my LED's.

    You've added a few other observations which I really do appreciate. I'll need to study your comment a few more times for sure.

    Perhaps in the mean time Sheldons will make another appearance and give me his rendition of a better schematic.... :b

    Thanks guys....
     
  17. ImWolf

    Thread Starter Member

    May 26, 2013
    63
    2
    Hey Ron,

    The first thing that jumped out at me because of all the comments I've received on this project is that I've really done a poor job of providing specific information to you and the others attempting to help me.... I'll try to furnish much more detailed and exact information from here out.

    I'm usually a quick study at new lessons, but this electronic stuff has certainly thrown me for a few loops. I assumed because of beginning successes that I could just purchase the correct hardware, connect it all up, and it would all function "like magic". Obviously there is a lot of math involved though that I'm just being introduced to, let alone understanding how different components behave and interact with each other. I'm not a math slouch either, but I need to know what needs to be calculated.

    That being said, I'd like to start being much more exact in describing what I have to work with, what I desire in the end, and how I'm actually going about it...... and perhaps address some of your latest comments along the way.

    When you bash/correct something I'm assuming or attempting along the way, please know that you are not insulting me in any way..... I already know that I'm ignorant on many aspects in this field...... you are teaching me...... and I appreciate your time.

    So, I'm building a scale spaceship model kit. I want to light the windows, the engines, the hanger deck..... maybe more. It's a small kit but a 6V power supply built by bridging two CR2032 lithium batteries will fit just right into the hull space available, and an access area. I can go up to 9V by adding a third watch battery, but a standard 9V cell is too big to fit in there.

    As I understand it, I can power as many components as I wish off my power supply as long as I wire them in parallel..... I just lose battery life as I add more lighting?

    If the above is true, then I just need to make sure I can get this hanger bay landing light circuit working within the 6V (or 9V) power constraint?

    After much time on the web, the only example I found of what I wanted to build this is the schematic I originally posted. But I knew I wasn't going to use a 12V supply nor did I need the double LED's, so it was my bad for not mentioning this in earlier posts.

    From an earlier LED lighting project, I had determined that 150ohm resistors were the minimum to protect my 3.0V / 20mA LED's from being killed by the 6V power supply. So I always used at least this when powering the LED's in the many configurations I played with.

    After 2 weeks of failure, I was able to latch an SCR via a test schematic, but the resistor value I used was much higher than I expected. In my minds eye, more resistance means less current, so I did not understand why a 1k resistor at the gate allowed the SCR to trigger, when just applying the raw 6V current would not..... nor do I understand why you suggest increasing the resistance to 10K.

    As far as the PNP transistor (off switch) is concerned, and your comment of "if you get the PNP biased correctly".... I really don't know what this means Ron. Does biased translate to having emitter/collector polarized correctly? As I understand it, the PNP is a closed switch in the normal state, and it's a (pos, opp?) flow of current to the base which opens the switch?

    Anyway..... even though the SCR test latch went fine, when I tried to repeat this process on the other side of my board (after current was powering the AMV and going through the 4017) the LED did stay lit, but just barely..... very dim flickering light. Same result with 150ohm or 1k at the anode of the LED, while 1k and 10k at the SCR gate.

    Need to do some math?

    Thanks,
    Wolf
     
    Last edited: May 28, 2013
  18. MrChips

    Moderator

    Oct 2, 2009
    12,410
    3,353
    Quite often (too often) folks post on AAC a circuit or a situation requesting assistance but they fail to say up front what they are trying to do. You can save a lot of frustration and wasted time and effort if you say from the first post what you are attempting to do.

    Now that we know, there is more than one way to skin a cat.
    Skip the SCR and PNP transistor. You can drive LEDs directly from a CMOS chip @ 6V (along with current limiting resistors).

    LED basics. Most LEDs will operate at 20mA @2V (3V for blue LEDs). When looking for longer battery life, find the lowest LED operating current that will give adequate brightness. You can control the LED current with proper choice of resistor value. Use Ohm's Law to calculate the resistor value. Many high efficiency LEDS are quite visible with only 1mA current.

    If the supply voltage is Vs and the LED forward voltage is Vf,

    the LED current I = (Vs - Vf)/R

    where R is the current limiting series resistor.

    To create your latching LED feature, use a serial-in parallel-out shift register, such as 74HC164 or 74HCT164.

    For a clock source you can use an LMC555, or a two transistor oscillator, or a simple UJT circuit.
     
  19. ImWolf

    Thread Starter Member

    May 26, 2013
    63
    2
    Having just arrived at this forum MrChips, I have no clue how often you folks are frustrated by new comers..... However, In my first post I stated quite clearly that I wanted to build a sequencing light circuit where the lights stay lit through each cycle, that this was for a scale model kit, and due to space constraints I really wanted to stay with a 6V power supply....

    Perhaps you did not read my first post?

    It was then that I offered up an example schematic and stated that this was the only thing I could find on the web which might give me the end result I seek, but I was having problems with it and that I suspected the SCR's were not latching.....

    If at this point someone had wrote "Hey Wolf, there's a better way to do this", then perhaps I wouldn't have wasted time this last weekend on a futile task?

    I did buy some SCR's, PNP & NPN transistors, and some extra resistors that matched up to the schematic I found, but I doubt that comes to $5.00. In the end I just need a circuit that works, and the less components the better because I do need to fit all the garbage onto a rather small PCB.

    What I have in my tool box at the moment includes the above mentioned items, some 3909IC, 4060IC, 4017IC, 555IC, a fairly good assortment of LED's and resistors, and a small variety of capacitors. (I had seen a video somewhere in my travels where the author stated that using a 555IC was not so great for DC supply as that chip eats up battery life quick.... I don't know, but he sounded like he knew what he was talking about)..

    At this point I guess I'll look up the components you're suggesting to see if anyone has an example/schematic/video of what they can do or not do (since you didn't offer any), but for future reference, please read ppl's posts b4 you accuse them of wasting your time.

    Thanks again.....
    Wolf
     
    Last edited: May 29, 2013
  20. MrChips

    Moderator

    Oct 2, 2009
    12,410
    3,353
    Sorry Wolf, I didn't mean to jump on you.

    The frustration and wasted time and effort I am referring to is yours, not mine. What I mean is we could have avoided all the frustrations of having to get SCRs working when they are not needed at all.

    I am here to provide you with advice and show you a better and simpler way of doing something.
    You can take my suggestions or go ahead and continue to trouble shoot your circuit. Makes no difference to me.

    btw, I read your post completely before responding. I offered hints on how to test your SCRs, not how to redesign your circuit. I had assumed you knew why you wanted to use that circuit.

    And don't worry about hurting my feelings, I have a thick skin. Don't get so jumpy so quickly when you are the one seeking advice.
     
Loading...