10 minute timer to pick up a 12v relay

Discussion in 'General Electronics Chat' started by aughtago, Apr 13, 2012.

  1. aughtago

    Thread Starter New Member

    Apr 13, 2012
    25
    0
    Disclaimer- I am a real amature in electronics, I am a master electrician by trade, but have only dabled in electronics.
    I am designing a solid state circuit and basicaly need the equivilent of a 12V TDPU relay that is not the size of a juice box, low power consumption low $ and will operate in the 10-15 minute range.
    Despite all I read about NOT using too large of a capacitor, I tried to design my own circuit with a 555 timer. I just breadboarded it and it failed. Is there any hope for this circuit or is there a better option.

    I will try to attach my sketch from sketchup so you can see where I am starting. Note that I deleted the rest of the circuit and left only the timer portion, I need time to start when R contact closes and need to pick up the 12V TR relay after the adjustable time delay of about 10 minutes or slightly longer.
     
  2. crutschow

    Expert

    Mar 14, 2008
    13,006
    3,232
    For such a long period a 555 doesn't work well as you found out. Instead you can use a low frequency clock driving a counter such as this watchdog timer, which can be set from minutes to hours by adjusting the oscillator frequency as determined by the Rt and Ct values.
     
  3. aughtago

    Thread Starter New Member

    Apr 13, 2012
    25
    0
    Thanks, that looks like it might work for me. I'll have to pull up the data sheet for the 4060 and get familiar with it. At first glance it looks like I could drive my output relay right off of pin 14 as shown. my only question is on the input, as I do not have a pulse from the R contact, it closes and stays closed. In my previous circuit I had incorporated a design I found called a trigger network using R1, R2, C4, and D3. I hope something simular will work here.
    thanks again for the quick reply
     
  4. crutschow

    Expert

    Mar 14, 2008
    13,006
    3,232
    The CD4060 doesn't output more than a couple mA so you may have to add a transistor driver for the relay.

    You can make a reset pulse by using a series capacitor to the Reset with a resistor to ground.

    Will the circuit be continually powered or only during the time it is operating?
     
  5. aughtago

    Thread Starter New Member

    Apr 13, 2012
    25
    0
    Thanks again for your help
    Yeh I noticed the low current capability in the datasheet. I will use a transitor off of Q14 to drive the TR relay.

    Not sure I follow you completely, I spell it out in simple terms that I can understand and you can please confirm or deny. do you mean to take VDD through my R contact in series with a capacitor and the reset (pin 12). then add a resistor from pin 12 to ground? would I use an RC calculation to determine pulse width and thus capacitor and resistor value?

    Yes the overall circuit IS 12vdc supplied through a rectifer and connected to battery backup. The R (run) contact closes as the unit starts up and does not open until shutdown is called for.(cycles on t-stat or manual controls) . If other parameters are not met by the time(10-12 minutes) TR relay picks up then the unit will lockout.


    I tried to use the formula listed on the drawing you sent to understand the time delay calc so I could then re-design it for my needs. I could not make the formula result in 5 minutes, what am I doing wrong.
    • first I tried the T OUT formula as follows: 2.2 * Rt(33000) * Ct(.000000047) * 2 * N-1 (13) which equals less than 900ms
    • I tried the same formula using 430K as Rt (resistance total) with simular results
    • I also tried the formula from the datsheet T=2.2 * Rx *Cx which equals about 34ms
    • I then tried thinking out of the box and applied the Freq formula times the 13 stages which equals about 6.3 minutes (closeest I came to 5 min)
    • looking at the tables in the datasheet only confuses me more as thier examples using a 50pf cap and 200k resisitance only result in ns of each parameter.
    thanks again for your help!
     
  6. crutschow

    Expert

    Mar 14, 2008
    13,006
    3,232
    Yes for the reset, place the capacitor in series with a resistor to ground. IF the switch is SPST then you need a resistor on both sides of the cap to ground to provide a discharge path for the cap (see below). The reset pulse period is approximately R*C. 100ms should be sufficient so use a 0.1uF cap with two 1meg resistors.

    Reset Ckt.gif

    The time equation has a typo which is certainly confusing. It should be 2.2*Rx*Cx * 2^(N-1). Thus for the given values the time is 2.2*330K*0.047μF*2^13 = 279.5 seconds or 4.66 minutes (2^13 = 8192).
     
    Last edited: Apr 15, 2012
  7. Lundwall_Paul

    Member

    Oct 18, 2011
    220
    19
    Newark has a nice Time delay relay
    TDR782XBXA-12D, TDR782 Time Delay Relay/ DPDT 5 Amp Rating. easy to adjust between .1 sec to 100 hours DPDT.
     
  8. aughtago

    Thread Starter New Member

    Apr 13, 2012
    25
    0
    It looks nice , but still a little big for my application and pocket book. Thanks for the suggestion, it is far better than any TD relays I had found in my previous searching.
     
  9. aughtago

    Thread Starter New Member

    Apr 13, 2012
    25
    0
    attached is my draft of the circuit. I think I got thr reset right?

    The polarity of the transistor also gave me trouble. I assume that the output of the timer is positive, therefore had to move my TR relay coil and use a npn.

    I am not sure if i still need the protection diodes that I had for the 555 but they can't hurt. come to think of it, i should move D1 to between TR and Q1

    Thanks agaim for your help i would lost without it.
     
  10. crutschow

    Expert

    Mar 14, 2008
    13,006
    3,232
    The schematic looks ok expect for a couple things.

    You need a resistor in series with the transistor base (the base-emitter junction of a bipolar transistors looks like a forward biased diode so you must always have a base resistor to limit the current). Its value should be about 10 times the resistance of the relay.

    Also I don't see any purpose for D1. D2 is all you need to suppress the relay coil transient. Putting D1 between TR and Q1 has no effect since the transient is in the forward direction of the diode (the inductance wants to keep the current moving in the same direction so the voltage at the transistor's collector will increase).

    Edit: The reset circuit looks fine.
     
  11. aughtago

    Thread Starter New Member

    Apr 13, 2012
    25
    0
    Ok, I put the resistor in , got rid of D1 and added a capacitor as suggested in this great youtube tutorial i found.
    http://www.youtube.com/watch?v=8DMZ...UCGXQYbcTJ33Yu0y2941q0EzS3xC9b-ekJBunEBFZvrD0

    The video also explained how to calculate the correct resistor value. I need only now to update the ledend, order the new parts and bread board it again.

    One thing I had forgotten about when you asked me if this circuit was energized all the time. I do have a n/c manul reset button to reset a lockout, but in that case I would want to reset the seal in on R relay and reset the timer too, and I believe that will accomplish that, Right? I attached the entire circuit so you could see what I mean.

    Thanks so much for your help.
     
  12. crutschow

    Expert

    Mar 14, 2008
    13,006
    3,232
    I don't know what your total circuit is attempting to do, so can't comment on that.

    I don't understand the reset switch at the top of the page. It appears to go to just a capacitor connected to ground. Do nodes 20 and 21 go elsewhere? :confused:
     
    Last edited: Apr 17, 2012
  13. aughtago

    Thread Starter New Member

    Apr 13, 2012
    25
    0
    Background:
    This circuit is an "add on" to fix design flaws in my wood pellet stove. I was unable to read the PIC chip on the main control board (so I could assess and re-program it) and the factory has given up on trying to solve the problems associated with this stove. NEVER BUY A THELIN STOVE!

    Nodes:
    Yes, nodes 20 and 21 are from the 12V source. This entire circuit is a mod to the existing control. The round and square nodes you see are actually male and female 37 pin D sub connectors that will be used to "insert" this mod into the existing control circuit between the loads/souces and the control board. Many connections are not shown here, are un affected and will pass from one 37 pin Dsub to the other. Other circuits are tapped off and/or interupted by my mod.

    The hex shaped nodes are terminals for wiring between my new mod board and my add on control panel with the user interface switches and LED's. TB4 is another 2 terminal strip that allows connection to an new temperature sensor mounted on the stove exhaust.

    I am also incorporating many test points on my board to allow easier monitoring or troubleshooting. Including +12v, grd, Timer pin 9, and each of the Q steps off the timer.

    Reset:
    The reset is a normally closed, momentary open, manual reset button. The button is between this circuit and the main 12V power source. It is used to reset the R and LO relays which have "seal in" circuits. I think (hope) that this will also reset the timer to 0 by 1) deenergizing it, and 2) by opening the R contact (reset)

    it 'Aughtago'
     
  14. Lundwall_Paul

    Member

    Oct 18, 2011
    220
    19
    I would use a Time Delay Relay TDR782XBXA-12D it is DPDT which you can set the time delay from 0.1 second to 100 hours.
     
  15. aughtago

    Thread Starter New Member

    Apr 13, 2012
    25
    0
    OK I have been expirementing with this timer circuit on the breadboard for some time now to resolve my mistakes. I released the magic smoke from a couple components and learned a lot about this chip. I now have the timer operating correctly at 10min and 6 seconds, with only 1 tweak of the pot. I also learned a few lessons about this chip and have one remaining question.

    Many of the issues were from my carelessness as I laid it out on the breadboard;
    • I had 1 diode in backwards. (It became an LED for about 1/2 sec.)
    • I had missed a couple jumpers (no power to the chip)
    • I also got the transitor in backwards at one point during my troubleshooting.
    One of my design issues was due to not understanding the output of the chip. For some reason I though its output was 5V, but carefull review of the datasheet reveals it is actually very close to the input voltage (I.E 12V) Therefore I had calculated the drop resistor for the Base of Q1 to small (5.7K vs the 30K required). In the end I measured the Voltage drop acrross on of my three 10K resistors and calcualted about .4ma which is really close to the .34ma target.

    One Lesson I learned that proved valuable for troubleshooting. By placing 3 LED's from ground to Q4 (pin 7), Q12 (pin 1), and Q13 (pin 2) I could monitor the progress of the timer as follows;
    • Q4 would alternate at about 51 cycles per minute allowing me to see when the timer was timing and when it was reset.
    • Q12 nd Q13 help monitor the overall progress.
    • Q12 turns on at 1/4 the total duration (2 min 32 seconds in my case)
    • Q12 turns off and Q13 turns on at 1/2 total duration (~5 min)
    • Q12 turns on at 3/4 total duration (both Q12 and Q13 on)(~7.5 min)
    • Q12 and Q13 turn off when the final output Q14 turns on.(10 min , 6 sec)
    The other design issue was due to not understanding the reset function. I had it set to close R contact to start the timer, but in fact it needs be N/C and open to start the timer. By experimenting I found that it could be simplified by simply using one of the 1 meg resitors from pin 12 to ground and have R connected from pin 12 to Vdd.

    So I ask, Why do I need the C2 and R1 resistor?
     
  16. crutschow

    Expert

    Mar 14, 2008
    13,006
    3,232
    Glad you got it all working. You have (hopefully) learned the necessity for careful construction and checking of the circuit after it is built. ;) I typically use an ohmmeter to check all connections and mark them off on the schematic with a yellow highlighter as I go.

    The reset with the capacitor is for a power-on reset function. I missed the fact that you had a separate reset button. But I'm not sure I understand how you did it. The RESET input does need a momentary logic high for the reset function to occur.
     
  17. aughtago

    Thread Starter New Member

    Apr 13, 2012
    25
    0
    Yeh, I got a little summit fever, just wanted to see it work. After a week of troubleshooting, my 8yo grandaughter was helping me (runing the stop watch) when I finally got it all working. She couldn't understand why I got so excited to see a relay pick up and a test LED light up.

    I plan to leave the resistors and capacitor in the circuit. However, i noticed as soon as I powered the unit up it began timing. I began to experiment to figure out how to reset it and found that even with nothing but a jumper conected to pin 12 it would start timing but q4 output got unstable until I added the resistor to ground from pin 12. It would reset when I placed a jumper to 12Vdd then start timing as soon as I removed the jumper. The datasheet indicates "A high level on the RESET line accomplishes the reset function" does not say anything about requiring a pulse.

    Thanks again for your guidance, I would still be searching the internet, and most likely would have resorted to trying to cram the 12V TD relay others have suggested into the project (really tight fit)
     
  18. crutschow

    Expert

    Mar 14, 2008
    13,006
    3,232
    The datasheet does not say it requires a pulse because the RESET is a static logic function, that is, the chip stays in the Reset mode (count inhibited) as long as the RESET pin is high. And you must never leave a CMOS input open as it will drift to an unknown state, subject to noise pickup. So you reset the chip by momentarily pulsing the RESET high and then force the RESET low, either by connecting it or ground or with a resistor to ground, when you want to start the count.
     
  19. aughtago

    Thread Starter New Member

    Apr 13, 2012
    25
    0
    I know it has been a long time, this project was on hold this summer for other priorities.

    After bread-boarding the timer and working out a few kinks(mostly novice errors installing components backwards etc.) I also found that my R contact for the trigger had to be a NC contact and the timer would start when I opened it.

    I completed my circuit board and finally installed it in the stove for testing last night. I had 2 issues with the timer. I am hoping to get some more advice. see attached final circuit diagram rev H, and photo of the main board as well as the control panel in this project. Note- I will never solder that many wires directly to the board again, what a pain in the neck. I will use more terminal strips for any future projects.

    Note that during the bread-boarding stage I added a few LED's to some of the outputs of the timer for troubleshooting. This worked so well, on the main board, I added LED's to all Q outputs which helped tremendously during the testing phase to watch the timer operate. There is a test jumper on the board where they can be turned on or off. note that when allowed to run the timer finished(Q14 LED lit) at 16 minutes which is perfect for my application.

    ISSUES
    1) the timer triggers as soon as power is applied, I used the same components that were on my breadboard where it worked fine. I can press my reset button to momentarily de-energize it and the timer will start over. I do not want it to start timing until R picks up.
    2) the output from the timer TR picks up almost immediately and it cycles intermittently on and off (not consistently, however when it picks up it seams to be in unison with my timer LEDs changing state.) Again, I used the same components from the breadboard circuit (except on the breadboard I had 3-10K resistors in series, and on the final board I installed 1-30K resistor. I suspect I may have a transistor, or diode in wrong and will start looking there. Any other suggestions on what to look for?
     
  20. aughtago

    Thread Starter New Member

    Apr 13, 2012
    25
    0
    I'm still working on solving the, the only thing I have found so far is that R5 is a little big, (30K design,35K measured and new calc says 22K is right) ,but that would not create the symptoms I see.

    Any suggestions? (see previous post for details)
     
Loading...