Please help me with High Low voltage cutoff with time delay

Discussion in 'The Projects Forum' started by sakbd.2010, Oct 28, 2010.

  1. sakbd.2010

    Thread Starter Member

    Sep 19, 2010
    39
    0
    Hi,

    Power line fluctuation is very common in my country, so was looking to build a Hi/low cutoff circuit for my power inverter which do not have one.

    I hv found this circuit that looks nice,

    http://www.free-circuits.com/circui.../high-and-low-voltage-cut-off-with-time-delay

    or
    http://www.elecpod.com/powerinverter/2010/06243261.html

    I hv built some circuit before and can breadboard one if needed. But the one I'm interested in uses LM741 op-amp, but I got one LM393(comparator) available with me, so can someone please help me convert above circuit to use LM393(Comparator) Instead of LM741 op-amp.

    I have already got the linear supply as was in above circuit detail. I was hoping to simplify the circuit with a single ic LM393 Instead of two 741's.

    Also, I cannot find SL100 NPN transistor, But got some 2n2222a and BC547's, will they work instead of sl100 ?

    Please help me build this circuit, as I need the inverter and hi/low protection for it, in fact writing this without line power i.e. backup power from the inverter.

    My power line 230VAC 50 Hz

    I have got one more question, How to protect the power supply circuit and transformer when the power line goes beyond 270(some times it does) any ideas ?? ....

    I have also got some quad comparators HA17339, which I believe are equivalent to LM339. hOPE can someone help me in this regard...
     
  2. sakbd.2010

    Thread Starter Member

    Sep 19, 2010
    39
    0
    the actual circuit to modify...
     
  3. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    I attached a modified circuit which doesn't require dual power supply but just 12V only.

    The relay represented by R14 in the simulation, will operate to cut off the supply to your appliance. So you would need to wire the appliance supply in series with normal CLOSE relay contact.

    Judging from your post I think you can figure out most of the circuit without additional explanation.

    If there are over voltage or under voltage for a brief moment, the 555 will trigger but the relay will not operate.

    Only during prolonged(timing controlled by 555) over/undervoltage the relay will operate to cut off the power to your appliance.

    Post back for additional questions regarding the circuit.

    [​IMG]
     
  4. sakbd.2010

    Thread Starter Member

    Sep 19, 2010
    39
    0
    thanks...really appreciate your circuit.

    But I'm no designer, not very good with circuits. I do understand most part, but also got a lots of questions regarding the parts u hv used in the circuit.,

    1. TL431 : As per datasheet it's a "Adjustable Precision Zener Shunt Regulator" , looks like a small BC337 transistor!!, am I right ? if so, does it require a voltage devider, to feed it with exact voltage, how is it done in the circuit ?, please let me know. U do not hv to be sober! if I said everything wrong here, u can tell me all as a teacher!!, after all a'm learning electronics.

    2. M1 : Mosfet , As per datasheet it's a SMD 8 pin component, Is there any other p-ch mosfet substitute, i.e. a 3 pin type ? if not how do u connect this mosfet!! (I know stupid question but I'm a newbie)

    3. R14 = 200 ohm : If I use any relay with 400 ohm relay would that be OK, as it shall be used for A/C main switch of the inverter.

    4. 555 Pin 3 "OUT" : Just above D4 u show a connection point to "555_OP", can u pls. elaborate this connection(in detail if possible)

    5. Preset : "rtot 10k wiper .5", meaning set it to about half i.e. 5k, is that right ?


    Also, as u know this circuit is very old, most sites hv this in their archive. There are other simple one's with only transistors or with a one or two 555 ic's, I'm sure U understand how it is better then them, Can you help me figure this part!!(if possible)

    At the last but not the least, I must thank you for your time and effort in helping me for the circuit, highly appreciate.
     
  5. sakbd.2010

    Thread Starter Member

    Sep 19, 2010
    39
    0
    TL431 :If it's what I think it is, here pin 2(ref) shall be connected to pin 1(cathode) and pin 3 shall go to ground. But, R6 and R7 confusing me, Please let me know how are they working together TL431.

    Lot of questions., sorry about that, I really am into building this one as soon as possible.
     
  6. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    No problem. I'll try my best to get you the answers.

    The TL431 provides a very stable 2.5V reference voltage for the LM393 comparators.

    The TL431 is just a fancy zener diode which can have its zener voltage adjustable by two external resistors. However, it already comes with internal 2.5V zener voltage without any adjustment. Adding resistors would enable user to change the zener voltage upwards. In the circuit, we just want 2.5V so no resistors is needed and I just connect the "adjust" terminal to the output pin.

    R6 and R7 is *NOT* there for zener voltage adjustment but together with C4 to filter the supply ripples to enable the TL431 to work more efficiently.

    The part number is just for simulation purposes. You can use any P type MOSFET with current of a few hundreds of mA. You can even use a PNP darlington transistor if the P-MOSFET is difficult to source.

    Absolutely no problem there. The relay coil resistance value(how low it can be) is only limited by the maximum current capability of the semiconductor that drives it, in this case the MOSFET.

    It has no significant at all. It is just a label for me to plot the waveform on the simulation graph.

    Yes. It is a 10KΩ variable resistor, .5 means 50% setting so set in the middle. Its there so I don't have to use two 5KΩ separate resistors for the simulation as I might have to adjust it for the best simulation results. If the setting is .6, then it means a 4KΩ and 6KΩ resistors.

    If you look carefully, most site just copy among each others without quoting the source. Some of the sites then broke up the whole circuit into parts for whatever purposes.

    In short, the circuit functions like the following:

    1. the two LM393s, with one input fixed at voltage around 2.5V(by the TL431) and the other input connects to the variable resistor which divide down the DC supply voltage into around 2.5V. We will call this sense voltages. The DC voltage is related to the AC voltage by a fixed ratio. See the green comment on the circuit diagram in bottom left corner.

    2. the LM393s shares a common output via two diodes and adjustment of the variable resistors will change the input sense voltage so as to keep both comparators output high at normal external AC voltage.

    3. at any one time, the comparators output will both be high or one of them goes low(because of external over or under voltage).

    4. via a capacitor, the comparator output is coupled to the trigger input of the 555 timer, which forms a monostable, with timing of a few seconds. The 555 will trigger if there is either a over voltage or under voltage detection occurrence. Its pin#3 output will goes high.

    5. the final output relay is driven by a P-ch MOSFET, where its gate is drive by the LM393 output *OR* the 555 output via the diode D4. A low on the comparator output will drive the MOSFET and thus the relay. A high on the 555 output will drive the gate high thus prevent the relay from operating.

    So, even if the comparator wants to drive the relay to cut off the supply, the relay will not operate because the triggered 555 says not yet.

    6. if the over/under voltage situation has gone before the 555 timed out(a timing determined by R3/C3), the comparators output go high. Everything is good so no relay operation occurs. This is shown in the first part of my simulation(from t=0s to t=30s) where the abnormal AV voltage happens for a second or two and return to normal. Note that the output of the 555 changes but the relay does not operate, shown by no current in R14.

    7. if the over/under voltage situation persists, the 555 will time out and so the relay will now be driven by the comparator to cut off the supply to your appliance. This is shown as the remaining part of the simulation(t>30s) where the over/under voltage situation lasts for more than a few seconds. At around t=32, the circuit sense over voltage and 555 triggered. At t=38, 555 timed out and relay operates to cut off the appliance.

    8. after the 555 time out, if the supply returns to normal, the comparators output will go high again and the relay shut off.

    I hope my explanation is clear for you to follow.

    By the way, what is the actual over voltage and under voltage settings that you want the circuit to operate in real life situation?

    You will be having a difficult time to verify that circuit can really operate correctly unless you have a variac to actually create the under/over voltage situation at the AC input to the rectifier.
     
  7. sakbd.2010

    Thread Starter Member

    Sep 19, 2010
    39
    0
    Hi, Chung,

    Thanks again, very nice explanation...

    Regarding setting up the circuit with presets : You are right I don't hv a variac. But I can set the over voltage very easily with the inverter itself. I know it can supply anywhere from 250 ~320 VAC . I set it up for 260VAC backup, but It got preset to get the voltage as high as 320VAC and my computers and other stuff works just fine, though for safety I lowered it to 260VAC. so, High voltage setup is no problem. But I hv no way to simulate the actual low voltage setup as it cannot supply lower then 250VAC at 50 Hz.

    But U can still help with your computer simulator, as I did not see it before(sorry for that) your simulator can predict(not exactly real but close) and can tell what can be the ratio of AC TO DC as u mentioned and can determine the low side of the preset of the comparator. All I need to set this thing up for,

    High : 270 VAC; LOW : 170 VAC,

    NO CUTOFF zone : 270 ~ 170 VAC and corresponding VDC.

    I got stabilizer that can keep supply voltage steady at 230VAC for as low as 160VAC.

    So, can you please have your simulator run at 150 VAC and 170VAC and also 190VAC(I think as per ur exp. 9.9VDC for 190VAC and preset at 5.8kohm) and tell me what is the preset value for the U6(Preset) that way I can try this live when it occurs and see what the circuit does!! I will hv a realtime cheap multimeter ready at all times for the A/C supply. Meanwhile I'm also thinking other ways to simulate this, some friend of mine gave me an idea about setting up two transformer to simulate this...I will hv to investigate this further.

    P-Channel Mosfet : I do hv some PNP power transistors i.e. BD140, BD242, B649A(2SB649). As per u, I Can probably use one of them, if not I shall get a few P-ch mosfet no problem.

    I hv got most of the parts except few resistors, I will need a day or two to get them over to me. I already go the TL431 today...
     
  8. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    You don't need complicate setup to correctly set the VRs trip points.

    You'll just need a 470Ω VR and two to three 1.5V dry cells, a a jumper similar to those commonly found on mother board for option selection. This makes up you test gear.

    With the arrangement below, you can set the voltage to simulate over/under voltage.

    Removes the jumper and plug in the test gear. Set one trip point then remove the test gear and plug it in the other way. You can then have 12V +/-4.5V to represent real life AC voltage.

    [​IMG]
     
  9. sakbd.2010

    Thread Starter Member

    Sep 19, 2010
    39
    0
    hello chung,

    Thanks and sorry for late reply.

    I see what u'r saying. The way u'r setting it up, the 4.5v or 5v shall be added to the vdd or substracted...ingenios.

    In case of 3 dry cell's I can also use my extra computer PSU which supply both +12v and +5v.

    I got another problem, I brought a wrong transformer, its a 12vx2(no center tap) and it's total is 24v, guess cannot use to make the supply(unregulated 12v), I need a 6vx2 at 1amp.

    Meanwhile I can make the circuit and use my PC PSU to supply +12vdc and with your 470 vr/preset the +5v(from the PSU), would that work...what ya think ?...let me know

    Also if I use one of those PNP transistors i.e. BD140, what pin will go to what side ? U can just tell me, no need to draw if u don't want(base-collector-emitter).
     
  10. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    It is even better to have a 12Vx2 transformer.

    As the detecting circuit draws only a few mA so you can get away using a 7812 three terminal regulator to get a stable 12V. But when the relay operates, it will pull down the voltage at the input of the voltage regulator(7812) to about +14V or less. This is not good if you base your sensing voltage on this voltage.

    Therefore you can use the remaining winding alone to provide the sense voltage. This will ensure that the sense voltage really reflects the true voltage of the AC side. Notice the value Cx is not finalized yet as I need to do some simulation to decide its best value.

    Sadly you can't use the two voltages from the PSU in this way as they have common 0V and is joined inside the PSU. Simply speaking, you can't get +17V from the PSU.

    You need to use two to three dry cells or if the circuit is powered by AC, then you can use only the 5V output of the PSU to replace the dry cells.

    Not just any PNP transistor will work as the drive current from the LM393 is limited to a few mAs. That's why I have used a P-Ch MOSFET which require only a large voltage to drive. Alternatively, you can use a type of transistor called Darlington, which in effect is two transistors connected inside a single package. The BD140 is NOT a Darlington transistor.

    [​IMG]
     
  11. sakbd.2010

    Thread Starter Member

    Sep 19, 2010
    39
    0
    OK, After reading your last replay...

    I was going through my junk box and found something...

    It's A614Y(KSA614)...will they work..as a PNP Darlington..let me know..

    can u please give a bit more details...on the sensing voltage connection.."Therefore you can use the remaining winding alone to provide the sense voltage" --how do u think u do this part...I'm not clear on this yet..
     
    Last edited: Oct 30, 2010
  12. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    It is not a pretty picture to be building that circuit with junk box parts...

    Can a normal PNP works as a darlington? Maybe it will if its hfe is high enough. The only to find out is to try using it in circuit.

    The circuit is meant to detect over/undervoltage. So it need to see a voltage that is related to the AC voltage by a certain ratio. If that ratio changes, then what the circuit sees would not represent the actual AC voltage value. Any current drawn by the circuit will reduce the rectified voltage so the circuit would see it as undervoltage.

    A separate winding will ensure the smallest error in achieving that goal.
     
  13. sakbd.2010

    Thread Starter Member

    Sep 19, 2010
    39
    0
    Hi,

    Just read your reply...

    So the real power supply is the unregulated one u hv drawn in the bottom, right!!

    Separate winding meaning multitap transformer or two transformer, if we are gonna use multitap or two transformer, what is the method to connect them in the circuit.

    or you just meant the bottom rectifier circuit as a means to only test the circuit! not sure about your plan. please clarify..

    thanks
     
  14. sakbd.2010

    Thread Starter Member

    Sep 19, 2010
    39
    0
    got another question regarding capacitor C2 :

    You wrote '103n' , did u mean code-103, which means .01uF PPC or Ceramic or plastic. There is no 103nF capacitor in the market where i searched.
     
  15. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    A single transformer with two separate windings. e.g. 12Vx2

    Some transformer comes with one winding on the secondary side, some with two separate windings(total 4 connections or wires) and some comes with two windings but only three connections or wires(center tapped).

    Make sure you know the difference between them. You would want to use one with two separate windings and have four wires/connection on the secondary side.

    Sorry the 103n capacitor is actually 103, or 10nF or 0.01uF.
     
  16. sakbd.2010

    Thread Starter Member

    Sep 19, 2010
    39
    0
    OK, understood, so you are going to connect both secondary windings with the same +12vdc rail and ground at the same time ...

    So, I'm going to connect both secondaries in parallel with the circuit, one regulated via 7812 another without regulation...regulated one supplying steady +12vdc AND the unregulated one unregulated +17vdc...

    OR, Regulated one normal connection but the unregulated one through the jumper as per your drawing...just like the 3 drycell batteries.

    if not how are they be connected what are the connection points..please clarify...

    for the life of me I cannot really understand the logic behind this...may be I'm too thick at power supply thing...

    Hope I'm not making it too dull for your likings...sorry if I'am but I must understand it before I build this power supply..
     
    Last edited: Oct 31, 2010
  17. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    Not quite. Ground yes, but not the +12V rail.

    One winding is especially set aside for voltage sensing only. The other winding provide the +12V supply to your LM393 and LM555.

    Not in parallel, only the GND of both supply are connected together. See below. Only two signals from the red block goes to the main circuit.

    [​IMG]
     
  18. sakbd.2010

    Thread Starter Member

    Sep 19, 2010
    39
    0
    Hi chung.,

    Thanks for the clear explanation...understood every detail.

    One last thing...

    IF I cannot get a 4 wire two secondary winding transformer...Can I get two separate 12v transformer ???...

    Note two got two separate primary and secondary..

    Can we join their ground(after rectifier circuit) if need be or can we keep them separate.

    Do they really need common ground??...let me know thanks.
     
  19. sakbd.2010

    Thread Starter Member

    Sep 19, 2010
    39
    0
    More..

    On your red block diagram...the Capacitor Cx...is it 470uF?...

    You said the last time U need to simulate to figure the best value...could you do that and figured the Best capacitor value ?? Let me know.
     
  20. eblc1388

    Senior Member

    Nov 28, 2008
    1,542
    102
    Of course you can use two separate transformers but two windings secondary transformer are very common, as common as center-tapped ones.

    I lost you here. Can you say a bit more.

    See my post#15 about transformers.

    The value of Cx depends on what sort of transformer output voltage you will use.
     
Loading...