Time delay with 2 Transistors

Thread Starter

cevval

Joined Dec 6, 2011
63
The purpose is to delay relay (k1.1) action, a couple of seconds after base of transistor Q2 is energized depending on capacitor C6.
The circuit diagram is as attached.
It seems to work on breadboard. However,
Here are my questions:
1) Is it reliable with respect to life of components?
2) Is it needed to have a diode just before C6 connection?
3) You can also make comment on the value of base resistor (R13) of transistor Q1.
Your help will be highly appreciated.
Thanks, in advance.

cevval_2TransistorCirc2.png

Moderators note : removed unnessesary white space
 
Last edited by a moderator:

MrAl

Joined Jun 17, 2014
11,342
The purpose is to delay relay (k1.1) action, a couple of seconds after base of transistor Q2 is energized depending on capacitor C6.
The circuit diagram is as attached.
It seems to work on breadboard. However,
Here are my questions:
1) Is it reliable with respect to life of components?
2) Is it needed to have a diode just before C6 connection?
3) You can also make comment on the value of base resistor (R13) of transistor Q1.
Your help will be highly appreciated.
Thanks, in advance.

View attachment 135847

Moderators note : removed unnessesary white space

Hello,

First, you should have a collector resistor so that the transistor does not have to be over taxed when turned on when C6 is discharged.

Second, when the input is more than about 6v and the cap charges up and then Q2 is turned off by bringing the base to ground, there will be a reverse bias on the emitter base diode equal to the voltage on the cap. Thats not a good idea, so keeping the input at less than 6v is a good idea.

Third, you have to make sure that your relay turns off fast enough. If it doesnt, it could cause excessive arcing across the contacts which leads to pitting and early failure. The reason it may turn off too slow is because of the exponential time constant of the cap and R13. A Schmitt trigger would have been better but it does depend on how much current you are putting through the relay contacts also.
To get it to open (or close) faster, you can put a diode in series with the transistor Q1 emitter as long as the relay still operates normally. The diode forces the turn off action to occur higher up on the exponential curve and thus it has a steep slope and thus the transistor turns off cleaner and thus the relay turns off faster. Keeping the diode biased with a resistor to +Vcc helps too to keep the diode voltage somewhere near nominal.
 

Thread Starter

cevval

Joined Dec 6, 2011
63
Hello,

First, you should have a collector resistor so that the transistor does not have to be over taxed when turned on when C6 is discharged.

Second, when the input is more than about 6v and the cap charges up and then Q2 is turned off by bringing the base to ground, there will be a reverse bias on the emitter base diode equal to the voltage on the cap. Thats not a good idea, so keeping the input at less than 6v is a good idea.

Third, you have to make sure that your relay turns off fast enough. If it doesnt, it could cause excessive arcing across the contacts which leads to pitting and early failure. The reason it may turn off too slow is because of the exponential time constant of the cap and R13. A Schmitt trigger would have been better but it does depend on how much current you are putting through the relay contacts also.
To get it to open (or close) faster, you can put a diode in series with the transistor Q1 emitter as long as the relay still operates normally. The diode forces the turn off action to occur higher up on the exponential curve and thus it has a steep slope and thus the transistor turns off cleaner and thus the relay turns off faster. Keeping the diode biased with a resistor to +Vcc helps too to keep the diode voltage somewhere near nominal.
Hi, Thank you very much for your detailed very technical explanations. I understand that your an electronics guru, where as I am just an amateur hobiest electronics guy.

1. Can you advice the value of a such resistor? Please. I assume it to be around 100 ohm. What you think?
2. From your second comment, I understand so: I put a 6.8V zener diode before Q2 collector. Pls. confirm this follows your recommendation. Please, see below revised diagram.

3. The relay capacity is 5A max. AC, however the load is not more than 700mA DC.
Sorry, I could not figure out your description and could not put on the revised diagram. If you can show this on a sketch, it will be most appreciated.
This shown revised circuit diagram seems to work on breadboard (not able to confirm relay speed improvement).
On the other hand, as I got the information from internet that, putting a 12V zener diode in series with relay clamping diode reduces relay turn-off (release) speed time one third.
2TransistorCirc_R1.png
What do you think about this?
 

Attachments

MrAl

Joined Jun 17, 2014
11,342
Hi again,

What are all those zener diodes for?

No problem that you are a hobbiest. If you are willing to learn you'll learn more than you ever thought possible i bet and have some fun doing it :)
 

Thread Starter

cevval

Joined Dec 6, 2011
63
Hi again,

What are all those zener diodes for?

No problem that you are a hobbiest. If you are willing to learn you'll learn more than you ever thought possible i bet and have some fun doing it :)
Zener D1 is for to reduce voltage around to 6V which you mentioned as second in your reply.
Zener D2 is for to get relay to open (or close) faster, as I get this information from internet, because I couldn't figure out what you mean in last paragraph in your comments.
 

Tonyr1084

Joined Sep 24, 2015
7,829
Well, here's a delay start rig using a relay I made a number of years ago. Used it on my truck. When I started the truck, my gauges with memory would come on, then when I cranked the engine the gauges would blink out. One of the gauges would not tolerate the interruption and would reset the color to default. So I decided to build a delay start rig so that it wouldn't turn the gauges on for four seconds (approximately). Notice I used a 24 volt coil relay. It worked just fine on 12 volts. Actually, the real voltage was likely 14.4 after starting, then around 13.8 during driving. And I used all stuff I've scavenged from other items. The relay was out of an old dish washer.

When power switched on, C1 (I know, not labeled - but it's pretty prominent) would take a charge through R1. When the voltage was sufficient in C1, current would then flow into the base of Q1, turning on the relay. When the relay energized it would self latch using voltage from the normally open contact through a diode, and thus, would remain on regardless of what the transistor was doing. When switching the vehicle off, the BEMF (Back ElectroMotive Force - or "Flyback") would be absorbed through D2.

[edit] IMAGE DELETED by - um - me.
 

Attachments

Last edited:

Tonyr1084

Joined Sep 24, 2015
7,829
Wait a minute! You want the relay to stay on for a few seconds after power is disrupted? What's to keep the relay active if that's the case? I must have missed something.

Looks like 12 - 15 VDC is always present. Power at R2 controls the relay. Am I reading that correctly?
 
Last edited:

ScottWang

Joined Aug 23, 2012
7,397
The below circuit is fade in and fade out for LED (or turn on/off for relay delay) about 3~4 seconds when the sw1 turn on and turn off.
 

MrAl

Joined Jun 17, 2014
11,342
Zener D1 is for to reduce voltage around to 6V which you mentioned as second in your reply.
Zener D2 is for to get relay to open (or close) faster, as I get this information from internet, because I couldn't figure out what you mean in last paragraph in your comments.
Hi,

Yes a diode and zener of 12v across the relay coil will allow the magnetizing current to decrease roughly 12 times faster than with a diode alone so that part is a good improvement.

There are other things that can be changed to improve the circuit too, but i would need to know the level of your drive signal that drives Q2 through R2. What kind of signal will be applied there? I need the min and max values. Also, if there is some flexibility based on what will drive that it would be good to state here too. For example, do you have to have the cap change when the input level is high, or can you change it so that it charges when the input signal goes low instead? I good improvement might come easily if that is possible.

The slow turnoff i was referring to is not improved much by the extra zener across the relay. The low turnoff i was referring too comes from the transistor Q1 turning off too slow because of the slowly discharging capacitor voltage. This means some improvement would be good there too and there are a couple ways to do this.

Another issue is the size fo the cap itself and the series resistor going to Q1. The time constant should be long enough to get the required time delay you want. Assuming 2 seconds delay and if we approximate the time as 4 time constants, then the required value is 104uf so 100uf seems reasonable for now, however this may have to be increased to 200uf later once we implement the improvement for the snap off of transistor Q1.

The most important question first though is what kind of drive are you using now, and how much can you change that if you need too, such as inverting the logic signal there?
 

Thread Starter

cevval

Joined Dec 6, 2011
63
Wait a minute! You want the relay to stay on for a few seconds after power is disrupted? What's to keep the relay active if that's the case? I must have missed something.

Looks like 12 - 15 VDC is always present. Power at R2 controls the relay. Am I reading that correctly?
Yes, you are reading correctly. Normaly, when no volt at R2, relay is off position.
 

Tonyr1084

Joined Sep 24, 2015
7,829
OK, then you can ignore my post with the delay start relay.

Are you tied to the idea of using transistors? My solution, if you are not firmly tied to using transistors, would be to use a MOSFET. Sorry, I'm not real experienced with them, but an input at R2 can charge a capacitor. When the input goes away the capacitor can hold the MOSFET on until the voltage in the cap has discharged low enough to switch the relay off. The thing I'm just learning myself about MOSFETs is that they can act like a switch OR like a resistor. With the capacitor discharging the FET will go from acting like a switch to acting like a resistor. When it's acting like a resistor it has to dissipate sufficient power, so you need to select one that can handle that extra heat. When the cap loses its charge the FET will be completely off. The neat thing with the relay is that the contacts don't suffer changing voltage. The relay switch is on until it's off.

Let us know if you can consider an alternative design.

In your initial drawing, you might change Q2 from a BJT (Bipolar Junction Transistor) to a MOSFET. Q1 won't draw as much power as the relay does, so you can continue using Q1 to activate and deactivate the relay. I don't have time to make a drawing right now. Maybe later. Depending on your response. :)

OH! Also, what signal level is coming in at R2? Voltage? Current (if applicable)? What's the control signal source, how is it being triggered?
 
Last edited:

Tonyr1084

Joined Sep 24, 2015
7,829
Well, here's my shot at this problem:

Delay OFF.jpg

I've left components and values out of the picture. You can engineer your own circuit values. The time constant of R1C1 will set your delay off time. Q1, a MOSFET, will draw no current. D1 will prevent any back feed to your controlling circuit (Signal). Q2 is turned on by Q1. R2 limits the current through both Q1&2. Q2 turns on the relay. When "Signal" is lost (goes low) R1C1 holds the gate of Q1 high for a set duration. When C1 is discharged Q1 will shut down Q2 and the relay will fall out.

One big part of why I've left out component types and values is because I'm not real familiar with all the types of components and what would best suit the purpose. I'd have to do a lot of research to decide what MOSFET I'd use. As far as Q2, I don't see any reason why it can't be a 2N2222. If your signal comes from a current sensitive device then you'll want to add a resistor in series with D1 so that when it sees the fully discharged C1 you won't draw too much current from your signal source.

Hope this is something to consider.

Tony
 
Last edited:

MrAl

Joined Jun 17, 2014
11,342
Hi,

I was asking what kind of signal we have to work with, that drives the input. We need to know this so that we can properly interface that to the delay circuit. As it stands now, we dont know the voltage logic levels nor the current available. We need to know both to be sure the new modified circuit will work properly.
 

Thread Starter

cevval

Joined Dec 6, 2011
63
Hi,

I was asking what kind of signal we have to work with, that drives the input. We need to know this so that we can properly interface that to the delay circuit. As it stands now, we dont know the voltage logic levels nor the current available. We need to know both to be sure the new modified circuit will work properly.
Hi again,
Sorry for delayed reply. Life must go on and other things needed to be done.
1. First good result is that; additional zener diode across the relay coil is for sure to be there always, but not in my case, it is a different mater and that is Q1 operation time, which I understand now.
2. Coming to your questions asked in previous reply; I try to answer as much as what I understand from your very technical expressions.
a) R2 is connected to LM 723 VO terminal but there is a 6,8V zener in between. Most probably, VO=VCC, and from datasheet I= 25mA max. Well, I have already tested with a cap in between zener and R2. Not worked as I wished. I also tested, by putting a diode and a cap on LM 723 VCC line and Input . Not worked also.
b) As in my sketch, Cap value of 100uF is good enough for my case.
Summary of my case story is that: This is a low voltage level sensing circuit. If voltage supply level drops to set critical level VO goes to low. But, during normal operation, due to some other loads on same power supply, when one of it comes on, I believe that input signal voltage (v+) fluctuates below the set voltage for a very short period, in this case relay off and on again momentarily, which I don't want.
Maybe, there is a way to slow response time of LM 723, which I can not manage.
 

MrAl

Joined Jun 17, 2014
11,342
Hello,

So what is the output voltage of the 723 then?
A zener in series will prevent the cap from charging up too high, but i have a feeling a common emitter circuit would do better as then you could let the charge charge higher.
 

Thread Starter

cevval

Joined Dec 6, 2011
63
Well, here's my shot at this problem:

View attachment 136342

I've left components and values out of the picture. You can engineer your own circuit values. The time constant of R1C1 will set your delay off time. Q1, a MOSFET, will draw no current. D1 will prevent any back feed to your controlling circuit (Signal). Q2 is turned on by Q1. R2 limits the current through both Q1&2. Q2 turns on the relay. When "Signal" is lost (goes low) R1C1 holds the gate of Q1 high for a set duration. When C1 is discharged Q1 will shut down Q2 and the relay will fall out.

One big part of why I've left out component types and values is because I'm not real familiar with all the types of components and what would best suit the purpose. I'd have to do a lot of research to decide what MOSFET I'd use. As far as Q2, I don't see any reason why it can't be a 2N2222. If your signal comes from a current sensitive device then you'll want to add a resistor in series with D1 so that when it sees the fully discharged C1 you won't draw too much current from your signal source.

Hope this is something to consider.

Tony
Thanks, Tony. It seems a better solution because, less components.
Signal source gives max. 25mA. I am not very familiar with mosfets. However, being an amateur hobbiest electronic guy with no education get on electronics, Can you advise component types and values as much as possible?
 

Tonyr1084

Joined Sep 24, 2015
7,829
I'm kind of like you. I know how these things work (albeit not all that well) but when it comes to knowing what components to use, such as a "Logic Level" MOSFET or other - I have practically no experience. I've messed with MOSFETS and therefore understand them a little, but you'd be better served if someone else could answer your question. As far as the RC timing part of it - again, it's a lack of experience on my part. I might bodge something together on my bench and see how it works. If it's close I will make adjustments. Sometimes they just don't work at all and I'm left wondering what went wrong. Often it's just my numbers. But with practice I'm getting better.

Another solution (instead of the MOSFET and transistor) could be to use an IGBT (Insulated Gate Bipolar Transistor), sort of a hybrid MOSFET and BJT (transistor). In fact, I think I have an IGBT here on my bench somewhere, but they look like any other transistor. I'd have to find it and identify it by a spec sheet. I'm sure I put it some place safe. Safe from me even.
 
Top