Simple power on / power interrupt trigger suggestion?

Thread Starter

Juhahoo

Joined Jun 3, 2019
226
I have a 12V car battery, and the device I'm implementing requests 3 pins, + and - and positive enable pin and this enable should connect after the + and - is connected so there should be a delay of some time. But because I want to use crocodile clips there is no chance for 3rd pin. So I try to implement a circuit that does the job for the enable pin automatically. Do you have good ideas?

Below is my fast sketch as a suggestion. It has a analog enable output done by RC which is fine.
When + is connected, C2 is fast charged, T2 pulls T1 base low, C1 starts charging.
When + is disconnected T2 no longer conducts, C2 will discharge via R2, R5 and T1 rapidly discharging C1.

1628675405742.png
 

LowQCab

Joined Nov 6, 2012
1,189
How do You know that a delay is required ?
How long must the delay be ?
It looks like your Enable Pin needs a Switched-Ground with a Pull-Up-Resistor,
is this correct ?
.
.
.
 

Thread Starter

Juhahoo

Joined Jun 3, 2019
226
This is a similar circuit that is to be implemented, ignore the 5V in this case, it will be 12V. I need to replace the "short" pin function by automatic "delay". The delay is supposed to prevent unknown behavior when everything is powered at the same time. The key is that when supply is interrupted by bad contact etc, the ON pin goes down immediately.


1628683149809.png
 
Last edited:

Tonyr1084

Joined Sep 24, 2015
6,374
Your circuit: Regardless of what C2 does, T2 will hold T1 off. C1 will not charge because T1 will already be holding it low. This will hold EN low almost as soon as you apply 12V. Does EN need to be held low to enable the circuit you want to delay the EN signal to?
 

LowQCab

Joined Nov 6, 2012
1,189
I don't understand.
You show a Circuit and then say that it's only "similar" to this Circuit.

What problem are You trying to solve ?
Do You have a proposed Schematic diagram,
and the actual part-numbers that You want to use ?
.
.
.
 

Thread Starter

Juhahoo

Joined Jun 3, 2019
226
I don't understand.
You show a Circuit and then say that it's only "similar" to this Circuit.

What problem are You trying to solve ?
Do You have a proposed Schematic diagram,
and the actual part-numbers that You want to use ?
.
.
.
Problem is written earlier. Its the supposed 3 pin connector with one short pin to have late contact. This should be removed and replaced with automatic delay.

Lets say this is THE circuit i'm going to use with 12V
 

Thread Starter

Juhahoo

Joined Jun 3, 2019
226
Your circuit: Regardless of what C2 does, T2 will hold T1 off. C1 will not charge because T1 will already be holding it low. This will hold EN low almost as soon as you apply 12V. Does EN need to be held low to enable the circuit you want to delay the EN signal to?
When T1 conducts, T2 is not, this allows C1 to charge.
Enable is positive active and LTC4211 has internal threshold for it.
 

Tonyr1084

Joined Sep 24, 2015
6,374
When T1 conducts, T2 is not, this allows C1 to charge.
Enable is positive active and LTC4211 has internal threshold for it.
I disagree. When 12V is applied, R3 & R4 provide some sort of voltage divider for the base of T2. Let's assume the current on T2 is sufficient to turn it on. When T2 turns on (immediately when 12V is applied) the collector will pull the base of T1 low, preventing it from turning on. C1 will charge. What you need EN to do? Go high? Or go low? Until I'm sure of that I can't say for sure how your circuit will work out.

I'm not following how you are coming to the conclusion that when T1 conducts T2 can't. I see some voltage going to the base of T2. This voltage, depending on the resistors in circuit, will result in a current that turns T2 on. When T2 goes active T1 will be hard grounded at its base pin. That will allow C1 to charge, and EN will go high. If that's the intent, EN going high then you don't need all the extra components. But this is much easier to do with a comparator. Whether you want a high or a low at EN - the comparator can be easily configured to give you either condition. Just a couple resistors and a capacitor.
 

Thread Starter

Juhahoo

Joined Jun 3, 2019
226
I disagree. When 12V is applied, R3 & R4 provide some sort of voltage divider for the base of T2. Let's assume the current on T2 is sufficient to turn it on. When T2 turns on (immediately when 12V is applied) the collector will pull the base of T1 low, preventing it from turning on. C1 will charge. What you need EN to do? Go high? Or go low? Until I'm sure of that I can't say for sure how your circuit will work out.

I'm not following how you are coming to the conclusion that when T1 conducts T2 can't. I see some voltage going to the base of T2. This voltage, depending on the resistors in circuit, will result in a current that turns T2 on. When T2 goes active T1 will be hard grounded at its base pin. That will allow C1 to charge, and EN will go high. If that's the intent, EN going high then you don't need all the extra components. But this is much easier to do with a comparator. Whether you want a high or a low at EN - the comparator can be easily configured to give you either condition. Just a couple resistors and a capacitor.
Enable is HIGH and R1*C1 is the timing for it. The thing here is the fast discharge of C1 when power is interrupted and reconnected in rapid pace. C2 will hold energy for driving T2 base in power interruption and discharging C1. D1 and D2 is there to prevent T2 of getting energy from C1 and C2.
If you don't have fast C1 discharge, C1 is already charged if power is returned quickly after disconnecting and there will be no timing.
 
Last edited:

Tonyr1084

Joined Sep 24, 2015
6,374
This is how I'd go about the delay: Option 1 is adjustable, Option 2 is a fixed delay. The delay is set by R3/C1 and VR1 selects the point at which the Comparator switches from low to high. When the voltage at the inverting pin ( - ) is set at 50% of the supply voltage (12 volts) via VR1 C1 starts charging through R3. As the voltage on the non-inverting pin ( + ) goes higher than the set voltage from VR1 EN switches from low to high.
1628691298053.png
I think you briefly mentioned an interruption in power. Choosing too high of R3/C1 values will retain their state possibly too long for your liking. This may cause unexpected behavior. Adding a transistor to short C1 will eliminate that issue, but the transistor needs to be rated to handle the sudden rush of current from discharging C1. Going too low on VR1 or R1 & R2 will draw more current than you may want. I'd go with a 10KΩ (VR1) or two 5KΩ resistors (R1 & R2). That way you're never drawing more than 12mA (assuming your 12V is constantly at 12V) through the reference voltage. Higher voltages will result in higher current. R3 should be low enough to quickly charge C1 and C1 should not be so high that it takes a long time to charge up. Smaller values of C1 and higher values of R3 would be preferred IF you go with a shorting resistor to drain C1 in a momentary interruption.
 

Tonyr1084

Joined Sep 24, 2015
6,374
Enable is HIGH and R1*C1 is the timing for it. The thing here is the fast discharge of C1 when power is interrupted and reconnected in rapid pace. C2 will hold energy for driving T2 base in power interruption and discharging C1. D1 and D2 is there to prevent T2 of getting energy from C1 and C2.
If you don't have fast C1 discharge, C1 is already charged if power is returned quickly after disconnecting and there will be no timing.
When power drops out - C2 drives T1 to short C1. OK, I see that now. Looks like you've answered your own question.
Do you have good ideas?
See post 12. It's not complete, but adding a single transistor to short C1 will be easy to add in.
 

Thread Starter

Juhahoo

Joined Jun 3, 2019
226
See post 12. It's not complete, but adding a single transistor to short C1 will be easy to add in.
Sure comparator circuit is very common in triggering devices, but adding a single transistor to this circuit to discharge C1... hmmmm.. I believe you need to come very close to my solution... but surprise me !
The reason I did not want to use op amp is the fact that my circuit is easy to scale higher or lower voltages without any additional components making it very versatile.
 

Tonyr1084

Joined Sep 24, 2015
6,374
Option 3 uses T1 & R4 to discharge C1. R5 is to hold T1 low when there is no power. That's what discharges C1. Your circuit dead shorts C1 to ground through T1. That can blow T1 out. Too much current. Unless you use a really big transistor. So far we haven't talked about any values of components. As for scalability, a mix of option 1 (with VR1) and option 3 should be as scalable as your circuit. I have 8 components; you have 11. Not that component count matters, if you have the space you can use 100 components.
1628692621070.png
You asked for "Good ideas". Well, this is MY "Idea" good or bad.
 
Last edited:

Thread Starter

Juhahoo

Joined Jun 3, 2019
226
Option 3 uses T1 & R4 to discharge C1. R5 is to hold T1 low when there is no power. That's what discharges C1. Your circuit dead shorts C1 to ground through T1. That can blow T1 out. Too much current. Unless you use a really big transistor. So far we haven't talked about any values of components. As for scalability, a mix of option 1 (with VR1) and option 3 should be as scalable as your circuit. I have 8 components; you have 11. Not that component count matters, if you have the space you can use 100 components.
View attachment 245382
You asked for "Good ideas". Well, this is MY "Idea" good or bad.
Isn't C1 feeding T1 base through R3 and op amp input protection diodes? This would effect the speed of the circuit to function.
Scalability has limits, my scale extends beyond any op amp VCC range, so you need additional components.
My circuit is (almost) reverse protected, to do that you need extra components.. so we come closer with component count :)
It's not about the component count really, it is the functionality. If it requires 100 components to do the job, then I go for it.
 

ScottWang

Joined Aug 23, 2012
7,072
Option 3 uses T1 & R4 to discharge C1. R5 is to hold T1 low when there is no power. That's what discharges C1. Your circuit dead shorts C1 to ground through T1. That can blow T1 out. Too much current. Unless you use a really big transistor. So far we haven't talked about any values of components. As for scalability, a mix of option 1 (with VR1) and option 3 should be as scalable as your circuit. I have 8 components; you have 11. Not that component count matters, if you have the space you can use 100 components.
View attachment 245382
You asked for "Good ideas". Well, this is MY "Idea" good or bad.
The Vb of T1 connected to the Vcc directly?
 

Tonyr1084

Joined Sep 24, 2015
6,374
1628697312366.png
The Vb of T1 connected to the Vcc directly?
Educate me - - - Is that a problem? D1 will prevent C1 from back feeding T1. As long as the base is held high (I believe) T1 will not conduct. Only when power is lost will R5 turn the base of T1 on and discharge C1. Am I wrong? Like I said - educate me.
 

Tonyr1084

Joined Sep 24, 2015
6,374
Scalability has limits, my scale extends beyond any op amp VCC range
What's the range you want to possibly extend to? More information will prevent us from giving you bad advice. And when we transcend into areas where I know I don't know - I'll shut up. I'm no expert on this stuff. The biggest killer for me is math. I graduated high school with remedial math skills, and it wasn't until college that anyone (teachers) discovered that I didn't simply know the order of operations. So I've been handicapped mathematically speaking. That's why AC theory was the drop-out point in my studies. Had I attended a better school things would have been different. I still remember the name of the one teacher that used to humiliate any student that got poor grades in math. That was often me - and that turned me off to math altogether. Had I a different experience I would have gone much further in electronics. As it is I'm just a low level hobbiest. But I try harder.
 

Thread Starter

Juhahoo

Joined Jun 3, 2019
226
What's the range you want to possibly extend to? More information will prevent us from giving you bad advice. And when we transcend into areas where I know I don't know - I'll shut up. I'm no expert on this stuff. The biggest killer for me is math. I graduated high school with remedial math skills, and it wasn't until college that anyone (teachers) discovered that I didn't simply know the order of operations. So I've been handicapped mathematically speaking. That's why AC theory was the drop-out point in my studies. Had I attended a better school things would have been different. I still remember the name of the one teacher that used to humiliate any student that got poor grades in math. That was often me - and that turned me off to math altogether. Had I a different experience I would have gone much further in electronics. As it is I'm just a low level hobbiest. But I try harder.
This is targeted for 12V, but I ended up to this circuit so I can extend it up till ??? by properly choosing component values for the range. The other method could be op amp, timer IC or some other more sophisticated system, but simplicity is good.

Thumbs up for your effort, keep learning, keep searching and make mistakes. I cant tell how many things has blown up and malfunctioned because of try hard effort of mine. But there is always something that has worked and gave me satisfaction. You are never ready or fully learned.
 
Top