Please critique my delayed pulse circuit

Thread Starter

MrSoftware

Joined Oct 29, 2013
2,188
If someone(s) would please critique my pulse circuit below, I would appreciate it. The simulated output using generic parts from the simulator is good, hopefully real parts will be close. The goal is; upon powering up a 5-ish second delay then pull the net Pin-4 (at Q1) low for 0.5-ish seconds. The purpose of the circuit is to electrically "push a button" on a car dashboard (by pulling Pin-4 low) after the driver starts the engine. The 5 second delay is to allow the electronics in the car to fully boot before pushing the button. I'll likely use a 556 instead of the pair of 555's, but I'm very open to suggestions for something better.

The power source is a bit of a black box (the car schematic is short on details) but could be a pin from the ECU, so low power consumption is important. The net Pin-4 is an ECU pin pulled-up by an unknown pull-up resistor (estimated by R1 here), so I'm using a mosfet to pull it down.

upload_2019-7-19_10-44-41.png


upload_2019-7-19_10-45-6.png
 

Attachments

jeff3385

Joined Jul 19, 2019
2
Didn’t follow all the logic but some observations:
  • Auto voltage running is closer to 14v
  • Auto voltage is noisy. You could see bounce.
  • Also, test the circuit when cycling power quickly on/off/on.
  • Generally, RC constant for long timing applications is not ideal.
If you're only making one, consider using an arduino or other cheap micro.
In either case, you should condition power for running electronics off 12v auto. Easily sees large voltage spikes.
Why is PR4 voltage changing is your simulation?
 

Thread Starter

MrSoftware

Joined Oct 29, 2013
2,188
Thanks for the replies! I did run the simulation at 14v too and the results are very similar (precision of the timing's isn't critical). Quick on/off is a great point, I'll check how that affects things.

PR4 is actually current draw, I apologize for not making that clearer.

Logic wise; Initially the output to A1 is high, preventing A2 from triggering. At around 5 seconds A1 turns off (dark blue line in the chart), pulling TRI on A2 low which would normally trigger it. However THR is higher than 2/3 VCC so it does not trigger, yet (light blue line on the graph). But now that the A1 output is pulled low, the voltage at RST and THR (PR2) starts dropping as C4 discharges through R6 (the light blue line on the chart). As the voltage to THR drops past by 2/3 VCC, A2 becomes triggered and the A2 output goes high (PR1, the green square wave on the graph). As the voltage into RST continues to fall and passes 0.7v, A2 resets pulling its output (PR1) low again.

The logic pans out in the simulation, but if you have reason to believe that it won't work this way in reality then please do let me know. I haven't used a 555 in a real product before so I definitely welcome any tips from experience.

Edit-> I'm trying to do this without a processor to keep price of components, and price of production low (no programming required).
 

jeff3385

Joined Jul 19, 2019
2
I have no doubt you can tweak this to work in controlled circumstances. Myself, I'd use op-amps and/or counters, flops as first stab if sticking with discrete design.

In an on/off/on situation, you could have C4 not discharged before it gets slammed again with 14v. This pumping action could blow the timer, I think. Good luck!
 

BobaMosfet

Joined Jul 1, 2009
2,110
You can figure out how to do the delay any way you like. Carefully done, with some noise immunity filtering and RC circuit will work fine. It doesn't have to be exact. 5 seconds, or 8, probably doesn't really matter, right?

For a clean pulse, you can always use an opto-isolator with a pull-up on the collector output, to get a nice square wave at digital logic.
 

Picbuster

Joined Dec 2, 2013
1,047
If someone(s) would please critique my pulse circuit below, I would appreciate it. The simulated output using generic parts from the simulator is good, hopefully real parts will be close. The goal is; upon powering up a 5-ish second delay then pull the net Pin-4 (at Q1) low for 0.5-ish seconds. The purpose of the circuit is to electrically "push a button" on a car dashboard (by pulling Pin-4 low) after the driver starts the engine. The 5 second delay is to allow the electronics in the car to fully boot before pushing the button. I'll likely use a 556 instead of the pair of 555's, but I'm very open to suggestions for something better.

The power source is a bit of a black box (the car schematic is short on details) but could be a pin from the ECU, so low power consumption is important. The net Pin-4 is an ECU pin pulled-up by an unknown pull-up resistor (estimated by R1 here), so I'm using a mosfet to pull it down.

View attachment 182037

View attachment 182038
I think that using a Pic 12f with adc will solve your problem.
It will;
- run on 3V run from capacitor when 12V power fails ( time depending from cap size)
-charge capacitor from 3V supply
-power supply 24 to 3v with negative clamp ( start motor pulses) and coil will suppress most of the line crap.
- the pic will measure the voltage and start when needed.
- pulse length fully controlled by pic.
- number of components low lm+pic+fet+coil+a few resistors and caps price pic + lm <two dollar
- program takes you less than 30 minutes.

Picbuster
 

crutschow

Joined Mar 14, 2008
34,284
Here's the LTspice simulation of a delayed pulse circuit using one 555.
The output pulse sets the Trig input high through D1, so the 555 only generates one pulse.

upload_2019-7-20_11-48-26.png
Edit: Restored deleted schematic.
 
Last edited:

Picbuster

Joined Dec 2, 2013
1,047
Not counting the time it takes to understand the operation of a microprocessor and learning how to do programing. :rolleyes:
You are correct however; we talk here about Mr software whom seems to have knowledge about hardware and solder work.
Mr software is very good in my opinion. A little program like the mentioned one takes him less time then scratching his ........... backside.

Picbuster
 

eetech00

Joined Jun 8, 2013
3,859
Hello,

Here's another version. This can also be done with a NE556 or CMOS version of the timers.
I've added a little reverse polarity and overvoltage protection.

DelayedPulseOnPowerUP.png
 

crutschow

Joined Mar 14, 2008
34,284
You are correct however; we talk here about Mr software whom seems to have knowledge about hardware and solder work.
Mr software is very good in my opinion. A little program like the mentioned one takes him less time then scratching his ........... backside.
Yes, I suppose someone with that name would not have a long learning-curve to program a PIC. :oops:
 

Thread Starter

MrSoftware

Joined Oct 29, 2013
2,188
Thanks everyone for the replies! Yes the first ~20 years of my professional career were primarily software, but in this case I would like an all hardware solution with no programming (flashing of chips) step required. Side note; I originally chose the name not to say I'm good at software, but to say I'm not that experienced with hardware. ;)

@crutschow thanks for the single 555 solution! I attempted that but the double timer solution was more obvious so I abandoned it, I'll run your idea through the wringer and see if anything falls off.

@eetech00 thanks for the improvements to the double timer design. I'm going to give it a close look so I can understand what you did and apply your improvements to @crutschow's single timer idea.
 

crutschow

Joined Mar 14, 2008
34,284
Note that one thing not reliably tested by the simulation, is possible false triggering of the 555 during power-up.
If this occurs, a larger C5 capacitor on the CV input should suppress that.

Otherwise an RC circuit to the RST reset input may be needed (R to V+ and C to ground).
Try 100k and 47μF
 
Last edited:
Top