To start with I'm a little rusty with electronics, I haven't done much in 20 years and it was only a hobby back then.

In my car I have a head unit with a touch screen which relies on software to run it.

My problem is that it starts to load the software when you turn the key to accessories but when you go to start the car it cuts power to the head unit which hasn't finished loading the software. After the car has started (key returned to the on position) the power is restored to the head unit and it tries to load the software again.

This cut in power, I believe, is causing the head unit to malfunction, it messes with the software.

So I need to supply power to the unit when the key is turned to ACC or On position but held on while the engine starts. After the engine has started the key will return to the On position and power the unit as normal. When the car is turned off then the unit needs to turn off.

I was thinking of powering the unit through a relay wired direct to the battery and then using the signal from the ignition, in parallel with a timer, to trigger the relay. If I use the loss of power when starting the engine to trigger the timer then the unit has already lost power and it would start loading again, so that would be of no use.

I tried to use the initial connection to power (when the key is first turned to ACC) to trigger the timer. If I am correct then with my circuit the timer is put into an endless 10 sec loop while the key is turned on and when the key is turned off only the timer powers the unit till the end of the time period. If the key is returned to the On or Acc before the end of the time period then the unit will stay on but if the ignition is turned off for longer than the time period then the timer stops and the unit turns off.

(This is also a diesel so needs the ignition to be turned on for a period of time before starting. I'm not sure if 10 secs is enough time but I don't think you can stretch it out further).

enter image description here

Can anybody verify if this circuit will work?
Is there a better way to do this?
If this is the best choice, is there anything wrong and what should I do to fix it?
(I wasn't sure if the 10k and 1k resistors connected to the base of the transistor were needed).

  • 5
    \$\begingroup\$ I don't know modern cars overly well, but my good ole Punto doesn't cut power – the battery is simply too weak to not take a significant voltage drop when the starter motor is used. \$\endgroup\$ Feb 10, 2019 at 8:47
  • 1
    \$\begingroup\$ I know that modules for the automotive industry are specified to be working when they take the voltage droop and oscillation of the ignition. (called cold cranking ) If I remember correctly a 12V can drop down to 4-5V. \$\endgroup\$ Feb 10, 2019 at 9:02
  • \$\begingroup\$ newelectronics.co.uk/article-images/image-library/93/… \$\endgroup\$ Feb 10, 2019 at 9:02
  • 1
    \$\begingroup\$ I doubt that the car deliberately cuts power to the radio. the issue you have is caused by the drop in voltage that happens when you start the engine. This drop make so that the radio unit either resets, or continues to boot with corrupted memory data. What I would do is use a 555, to delay the boot up of the radio to shortly after the engine is running. \$\endgroup\$
    – Elmesito
    Feb 10, 2019 at 9:52
  • 1
    \$\begingroup\$ A diode and some big capacitors will fix it. \$\endgroup\$
    – Indraneel
    Feb 10, 2019 at 10:07

5 Answers 5


I don't think this is going to work.

When your engine cranks it's typical to see a large drop in voltage supplied by the battery. Even if you keep the radio from turning off deliberately you're still going to see problems with brownouts. Most circuits do better if they're turned fully off than if they're dealing with this sort of brown-out and I think that's what the wiring in your car is trying to protect you from.

Does the head unit have a manual?

  • 1
    \$\begingroup\$ I didn't think about brownouts, thanks for pointing that out it's something to consider. My cars have always been good as far as that goes so I didn't think about it. My last car was an 80 series Landcruiser, it had the earlier version of the same radio and had no problems. But it wasn't wired to drop power to the radio when you started the engine, so I didn't have this problem. As for a manual yes there is one but it doesn't tell you much. \$\endgroup\$
    – Berallan
    Feb 10, 2019 at 10:41
  • \$\begingroup\$ By the way, you don't need the feedback from the output to the trigger. Or the feedforward from the ACC to output. The discharge pin should be all the feedback you need to create your monostable circuit with the 555. That should simplify your input logic. and let you remove the diodes and most of your combinational logic. \$\endgroup\$ Feb 10, 2019 at 21:55
  • \$\begingroup\$ Thanks Andrew, even if this isn't the right solution for my problem due to brownouts I would still like to know if it would work and where my mistakes are. \$\endgroup\$
    – Berallan
    Feb 11, 2019 at 0:32

Would it be possible to add some kind of isolated power source for the head unit, so it can receive uninterrupted power, even when there's a large voltage drop across the car battery. This could just be a small rechargeable lipo battery. You'd just need some kind of circuit to detect a voltage drop on the main source, and switch temporarily to the backup battery. The backup battery could be very small, as it only needs to provide power for a couple of seconds while the engine starts.

  • \$\begingroup\$ Possibly but I would have to give some thought to charging and switching as I wouldn't want a lipo to catch fire. But I do have 12v lead acid battery about the size of a 6v dolphin torch battery that I might be able to do something with. \$\endgroup\$
    – Berallan
    Feb 10, 2019 at 10:47

Here's a simple relay solution that might be worth investigation.


simulate this circuit – Schematic created using CircuitLab

Figure 1. The two relays hold off radio power until the starter cycle has finished.

How it works:

  • With ignition off both relays are de-energised. Radio is off.
  • When ignition is turned on radio power is still off as RLY1 contact is open.
  • When starter is energised both relays pick up and RLY1 latches. RLY2 is open and radio power is still off.
  • When starter is released RLY1 remains energised over its own contact while D1 prevents back-feed into the starter circuit. Meanwhile RLY2 drops out and power is applied to the radio.
  • Turning off the ignition re-initialises the complete circuit.

Note that this relies on RLY1 picking before the battery voltage drops due to starter solenoid pulling in causing the battery voltage to droop and RLY1 must remain energised at the low voltage. This should be possible so if it doesn't work with one relay then try another of a different type. If you have an adjustable power supply you could measure the pick and release voltage for each relay before testing in the car.

  • \$\begingroup\$ Assuming the relays worked as you intended, I can see how this would work but the radio would only work when the engine was running. But I do have an idea on how to solve that, Bridge the contacts on RLY1 with a momentary switch. Would have to remember to turn the key off before starting after using the switch. I wish the power button on the radio actually turned off the unit instead of putting it in standby or just turning off the screen, whichever it does. The momentary switch would also fix the issue if RLY1 did drop out. \$\endgroup\$
    – Berallan
    Feb 10, 2019 at 10:53
  • \$\begingroup\$ If you do come back to this post I had a question for you about another post that you replied to. electronics.stackexchange.com/questions/215418/… Can the LM317 current limiter be added to a car to charge a second battery at a different rate to the main battery? \$\endgroup\$
    – Berallan
    Feb 11, 2019 at 0:56
  • \$\begingroup\$ That has nothing to do with the question at the top of this page so we shouldn't be discussing it here. In any case, no, the LM317 requires several volts headroom so its output voltage must be at least several volts below its input voltage. \$\endgroup\$
    – Transistor
    Feb 11, 2019 at 7:19

It looks like this circuit would be pulling around 800mA through the coil and BJT which would probably blow both of them. Put a current limiting resistor in there.

I am skeptical that the car cuts power to the head unit unless you've confirmed this somehow. I think it would be more likely just being browned out when the car is starting.. A new battery with a better C rating (lower output impedance) might fix your problem if it was originally designed to work under starting loads. It could just be drooping more than expected. Was this always a problem?

The simplest solution I can think of, assuming the radio is browning out and not switched off intentionally is to just place a smaller 12V lead acid in line with the head unit power with a diode preventing current from going backwards to the starter motor. It will still stay charged and everything from the alternator so it only has to have enough capacity to run the radio while starting. Spec it for at least 3-4 times this though. Its maybe a little ghetto but should work.

The timer solution should would work too, although I haven't checked your circuit

  • \$\begingroup\$ I did have concerns as to power consumption of the circuit. Only had the car 18 months and from day one everything that isn't essential to starting the engine cuts out when you start it. Instruments and warning lights still have power. \$\endgroup\$
    – Berallan
    Feb 10, 2019 at 11:05
  • \$\begingroup\$ Just checked the small lead acid battery that I have, it's a 12v 7Ah. probably would do but don't know if the alternator would put too much charge into it. The alternator is something like 150A, not sure exactly but I know it's over 100A. \$\endgroup\$
    – Berallan
    Feb 10, 2019 at 11:54
  • 1
    \$\begingroup\$ Vehicles do deliberately cut power to all non-essential services while starting. The power conditioning required on the unswitched (battery) system is significant (and from the vendor's perspective, expensive). During cold cranking, the battery voltage my droop by up to 8V (quite apart from other issues); the switched system is cut off (cleanly) during such events. \$\endgroup\$ Feb 10, 2019 at 11:57
  • \$\begingroup\$ It probably is cutting power to it then. The battery / diode could still work but you'd have to put it in the right place in the circuit. Mainly it has to be after the starter motor in the circuit and somewhere that the battery will be switched off when the vehicle is off. 7Ah is a good size. To be safe you'd want to use a fairly beefy shottkey diode. You don't need to worry about the alternator overcharging it as the alternator should only produce voltages that are safe to be put on a lead acid. \$\endgroup\$ Feb 10, 2019 at 20:53
  • \$\begingroup\$ If you do this, charge the 7Ah guy first. If his voltage is too low the main battery will try to push too much current on him and blow the diode or damage your battery. If its charged first it should be fine since current cant go in the other direction. Eventually the two batteries will find an equilibrium. \$\endgroup\$ Feb 10, 2019 at 20:56

Thank you all for your assistance.

Originally the unit appeared to be working correctly with the exception of erratic behaviour which I thought was software related. It seems that after more time has passed some new symptoms have shown up and what the unit was doing is now happening a lot less frequently. I am thinking that I was wrong as to the cause of my problem.

This unit has four touch switches down the side similar to to what many early smart phones had at the bottom. Three of these have now stopped working. The interesting thing is that the three that have stopped working are the Return to previous screen, Mute and Volume Up (the Volume Down still works fine). The problems that I thought were software related were the unit going back to the home screen for no reason, the unit muting and un-muting itself, the volume would rise on it's own but never drop and the touch screen would freeze or sometimes think you were on the home screen when you were on a different screen, or it thought you were holding your finger on the screen when you only tapped it.

It seems too much of a coincidence that the actions of the unit just happen to be the same as the buttons that have stopped working for them not to be related. I'm not sure how the screen freezing and the other issues relate to the buttons but maybe the unit was overwhelmed by the signals from the faulty buttons.

I think the solution is to do one of three things

  1. disconnect the faulty buttons and hope the unit doesn't pickup noise due to the missing buttons (the unit has other ways of doing the same things)
  2. replace the faulty buttons
  3. replace the unit (after my frustration has been satisfied :) ).

In any case thank you all and sorry for bothering you with something that probably wouldn't have helped.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.