# Using very low voltage from dimmer switch to trigger 12 volt supply

I have a new stereo head unit in my car that auto dims when the headlights are on. The head unit does not have a variable dimmer, it is either on or off. I have a dimmer switch in my car that is used for all of the internal panel lights. I have measured the voltage from the vehicle dimmer switch as follows:

Position 0 is full brightness, position 7 is the dimmest (gray wire): 0 - 0.0 volts, 1 - 0.5 volts, 2 - 1.2 volts, 3 - 1.8 volts, 4 - 2.5 volts, 5 - 3.2 volts, 6 - 3.9 volts, and 7 - 4.6 volts

I measured the voltage of the headlight trigger wire that runs to the head unit and when the car is running it sits around 13.4 to 13.8 volts (orange wire).

In the past I have used relays for situations where I want an event to trigger something else, but through my research I am seeing that relays will require much higher voltage to trigger than I have. So I need a solution that allows for low voltage trigger, approximately 450 mV but, is also able to handle up to 14 volts. The solution also has to be able to handle up to 5 volts on the trigger side and wont lead to failure. I have been looking around and it seems that a transistor might be my solution, although even reading up on transistors and watching a number of YouTube educational videos on the use of transistors, I am still not sure of how to wire one in, and or if it the best option, plus I haven't worked with current switches before.

Ideally I would splice into the gray wire and run that to my switch (part I am here asking about) as the trigger, and then using the power from orange wire (vehicle side) as my power feed, then my output from the switch goes to the head unit.

Thank you in advance for any assistance!

Edit 1: The radio head unit has an input power wire that is used to trigger the internal head unit dimmer. The power wire from the vehicle is at 0 volts when the headlights are off, and then jumps to 13.4-13.8 volts when the headlights are on. So I don't believe the head unit can handle PWM. The dimmer switch in the vehicle has 7 different positions, with position 0 being all internal lights are at full brightness, and position 7 they are at the dimmest. I would like to trigger the head unit to dim at position 1 of the dimmer, but not have any damaging effects, if the dimmer switch were set to position 7. If that makes sense?

• What are you trying to achieve? To get the head unit to dim in response to the dimmer control? Feb 28 '20 at 20:24
• That is correct. Feb 28 '20 at 20:26
• But the radio head unit has only got a digital input - bright or dim. So what is your design specification? Positions 1 to 3 give dim and 4 to 7 give bright? Or do you want to try to make it dim with the dimmer by using PWM (pulse width modulation) if the radio head will handle this? Hit the edit link below your question rather than bury the information in the comments. Feb 28 '20 at 20:31
• How does the head unit know the headlights are on? Feb 28 '20 at 22:23
• The illumination wire (orange) is at 0 volts when the headlights are off and 13.4-13.8 volts when the headlights are on. I probably should have mentioned that the head unit is an aftermarket head unit. The OEM head unit dimmed with the dimmer switch, but this head unit is not designed for variable dimming, and so it's either full bright, or dim. Feb 29 '20 at 4:13

simulate this circuit – Schematic created using CircuitLab

Figure 1. A possible solution.

How it works:

• The circuit is based around a comparator, CMP1. When IN+ < IN- the output switches low. (A transistor connects the output to ground.) When IN+ > IN- the output switches high. (The output transistor turns off.) For this reason RLY1 is connected between +12 V and CMP1's output.
• R1 and LED D1 provide a simple, reasonably stable voltage reference of about 2 V for a coloured LED.
• R2 is a potentiometer and the wiper can be adjusted to give a reference voltage of between 0 V and 2 V into IN- of CMP1.
• When DIMMER < R2 setpoint CMP1's output will pull low and turn on RLY1, connecting the headlights to the radio dimmer input. The dimmer will work as normal with the headlights.
• When DIMMER > R2 setpoint the relay will be off and headlight selection will not affect the radio.

There is one thing to watch. The RADIO may be using the bulbs as a ground path and with the relay energised there will be no ground path and it may think the lights are on. If so you might need to add some resistance between RADIO and GND. You should be able to figure out a way of testing this before you start the project.

• Wow! Thanks for the schematic! I am a novice and am not too familiar with electrical schematics, so I appreciate the description. So to be clear, I would cut the orange wire that feeds between the headlight and the radio and insert the relay. The ignition wire, I could also pull from the radio wiring harness as the power when the car is on, and no power when the car is off, correct? Also think you could provide a suggestion on what comparator and relay I should be looking at getting for this circuit? Feb 28 '20 at 21:23
• Your understanding of the wiring is correct. For a comparator try the LM211. .Any 12 V relay that draws < 50 mA (coil resistance > 240 ohms) should be fine. Let us know how you get on. Feb 28 '20 at 21:48
• Awesome! Thank you for your help. I am currently purchasing the parts. In my cart I have a LM311N, a trimpot 10k Ohm with knob, and am looking at a G5LE for the relay, would that relay work for my application? If I am reading the documentation correctly it has a coil resistance of 360 ohms and a current draw of 33.3 mA at 12V DC, correct? I already have resistors and a colored LED. Mar 2 '20 at 19:18
• Have fun. I realise that I've omitted the snubber diode on the relay coil. I'll update later. Mar 2 '20 at 21:33
• What size snubber diode I need? Mar 4 '20 at 14:48

There are two types of dimmers used in cars. Traditionally, it would be a wirewound rheostat, that would drop some of the voltage supplied by the battery to the panel lights. This is pretty inefficient, and newer cars tend to have PWM dimmers, that work by switching the power to the lights on and off by a variable ratio, but at a high enough frequency that you do't see it. Blot of these methods are used with traditional incandescent panel lighting, and LED lighting that's common now, though using a rheostat with either is also not ideal, since the response of the lighting is non-linear, so the rheostat has to be calibrated to a given load.

The dimmer can be on the high or low side of the lighting circuit, yours is obviously on the low side. Which type it is isn't yet clear, since a meter is likely to show the average value if it's a PWM type. If you set the meter to AC, you might see a voltage at the grey wire that isn't zero if it's a PWM type. Where the dimmer is on the head unit is unknowable, but irrelevant, since you just want to trigger it using the headlight input.

If you're just aiming to trigger the dimmer at position 1, this is challenging, since you don't even have enough voltage present to operate transistor to switch a relay. you can acheive this with an comparator that has a reference voltage probably midway between the 0V for position 0 and the 0.5V for position 1 to drive a relay (or a transistor, since the current needed for this is likely low). Som filtering on the input would be needed to stabilize the signal if it's a PWM type.

• I typically at night have my cars dimmer switch set to position 1, and then during the day set to position 0, but there are occasional times where I set my dimmer switch to position 2-7, and so any solution I go with, I don't want to create an issue since the voltage increases from position 0 to position 7. How would I be able to tell the difference between a wirewound rheostat or a PWM dimmer? I can tell you that there are a total of 6 wires that run into the vehicle dimmer switch. Most of the wires are at 0V and don't change when the dimmer position is changed. Feb 28 '20 at 20:52
• A photo might be beneficial to determine what it is. Feb 28 '20 at 20:59
• I don't have a picture with me, but I can get one tonight. This is picture of the switch I found online. I can get a picture of the switch/ wiring if that would be more helpful tonight... ebay.com/i/… Feb 28 '20 at 21:03
• That's interesting. Are some of those used for illuminating the control itself? Feb 28 '20 at 21:07
• I would suspect yes, since the switch does have illumination. I suppose I tested the voltages against a ground and I believe the headlights were off. I can go back through and measure the voltages when the headlights are on if that would help. I sort of stopped looking when I found the wire that changed values with the dimmer switch. Feb 28 '20 at 21:16

I think all this flows from a measuring error.

I think OP measured voltage across the dimmer, i.e. between "hot" and "dimmed light hot"... instead of measuring the voltage (compared to common/chassis) of the "dimmed light hot". As a result, all numbers are upside down.

This has led to an assumption that lights are wired between hot and the dimmer, i.e. That the dimmer is dimming the negative/common. I don't think so. I think the actual dimmed-hot wire voltage to common/chassis is:

• 13.8 volts with lights on, dimming 0 (100% bright)
• 13.3 volts with lights on, dimming 1
• ...
• 9.2 volts with lights on, dimming 7
• 0 volts with lights off.

Lamps are simply connected between this wire and common/chassis.

And I think the head unit's orange wire is actually designed to be connected to the instrument lighting wire that works as described above. And it is supposed to dim just like the above. However, The head unit needs high intensity backlight to be visible in broad daylight, so when instrument panel lighting is off altogeher, it illuminates at 200%. This is normal for radios with certain types of displays. I mean both the 200% brightness and also the use of the orange wire as a dimmer input.

There would be no reqson for a car to be wired in this backwards way that would require a function inverter like this.

So I would simply connect the orange to the instrument panel lights (post dimmer) circuit and see if that does the trick. I bet it does.

• All measurements taken were with the red connection of a multi meter to the wire and the black connection connected to directly to chassis ground. The vehicle is a 2016 Subaru WRX. The dimmer switch connector has 2 plastic covers on each side, that once removed exposes the terminals. I removed the 2 plastic pieces, plugged the connector into the dimmer switch, turned the ignition on, and then connected the black wire of the multi meter to chassis ground, and the red wire to each of the 6 wires, recorded the volts, then changed the position of the dimmer switch, and repeated. Feb 29 '20 at 3:55
• I then disconnected the power cable from the back of the radio, again turned the car to the on position, and measured the orange wire (which according to the radio's pinout is the illumination wire) with the red connector of the multi meter and the black wire connected to chassis ground. I turned the ignition on with the headlights off, and observed 0 volts, turned the headlights on and 13.4-13.8 volts was measured. To clarify the dimmer switch has a 12 volt supply (12.3 ish when the car is on) that comes in on a different wire than measured, and did not change with dimmer position. Feb 29 '20 at 3:58
• As for the radio, I measured the illumination wire that tells the radio that the headlights are on, (and verified with measurement with respect to headlight on or off position) the radio then switches to dim. I am not sure how the radio determines what to sent the screen brightness to, all I know is that my options are dim, or full brightness. I am trying to manage the input to the radio, not manage anything internal to the radio. The radio is a Pioneer DMH-C2550NEX for reference. If I performed any of my measurements incorrectly, please let me know as I know a little, but still learning. Feb 29 '20 at 4:04