12v LED dimmer 'relay'? (Add more LEDs to existing LED dimmer circuit)

Thread Starter

fognozzle

Joined May 24, 2021
3
Hi, newbie here, I'll try to be descriptive yet concise!

My car has LED dashboard instrument lights, dimmable by a push button that cycles through 6 brightness levels. It remembers separate day and night settings.

I want to add a number of switches and gauges to the car that all contain LED illumination intended for a 12v circuit. I'd like to get them to follow the instrument lights, rather than use a separate dimmer.

I don't know how to interface the two circuits, or what voltage the instrument lights are running at (I assume they are being PWM, but I've just about reached the limit of my electronics ability). The instrument lights have a ULN2003A chip, and SMD LEDs that each have a 1k resistor. I've attached flyleads in parallel to an LED+resistor and my basic autoranging DMM reads the following, if it helps:

1 (max): -3.6v
2: -2.7v
3: -1.7v
4: -0.7v
5: 0.0v
6 (min): +0.8v

Thanks in advance for any guidance!
 

Thread Starter

fognozzle

Joined May 24, 2021
3
Thanks ElectricSpidey, and apologies for taking so long to get back here! I wanted to check out a couple of things and draw up a diagram, and it took a bit longer than expected.

The existing lights appear to use a constant +12v 'rail' equal to the battery voltage, and are dimmed on the ground side. The switches are most easily used the same way (but could conceivably be wired differently).

The gauges are wired the other way, and have a dedicated +12v illumination input with a common ground internally. So it seems like my ideal solution will be something like this:

dashlights.gif

I have ordered a couple of TC4420 to play with. Am I right in thinking I can just wire the IC directly into my circuit, and the output will just 'pulse' at the same rate and duty cycle as the input signal?
 

ElectricSpidey

Joined Dec 2, 2017
2,758
Ok, I'm going to get back with an update.

Some questions in the meantime.

Did you confirm it's PWM?

Are the common grounds in the gauges shared by other electronics of just the LEDs?
 
Last edited:

ElectricSpidey

Joined Dec 2, 2017
2,758
Just a caution...this system will not work if the existing system is not PWM.

So based on your ideal solution.

The drivers are TC4420 for the low side switching. (switches)
And TC4429 for the high side switching. (gauges)

Your BOM is as follows not including the board, wires, terminals, enclosure, fuse etc.

1 4.7k 1/2 watt resistor (you can probably change this value, depending on the existing load on the ULN2003 1k would be more ideal)
The two drivers and sockets (sockets are optional but I recommend using them)
2 1uf ceramic capacitors (they go directly across the supply connections of the drivers, not shown on the diagram)

Be sure to read the application notes and the data sheet for the drivers, there is some important information such as how the dual VSS, VDD & Outputs should be connected together.

You may be able to find a driver with complementary outputs to replace using 2 drivers as shown, I just don't know any off hand, and don't have time to start looking for one.

(the resistors and LEDs are reversed on my diagram, but this does not matter)

Any questions be sure to ask.

AAC_Guages.jpg
 

Thread Starter

fognozzle

Joined May 24, 2021
3
Thanks again, I really appreciate your detailed and generous responses!

Did you confirm it's PWM?

Are the common grounds in the gauges shared by other electronics of just the LEDs?
I'm assuming it's PWM, based on the weird voltages* I'm reading across the existing LEDs, the inclusion of a ULN2003A in the circuit, and the thought that PWM is probably the sensible way to design a board like this (the illumination LEDs are totally integrated into the design of the instrument cluster board, rather than a discrete system). But I'm mostly working on hunches here.

The new gauges are electronic (and sealed) and use the same ground for illumination and for other circuitry.

I'd prefer to dim the ground side of the switches, as they contain both 'background' and 'indicator' LEDs, the latter sharing its supply with the load. It would be great to dim both, but not the end of the world if the indicator wasn't dimmed.

I assumed the TC4429, being the 'inverting' version of the TC4420, meant that the duty cycle was inverted (and therefore of no interest to me). But you are saying the modulated output is (positive) on one and (ground) on the other - so I actually need one of each chip to achieve the desired result?

I will order TC4429, and digest the diagrams and info you have posted. Again it's very kind of you to include details of the additional components I should add - I have a working electrical knowledge, but my electronic ability is currently far too basic to figure out the finer details!

(*the voltages in my first post should all read mV, not V).
 
Top