Use LED to trigger relay?

Thread Starter

Ballo

Joined Jul 28, 2019
14
Hi, new user here! Know and use ohms law but I'm more familiar with digital circuits than analog.

Anyway I have an automotive problem. The sort version is I have an LED which I'd like to close a switch to a large relay. I already have a switch to the relay on the dash near the LED in question. I was wondering if there was a low voltage relay I could run in series do accomplish this simply.

Background: I rewired an ambulance (camper van) to isolate the battery with a voltage-sensing relay (basically a 500 amp relay that sits between the deep cycle bank and the starter battery that engages if either battery is being charged so both banks get charged by the alternator or solar). Now, I have a 35 amp 12v air compressor which I only want to run when the batteries are being charged (so as not to drain batteries or run it in the evening). I don't want to go through the expense of using a second voltage-sensing relay. I already have a full duty relay (came with the ambulance and is wired to a switch on the dash) so if I could add a relay inline to that switch which is triggered by the LED I'd be close to a complete solution.

An alternative could be to add another voltage sensing relay, but that would be like $80 I'd rather not spend plus the effort in installing it plus I'd have to route another switch to the dash.

Any input would be appreciated :)
 

djsfantasi

Joined Apr 11, 2010
9,156
Forget about wiring it in series. How much voltage is supplied to the LED? This will help you select a relay. What you should do is figure out how to wire a relay in parallel with the LED, so when you switch power to the LED, the relay is actuated as well.

That’s what you want, correct?
 

Thread Starter

Ballo

Joined Jul 28, 2019
14
Unfortunately my DC clamp meter went missing (I may have left it on a picnic table in the middle of nowhere, Nevada) so I'm not sure yet. I have another non-clamp meter but all the probes were in the case with the clamp meter :rolleyes:

The reason I was considering series is I don't want to overly tax the circuit driving the LEDs, and I don't care if the LEDs are dimmer. It's just an indicator that the charge relay is engaged.
 

Thread Starter

Ballo

Joined Jul 28, 2019
14
I should add that it drives two bright red LEDs and I only need one, so I have some power to play around with. I'll see if I can find my DC clamp meter or borrow one
 

djsfantasi

Joined Apr 11, 2010
9,156
Unfortunately my DC clamp meter went missing (I may have left it on a picnic table in the middle of nowhere, Nevada) so I'm not sure yet. I have another non-clamp meter but all the probes were in the case with the clamp meter :rolleyes:

The reason I was considering series is I don't want to overly tax the circuit driving the LEDs, and I don't care if the LEDs are dimmer. It's just an indicator that the charge relay is engaged.
Putting a relay in parallel may actually lessen the impact on the LED. The relay current may be greater than the LED maximum current. Also, if you put in in series, you’ll have to recalculate the current limiting resistor for the LED and if it’s a 12V LED assembly, you may not be able to adjust the resistor.
 

Thread Starter

Ballo

Joined Jul 28, 2019
14
The clamp meter would tell me how many amps are going through the LEDs. Unfortunately, I think I lost it for good and am now researching a replacement.

I don't care about the LEDs, only the driver which is embedded in an expensive relay
 

djsfantasi

Joined Apr 11, 2010
9,156
The “driver” for what? The LED?

Is it a single LED and what color? A rough estimate would be 20mA for the LED. I can’t predict the current draw of a driver.

If you want to activate a relay with the same switch as the LED, you don’t need to know the current draw of the LED when the relay is in parallel.

If the power is from the automotive supply, you have enough for a relay. Don’t overthink the problem. Just connect a 12V relay in parallel with a sufficient contact rating.
 

Thread Starter

Ballo

Joined Jul 28, 2019
14
They're red.

The reason I want to measure current is as not to over tax the driver. I have no idea how the driver is designed and it's only designed to run the two LEDs. Putting anything in parallel will increase the amps and tax the driver.

The driver will likely be from some IC so please don't think it will be 12v based. I may have to use a 1.5V relay.
 

djsfantasi

Joined Apr 11, 2010
9,156
They're red.

The reason I want to measure current is as not to over tax the driver. I have no idea how the driver is designed and it's only designed to run the two LEDs. Putting anything in parallel will increase the amps and tax the driver.

The driver will likely be from some IC so please don't think it will be 12v based. I may have to use a 1.5V relay.
And there’s the problem! Putting it in parallel will not increase the amps through the driver (if you wire it correctly). Unless the driver is before the switch... Putting it in series may have a larger current draw than a parallel circuit. Depends on your wiring. Uh

It’s time to ask if you have a schematic of the existing wiring. Do you?
 

Thread Starter

Ballo

Joined Jul 28, 2019
14
So what you're saying is either wrong or counter-intuitive. If I place items in parallel the overall resistance on the circuit will lower and the amps will go up. Maybe you're envisioning some kind of complicated setup where each LED is individually driven which isn't the case. If I place anything in series that can only increase the resistance. Tell me where I'm wrong in my thinking.

I did the wiring myself and there isn't much to show. The LEDs are wired in parallel internally in the switch:
http://assets.bluesea.com/files/resources/instructions/990180180.pdf
(page 3 has the wiring)
 

djsfantasi

Joined Apr 11, 2010
9,156
Remember voltage drops/divides in series and stays constant in parallel. Current divides in parallel and remains constant in series.

Then, LEDs are constant current devices. They are not resistive in nature, so you can’t think of overall resistance lowering in a circuit with an LED.

If you place a relay in series with an LED, the relay current is likely to be too high and cause the LED to fail. Remember, current is constant in a series circuit.
 

Thread Starter

Ballo

Joined Jul 28, 2019
14
Yes, I'm well aware voltage drops in series. That's why I said dimmer LEDs would be acceptable. Current divides in parallel only if the circuit is current limited. I have no way of knowing if this is the case. If it isn't, the overall current will increase. This is also why I want to measure the current current (no pun intended) so as not to go over that.

Tell me where I'm wrong.

The only time I've heard of an LED burning out in parallel is because when one LED fails, the others are over-driven. This isn't due to a short, but the opposite. I have never heard of an LED burning out because more resistance is placed on the circuit in series, which could only limit amps.

The only difference between an LED and a resistor that I can see is that before it reaches the threshold to emit light, it's essentially closed, but after that point it's linear like a resistor:


The only downside to series that I can see is if that threshold can't be reached in which case the relay won't work at all.
 

Alec_t

Joined Sep 17, 2013
14,280
You could use the forward voltage across the LED (about 2V for a red LED), or the current through the LED (let's assume 15mA) to switch on a transistor which, in turn, switches on power to the relay. The type of transistor (npn, pnp, Darlington etc) would depend on how the LED is currently wired (a wiring diagram would be helpful) and the relay coil current needed.
 

Thread Starter

Ballo

Joined Jul 28, 2019
14
I posted the wiring diagram URL.
There appears to be a common positive with an LED ground. Both LEDs are wired in parallel in the switch. I could disconnect the LEDs for all I care; the switch I want to power with the relay has a light (12v incandescent likely).

You don't think a 1.5VDC relay would work?
 

Alec_t

Joined Sep 17, 2013
14,280
The relay in that post #18 link requires a coil current of >90mA, so would not work in series with the LED(s).
Latching relays generally require either dual-polarity pulses to set/reset them, or have separate set and reset coils. Neither type seems easy to interface to your circuit.
One complication with using the LED(s) to cause your relay to switch is the fact that they have multiple functions, as per this table from the linked datasheet:
LEDindications.PNG
You would need a circuit which could distinguish between LEDs on continuously and LEDs blinking. Doable, but not as simple as you'd probably like.
 
Top