ULN2803 relay and led together

Question

Using diferent voltages for led and relay ULN2803

Hello guys, is it possible to use the wiring as shown in the picture? I have a ULN2803 and I would like the output to control the relay at 24V and the led at 5V. I have a little concern about the function of common-cathode clamp diodes.

Answer 1

The relay will hold the output of the ULN2803 at +24V while off, which will reverse bias the LED by 19V, potentially destroying it. If you insist on using separate supplies for the LED and relay, then you must protect the LED:

^{simulate this circuit – Schematic created using CircuitLab}

Diode D1 has a reverse breakdown voltage well above 24V, and will keep reverse LED current in the picoamps. R1 can be omitted, but it ensures that the voltage across the LED stays zero, regardless of which diode leaks the most.

Unless you have a specific reason not to, use +24V to power both relay and LED:

^{simulate this circuit}

Answer 2

The only problem I can see in your configuration is the reverse voltage across the LED.

ULN's output are open collectors so when the output O1 is not driven it'll be floating. Since there's going to be only a little leakage current, the voltage drop across the LED's cathode to anode will be approximately the difference of the supply voltages i.e. 19V.

No details about the LED are given but I'm assuming it's a single-chip, single-colour LED. Most of them today have a reverse breakdown voltage rating of 5V. If that's the case for your LED then it could be something to consider ⁽¹⁾.

There are few things you can try:

• If there are some unused outputs of the ULN then you can use one of them for the LED. Just short the respective inputs so you can use them as a single unput, and leave COM unconnected.
• You can use the same output for both the relay and the LED but with an extra Zener dropper if you worry about the LED resistor dissipation:

^{simulate this circuit – Schematic created using CircuitLab}

• The coil's high DC resistance makes me think about the possibility of using the LED and the relay coil in series. With a voltage drop of 3V (presumably) across the LED, the relay should still be energised (pickup voltage is 75% in the datasheet), and the current will be about 7 mA. If this is enough for the LED to give good illumination then why not consider?

---

⁽¹⁾ Years ago I did some experiments on Zener diodes and LEDs. In my experiments, the 3528-case red LEDs from different brands generally broke down at >20V. The white and blue (they are the same thing) ones generally survived 30V. But this shouldn't lead a conclusion like "Oh, 19V should be fine, then.". I'd never ever consider keeping the operational conditions close to the abs max ratings.

Answer 3

Table 1. Extract from datasheet.

The 8.3 mA coil current is a good match for the LED which gives the possibility of wiring both in series. You'd drop a couple of volts but the relay is guaranteed to operate at 24 × 75% = 18 V.

^{simulate this circuit – Schematic created using CircuitLab}

Figure 1. The alternate circuit. D1, built into the driver, will protect against inductive voltage spikes.

Answer 4

You could add a single BAT54C and get 'belt and suspenders' protection of the LED.

^{simulate this circuit – Schematic created using CircuitLab}

You probably don't need the 1N4148 in your schematic since the ULN has built-in catch diodes, so the parts count could remain the same.