The diode test function is probably putting out a voltage of 1 to 1.5 V or so. This is below the forward voltage of the LEDs so they will barely light, if at all.
You can check the diode test voltage as follows:
- Switch the meter to diode test.
- Switch your second meter to DC volts and measure the voltage across the probes of the first meter. That is the voltage being applied to the diode under test.
- The '1' is an 'over-range' indication telling you that the voltage on the probes is the test voltage and that not enough current is being drawn. Effectively the probes are open-circuit.
- Now switch the second meter to mA and repeat the measurement. You'll probably read a few mA. This is the maximum current your meter supplies on diode test into a short-circuit.
You can swap meters and repeat all the tests to see if they have different test voltages.
3.3V "popped" the first one I tried: I'm guessing that range is for across their range of LED colours.
No. LEDs need to be driven by controlled current and not connected directly to a voltage supply.
Figure 1. LED current versus voltage for various colours. Source: LED I-V curves.
From the graph we can see that connecting a red LED to a 3.3 V supply would cause the current to go off the chart. A blue LED might survive the experience due to its higher forward voltage, Vf. There's a video I made on the linked page which demonstrates how to draw the graph and you may find it useful.
You might also find Ohm's Law - resistor calculation useful too.
From the comments:
I understand, but the LEDs said their Vf was "1.8-3.4V", so I put that to the test - if they did work @3.3V, I wouldn't need a resistor, is that right?
You've already found that you've blown some due to overvoltage so, no, it's not a good idea.
So my guess was that some colours in their range had a Vf of 3.4 V.
Yes, but at what current and what is the variation? See Variations in Vf and binning for more on this.
Is that graph true for all types of LEDs - all red LEDs are the same?
No. I generated it from datasheets for specific LEDs. We recommend, "No datasheet? No sale!" for reasons that you are learning.
Wouldn't I need to know the max forward current here? (I don't have Vf or Ifmax.)
Yes, but 20 mA is safe for most 3 mm or 5 mm LEDs.
To measure the Vf you can do a simple test.
simulate this circuit – Schematic created using CircuitLab
Figure 2. A simple Vf measurement circuit. DUT is the "diode under test". VM1 can be your multimeter on V DC range.
The 560 Ω resistor will limit the current through the LED to about 10 mA. (Actually, for a 2 V LED the current will be about 12.5 mA and for a 3.5 V LED the current will be about 9.8 mA. This will be good enough for most purposes.)