# Derive voltage of LED array?

I'm trying to diagnose a faulty builder's worklight.

It has a single board, a 8.4V battery, an external 8.4V DC input, and a LED Panel.

What voltage would you expect at the LED panel?

I'm measuring 8.2V supplied to the LED panel, when powered by battery to 8.4V when the DC-mains lead is connected. An 8.4V test LED borrowed from a bike light works fine, but the LED panel pictured does not illuminate.

The LED panel physically has 6 rows of 6 LEDs, but trying to look at the traces suggests its 4 rows of 9 LEDs but "folded up"

I don't have any test PSU capable of doing more than about 19V, and the LED panel does not illuminate at 8.4V.

Which is likely faulty - the LED panel or the driver board?

• You need to figure out how the LEDs are connected together to work out how much voltage the panel needs. Each LED typically needs somewhere between 3V and 3.6V to operate. Jul 21, 2022 at 11:30
• If that large cap near the output is in parallel with the output, it's voltage rating will give a clue. Jul 21, 2022 at 12:22
• 9 white LEDs in series would require about 27V (3V each) to get anything more than a dim glow. Jul 21, 2022 at 12:45
• One broken LED may cause faulty condition to one row, not all LEDs. My guess the PS is broken. Jul 21, 2022 at 13:17
• Wonderful - thank you all, I'll progress this and update. Jul 21, 2022 at 22:33

Based on the cap voltage and your belief that there are 4 rows, there are probably 9 LEDs in series. Then the 4 rows are in parallel. LEDs don't share current well normally because of slight differences in voltage drops, but if they are matched, then this is acceptable.

I agree with Simon B, about 3V each, maybe a few tenths more for high brighness.

It seems unlikely that 4 LEDs have failed. So I would suspect the power converter.

Further troubleshooting. I would try to light each LED individually with a 5V power supply and a 100 ohm resistor. With needle probes you should be able to make contact with the LEDs. Be sure to get the polarity correct.

The LED array is definitely 9 LEDs in parallel, with 4 banks in series.

It lit successfully with a 19.5V laptop PSU as a source, but would not light at all with the 8.4V from the battery nor with 12V from another PSU.

I would have expected it to light on 12V, based on 3V per LED and 4 in series.

(yes I need to buy a bench power supply, its on the list. At the moment I have an old ATX PSU connected to banana sockets, with meters on the 3.3, 5 and 12V lines.)

I'm going to blame the PSU/driver board as "gone dead" and possibly the tactile membrane switch pane on the rear as well.

Repair plan is to fit a 2s Lithium battery charger board and a separate step-up converter board from 8.4V to around 19V. And a physical switch.

Turns out unit is 5 years old, and sat around unused all that time. It was not charged before use either. No idea if this contributed to its failure after running once for about 4 hours successfully.

• It is probably a boost converter in which the switching element has failed, so that the raw 8.4 VDC from the battery is fed to the output through a series diode. Some LEDs will just barely light at 3 volts, and the output voltage of the switcher will be one diode drop less than the 12V applied (which may also be a bit less). There seems to be a MOSFET near the output, that has either failed open, or the drive PWM has failed. Those possibilities should be easy to analyze, and possibly fix. circuitdigest.com/tutorial/… Jul 24, 2022 at 23:47
• You really need a current regulated boost converter. A fixed voltage may drive the LEDs too hard, especially when they heat up and forward voltage drops. Jul 24, 2022 at 23:52