# How to use 350mA constant current driver with a 180mA LED

As part of a domestic down-lighting system I've got a 350mA Constant Current driver, but i'm advised the LEDs I want to use should be driven with a 180mA CC driver. Changing the driver or LED's is going to be tricky.

1. Can anyone see any problem with putting a resistor in parallel with the LED to drop the current from 350mA down to the recommended 180mA?

2. How would I go about calculating the appropriate resistance?

The forward bias of the LEDs is around 0.6V.

• Besides .6Vf probably being a very deep infrared LED this doesn't sound like a good idea. Unless you are colinfurze you also don't put a V8 into a lawnmower. When its not designed to do the job, don't force it to do the job. Jun 23 '17 at 8:38
• If you have more of the LEDs and I mean Identical ones and you operate them under similar conditions, meaning at the same temperature. Then you can put several in parallel. Theoretically 2 in parallel would be enough but I'd incorporate some margin and use 3 in parallel. Using a resistor in parallel is tricky and wasteful as that generates heat and no light. The best solution is still using the right LED driver though. No way around that. Jun 23 '17 at 8:48
• I do have more LED's, from the same supplier and batch - I was put off putting them in parallel because of warnings I read about variable brightness. The resistor is obviously wasteful. If the current was split exactly, I would end up with 175ma through each LED - which is less than 180ma. So why would you put 3 in parallel? Jun 23 '17 at 9:03
• really, @iclegg, don't. Simply get the correct constant current source. This is all a really bad solution to your problem. Jun 23 '17 at 9:03
• I have not been able to find a 180ma constant current DALI DT8 CCT driver :-s Jun 23 '17 at 9:05

Can anyone see any problem with putting a resistor in parallel with the LED to drop the current from 350mA down to the recommended 180mA?

yes. That won't work out, or at least, it's hazardous.

Diodes, unlike resistors, reduce their effective resistance when heating up. Thus, after initial "correct" setting, the current through the diode will increase, due to the temperature increase and the resistance decrease.

That, in turn, means the diode gets hotter. And then, its resistance reduces, further increasing the current through it…

A vicious circle, often called thermal runaway. It usually ends with the diode failing by burning out.

One thing you could do is have another constant current source feeding the diode, and a "waste power" resistor in parallel to that. But it's an ugly solution that's not going to be easier than ripping out the original CC source.

If you cannot find a "constant-current" 180mA dimmable LED driver with a DALI interface which presents itself as a Device Type 8 (IEC62386 part 209 Colour Control) you could use a combination of the iLumTech TW Module and a DALI dimmable LED driver which can be set with a resistor for maximum 180mA such as LCA 10W 150–400mA one4all C PRE

There aren't many Device Type 8 drivers on the market so you are going to have to choose your LEDs to match the available driver rather than the other way round.

As Marcus pointed out in his answer, paralleling LEDs risks thermal runaway in whichever one gets warmer quicker.
However, commercial LED "light engines" routinely parallel LEDs or strings of LEDs without trouble.
There are a couple of things which can make this work:

1. Use LEDs from the same batch - they will have very similar V/I curves and will 'share' better
2. Mount them together on a substrate which spreads heat well - like an aluminum core PCB (best) or thin FR4 with lots of copper firmly bonded to a heatsink - you want them to be at the same temperature
3. Under-drive them. In your case paralleling 2 180mA on a 350mA is still a little tight. As bimpelrekkie suggested, rather use 3.

If you're not able to follow #2 very well, then a small series resistor for each LED can help to balance the current between them (the resistors' PTC compensates for the LEDs' NTC) when their temperature differ.
Probably something in the 1-3ohm range.