So what is going on with the Nichia219 on copper?

Yes, the lumens peak around 3A (the lumens/watt probably peak around 100mA ) In the table that is also the case, but you may have read the voltage colomn as current?, the current is the first colomn. And I am only glad when people learn something from what i write down , just a few months ago a lot of the number stuff was new to me, I was glad to learn from others as well

ha! You’re right. I was reading volts as amps. Thank you. Makes sense.

I don’t know why I didn’t realize that. The testing done at CPF showed the 219’s efficiency drops pretty fast as current rises. They only tested up to 2A if I recall.

I’m excited to drive one at 3A. I’ll report back on how it goes. Thank you for work you did here again. :slight_smile:

I managed to get my DMM working again. So I did a tailcap test on my Nichia219 Z8.

I’m a little confused by the results. Hope someone can explain it to me

Stock Trusfire Z8 with the XM-L = 1.82A at the tailcap

With the nichia219 = 0.94A at the tailcap. But with the exact same driver and batteries. The only change was the LED & star.

Does this sound right?

Unfortunately this is exactly what is the pain with the high Vf of the 219. Lineair drivers like the 7135 regulator Nanjg driver adapt the output voltage to the Vf of the led so that they provide constant current, also with the Nichia 219 (as long as the battery can provide the voltage needed -another problem with the high Vf-). But drivers like the Z8 driver, and also cheap boost drivers as found in AA-lights or cr123 lights put out a current that is (also) dependant on the Vf of the led. So the 219 with its high Vf (that is: Vf is high at the higher currents) will always disappoint. I used a BLF-mini driver in a 219 build, in the Trustfire Mini01 it puts out a satisfying 2+ amps to the XM-L, with the 219 I only get 1.3A (which is actually a nice current for the 219 on a alu board).

The Vf of the 219 is high even under 1A, for example 3.2V for for 270mA.

With the 219, if you want more light, add more 219’s. Adding current makes more heat than light. 1-1.5Amp is a sweet spot. Lumen/watt really drops fast with these.

The B10 219 Illumination Supply has are the ‘low Vf’ lot. 1Amp drive is about 3.5 volts. (edit; mine are closer to 3.3V at 1Amp)

True, but djozz’s setup on copper appears to have extended the efficiency out. The angle on the chart is still pretty steep going up to 2.5+ amps.

Yeah, take my comments with a grain; If you are after absolute max regardless of current/heat/efficiency, there is still room to push, (maybe 2-3 amp?) But the Hi CRI 219 benefit vs. XP-G2 or XM-L de-domed; these are much more suitable for pushing hard. At least you are rewarded with lots of Lumen for your efforts…

True, but it’s still interesting to explore and break into new ground. Sometimes neat little goodies are found.

3.025V at 270 mA. It is 3.2V at 500 mA.

Personal measurement? Because you can see I took the data from the OP.

Yes… also confirmed by:

In the OP it is also said that my data are not quite calibrated and that I suspect that the actual Vf is about 0.15V lower than what the power supply measures, so at 270mA that would be 3.05V (and that comes close enough for me to the 3.025V that is undoubtedly more accurately measured than I did )

Should have ignored it ...

ok then, so using all of the info in this thread. Can anyone recommend a good driver for getting some good lumens out of a Nichia219?

I’m thinking in a host running 2x18350 IMR’s and suitable heatsinking.

perhaps the ld29 driver from intl-outdoor, or this one, also from them. I think the ld29 is nice, because 2.8A is just a bit more optimal than 3.0A. The driver is 20.3mm though, not suitable for every host.

I had a driver similar or same as the ld29 that I bought for an XM-L Maglite build, outputs a claimed 2.8A I’m running it with 2x26650’s and it seems to work fine with XM-L.

Tried it against my remaining Niachia I got from Nightcrawl, it’s on a tiny 10mm board, but to test it I attached it a big round bar of aluminum (1.5” plus by 4”).

Anyhow on High it goes blue. I won’t dare let it run for more than a fraction of a second. I assume the LED simply doesn’t like being driven at 2.8-3.0A???

EDIT: This is the driver, although the URL has changed and it now says 3.0A not 2.8A. But the top picture matches up which didn’t used to be present.

Hmm, was the led on a traditional aluminium board with dielectric layer between the heatsink pad and the core of the board? If so, it sure will not like the 2.8A. That current just works ok for sinkpad type boards.

Just the traditional type. I just wanted to see if it’d work and be worth putting into my Mag build instead of the XM-L. I think the answer is no.

I’ve just tried it wired up direct drive to a 14500 ICR and it seemed to work fine, so thankfully it doesn’t appear fried. I have a host coming in from DX which I think I’m going to put it and make it direct drive instead.

These numbers differ quite a bit from foxtrot824’s: LINK