SFT40 Gen 2?

Sadly, they are basically impossible to obtain, I tried to buy some samples from a LED supplier and failed.

AFAIK the D3AA driver could work at 6v without problems, so it would make more sense to use something like an XHP50.3HI or SFT70 instead, both being modern and readily available with standard footprints.

1 Thank

That makes sense.

Wouldn’t using a 6V LED reduce the driver’s power from ~18W to ~12W?

I don’t remember exactly, but it can probably be “fixed” by swapping the sense resistor, the same way that Hank changes the current on many of his drivers to accomodate the emitter used.

1 Thank

I tested an SFT40 gen2 6500k in a C8+ host@8A, yielded 2,300 lumens\177,000cd@30s. This is an improvement over the other SFT40s I’ve tested in C8 and M1 hosts. The only light I have that beats it in cd is a C8 SFT-25R@8A that puts out 1,713 lumens\190,600cd@30s.

I should test a gen 1 SFT40 6500k C8 to be fair, but regardless, I am pleased with it’s performance.

9 Thanks

NIce!
How did you know is Gen2 LED?

Share some test data from Nightwatch.


3 Thanks

I wonder why these values are very low compared to my tests (for example: SFT-40 N4 6500 K @12 Amp = 2557 lm from Nightwatch vs. 3090 lm from my test). So keep this in mind, if comparing these.

1 Thank

It may be due to the difference in devices, and accuracy and calibration. As far as I know, Nightwatch has an integrating sphere, but I don’t know the size. Maybe the results will be different from lumen tubes and integrating spheres of different sizes.

And of course, no matter what, I think it is still meaningful to compare the data horizontally as long as the test is done under the same setup.

Since I have also a sphere I think this could be also the heatsink used. (Of course the accuracy is also a thing.)
Especially the different maximum current in the tests is noticeable in this regard.

3 Thanks

I am trusting KD, I bought several there, in 5000k, 5700k and 6500k. I don’t truly know if these are just slightly better bin LEDs or actual gen 2 emitters. There are design files on Luminus website referencing SFT-40 Gen 2.

I want to know what is the N5-235UHP? I want to get one to test, any info on what/where to get one?

1 Thank

N5-235 and other N-prefixed’s are Nightwatch’s custom emitters. N5-235 is basically 5050 version of Luminus SFT-25R. They are not currently available in the West, but may be available in the West someday along with Nightwatch flashlights.

3 Thanks

Thank you for sharing! Looks promising.

This is the emitter I have been waiting for!

What is odd about this test is the 6500K G1 measuring lower than the 5700K F9.

The 5700K F8 sample of SFT40G2 that I installed in an M21B with a Convoy 10A buck driver has great tint, but is not brighter than a Gen1.
I have a 5000K G1 on the way. I hope it is better.

1 Thank

This emitter you waiting for is already available - the “Yinding 5050” are the exact same thing. Even with same LED chip.

2 Thanks

Not really, “same chip” does not mean “same emitter”, or even really “same chip”.

Many Chinese round die emitters look completely exactly the same, but use different chip manufacturers (SanAn, LatticePower), different product generations, and different emitter manufacturers’ processes and phosphors.

This information is not transparent and cannot be known before real testing. In other words, I don’t think trying to find “commonalities” for these Chinese round die emitters is better than just using their trade name, because they may indeed be different.

A test by an enthusiast said, N5-235(3.9V@10A) has a lower Vf than Yinding L50(4.05V@8A), and may be a newer chip than L50, but I heard it is a little bit green, DUV +0.0044 at low mode, DUV -0.0056 at 40W but CCT 10000K.

Yes, it is indeed odd, it may be due to individual sample differences.

1 Thank

That’s the real problem with all these Chinese domestic market LEDs. There is simply nothing known. “Yinding” or ‘Yinding L50’ is also just a placeholder name that has no meaning whatsoever. At least two different LED chips are visible in my picture, but the manufacturers are still unknown. Is this information also unknown in China or is there simply hardly any talk about it?

Generally unknown, they are not discussed much, these “Yinding L50-like” emitters are mainly used in cheap zoomies, and there seems to be countless manufacturers producing these white silicone packaged “egg yolk” emitters.

How many of these manufacturers do you think there are in China? Is it as I suspect that basically anyone with some money, some employees and good business connections can buy such machines and start their own LED manufacturing? I can’t really assess what the mentality is like there. Or are there clear rules and guidelines for LED manufacturers or does it require a lot of initial capital?

I don’t know much about this, but there are probably a lot of LED manufacturers out there. If their products are for OEM use or are designed for the low-end market, they may not provide much data. Or because of consistency issues, the data is different from batch to batch, so the data doesn’t mean much. Olight and Nitecore may have found better manufacturers to solve the consistency problem.