519A vs B35AM in S21A

I have been carrying a Convoy S21A with a Nichia B35AM emitter for a few weeks. It’s a really nice light. A little bit of throw, but not a pencil beam so it’s useful up close, almost no tint shift, perfect color and tint (4500K).

My only 519A light ATM is a T2. So, it’s not as bright as the S21A, obviously, but I’m kinda curious what all the 519A fuss is about. I’ll probably end up getting an S21A with a 519A just to compare, but, according to the specs and others’ experience, what should I expect? I suspect the 519A will be less throwy than the B35AM since the 519A has a dome and the B35AM doesn’t. From the limited real-world specs I’ve seen, the the B35AM is probably more lumens, but not enough to make a big difference. Does that sound right?

Is one likely to run cooler, more efficient, or longer than the other, all other things being equal? Any other already-known differences that are significant enough to notice?

Even if I end up deciding the B35AM is my current Most Favored Emitter, I’m sure I’ll find a use for a 519A torch, and won’t regret buying it. Just trying to do a little homework first.


Throw: domed 519A compares to B35AM. Although B35AM is domeless, it has a much larger emitting surface.

Output: B35AM is a bit higher (more efficient), but the difference is not super noticeable.

Tint: 519A will have more tint shift at 4500K. The tint shift is not as bad at 5000K and 5700K, but won't be as good as with B35AM. The hotspot will be a bit more yellow compared to B35AM. If you end up disliking the tint shift, dedoming works wonders for 519A.

Overall runtime and efficiency: I really don't know, as this depends on both emitter and driver, which are different across these two lights.

519a is super-easy to Dodome. Dedoming makes the color temp approximately 1000K warmer, reduces DUV, reduces lumens and greatly increases the throw.

519a is my current favorite emitter. It’s outstanding!

Yeah, IIRC, 519A is 3V, and B35AM is 6V, so the B35AM would require a boost driver. Right? Dunno exactly what’s in the S21A driving a 519A, or what difference that would make.

Dedoming is probably not a good task for someone with permanently numb fingertips. It’s too bad Simon doesn’t dedome on demand.

But, since you like the 519A, what do you like better about it than the B35AM?

Firelight2 may or may not agree with me, but here are a few things I love about the 519A, which IMO is the undisputed best high CRI emitter available right now.

  • Standard 3535 footprint and 3V layout. This makes it a direct replacement for arguably the largest class of flashlight-compatible emitters. Other LEDs with comparable levels of CRI and tint, such as E21A and B35AM, tend to have strange footprints that make them difficult to use.
  • Underrated CRI. Although rated 90 CRI and 80 R9, the CRI almost never drops below 95, and some CCTs (such as 5700K) consistently score above 97. 90CRI emitters from other manufacturers do not have this level of performance.
  • Versatility: while the stock emitter already has great tint and wide selection of CCTs, dedoming them creates a different beast with the best angular tint uniformity and nicest tint I've ever seen, while still maintaining high output unlike 219B. This is also THE easiest emitter to dedome due to its unique glass layer. Dedoming 2700K results in a rosy, high-output 2200K emitter that couldn't really be obtained any other way.
  • High output while maintaining extremely good color rendition and tint. Some high CRI emitters like LH351D have more output on paper, but at the cost of significantly worse color rendering (especially reds and browns) and much greener tint.

The B35AM is more efficient, is a great all around emitter, requires a boost/buck driver(which is great for regulation), etc.

Being domeless stock also means more consistent output and tint.

I completely agree.

Previously my favorite high-CRI emitter was 4000K SST-20, but only from good bins. Bad bins were extremely green and looked pretty bad.

However, having installed 519a both domed and dedomed in numerous lights I’ve concluded that it is superior to SST-20. Especially when dedomed the beam is fairly rosy with low DUV. It doesn’t look greenish even when run at low power. It makes even the best tint SST-20s look bad.

Unfortunately, I cannot comment on how 519a compares to B35AM as I do not have any lights with that emitter.

I like that! I ’m waiting my S2+ 519a with clip AND diffuser!

L.R.O. has reviewed the S21B 519A and the S21A B35AM. Comparing stills from a video is hard enough, but since ceilingbounce only shows un-calibrated, un-related percentages, it’s hard to tell if the B35AM is really more efficient, or if the driver just has different step-down levels. It does appear to run longer, though.

Oh yeah, the B35AM in the S21A is far more efficient as a whole setup.

While the B35AM S21A might me better on paper, it had an important problem: some of them were smoking when pushed at 100%.

This was addressed by changing the maximum current from 3A to 2.2A in the drivers which are being shipped now, so it is important to have this in mind.

The B35AM may seem like a slightly better choice, but the output is now lower than it was in the initial design and reviews, and has a thermal design problem which was addressed in the only practical way possible, using a workaround and not addressing the root cause.

This is why, for my S21A order, I decided to purchase the 519A option even if I would have preferred to try the B35AM.

Nichia specifies 1.4A as the max current, so I’m not terribly concerned that Simon overdid it in his hot-rodding by 700 ma initially (Simon’s message says he dropped it to 2.3A). The 3V 519A tops out at 5A, so that’s really a tiny difference. I’m more interested in efficiency than starting fires, anyway, and my post-April S21A is very bright enough.

I do envy the efficiency, but the tint shift with the 3A driver was pretty concerning, the whole situation did not look healthy for the LED:

I think your driver should be running at 2.2A, because that is what is advertised in the AliExpress listing.

Out of curiosity, is a tint shift unnoticeable in your post-April model? My 519A has a beautiful and consistent tint from 1% to 100%.

If you consider the base efficiency difference + the lower current, you are likely getting 50% more runtime than me, which is very neat.

You’re probably right. Even then, that’s only 12.8% less power than the 519A driver. At 3A, it was 18% more than the 519A. I suspect we’re talking differences that are completely invisible to the naked eye (except for the tint shift). And I don’t have the gear to measure the difference in light output from a 12% change in power.

No, there is no discernible tint shift at all in the beam profile, or at different levels. That’s one thing I like a lot better about the B35AM I have (in a S21A) than the 519A I have (in a T2). The beam profile of the T2 isn’t nearly as bad as some Cree fried eggs, but there is a a slightly yellow halo around the hot spot. It’s probably only visible on white walls, but I notice it. Maybe that’s just a T2 thing, I don’t know, but I think it’s more likely a dome thing, because the least shifty single-emitter lights I have are B35AM and FC40.

Now I’m thinking I need to order a 519A S21A, but throw in a M21B with a B35AM to keep it company.

Seems like the conclusion arrived at in general is that B35AM shouldn’t be driven up to 3A, or the LED will start to degrade and eventually blow. So Simon and Hank toned down the drivers to put out around 2.2A maximum, to protect the B35AM.

I’ve really enjoyed reading this thread. Thank you all for sharing your knowledge.

I have an XHP 70.3 at 5kK, in my new M3-c, as it is quite powerful and a good combo throw/flood. Based on videos showing the 519a’s color at 4500K, I didn’t pick up any “magic hour” rosiness from any of the 519a 4500k beam shots, and that seems to be the consensus opinion here (lack of rosiness). I am not the right person to attempt de-doming either. The B35AM does seem a very attractive tint.

I think I will order the S21D in 219b mostly for walking the dog at night, and an S11, for the car. I am kind of stuck between GT FC40 and B35AM for the S11. I do not need throw at all, the floodier the better.

gt fc40 will be much more floodier and poweful than very weak b35am. B35am gives only 1000 real lumen at almost 18w and it literraly burns at this power.

I really don’t understand the whole watts, concept. I am more concerned with how long can I shine at a desirable sustained lumen output (let’s say 500lm) until the battery makes the light turn off. According to a couple of reviews I read, in the Convoy S11,

B35AM @ 4500K at 35% is at 542 lumens, and runs for 174 minutes

GT FC-40 @4500K at 35% is 710 lumens and runs for 110 minutes.

both using LittoKala 5000 mAh bundled battery.

That kind of looks equally strong to me — one about 25% fewer lumens, but lasts about 50% longer. If anything seems like the B35AM is more “efficient”.

However, if as you say the GT FC-40 is floodier (which it seems like from review beamshots also), that makes my decision simple. So I will have two floody lights… S21D in 219b, and S11 in GT FC-40.

I guess the only decision left for me is to decide between 4500K and 4000K for the 219b.