Nichia 519A

[Pavlo]

@jon_slider

519a 4500k is twice as bright as 219b 3000k

for daytime use, the 519a 4500k has bumped my sw45k out of the workshop

Hard to guage from the photo, but would you say the 219b at 200 lens still offers better intensity over the 519a at 400 lumens in these two identical hosts?

[jon_slider]

@Pavlo

would you say the 219b at 200 lens still offers better intensity over the 519a at 400 lumens

good question, I did a test and the answer is no.

the hotspot on the 219b measured 1.6 lumens, and 1.9 lumens with the 519a.

the lights were 22” from the wall, the 219b set to 31 lumens, and the 519a set to 59 lumens (thats as close to double I could program).

.

bottom line is the LUX in center of hotspot, of the 519a is 19% higher than the 219b, when the 519a is set to 90% higher output.

otoh, at equal lumens, yes the LUX is higher on the smaller hotspot from the 219b

.

with the 219b, set to 39 lumens, the lumens in the hotspot on the wall reads 1.6

with the 519a set to 41 lumens, the lumens in the hotspot on the wall reads 1.2

so the 219b has 33% more LUX when both lights are at the same lumens.

the smaller hotspot of the 219b is more intense, but the larger hotspot of the 519a is more useful at indoor distances, because it illuminates a larger field.

[Pavlo]

Thanks for all the great info. I bet the dedomed 519a will compete with 219b in intensity but benefit from the higher brightness, all else equal.

[jon_slider]

@Pavlo

dedomed 519a will compete with 219b in intensity

thaks for prompting some interesting comparisons:

.

Though I like the warmer CCT (3400K), I do not like the small, irregular, hotspot of the dedomed 519a

I ordered some 3000k 519a to try next, so I can
Leave the Hat (Dome) On

Joe Cocker - You Can Leave Your Hat On (Official Video) HD

output comparison… 519a is 53% brighter than sw45k:
.

[cannga]

“Parcel dispatched” from Aliexpress - 519a and 219b both at 4500k. I hope to have fun with these bad boys.
Have same LED’s in Emisar D4V2 bodies with various Carlo optics (spot, clear, frosted) so the fun should easily double. Maybe even triple with the GT FC40.
30 bucks each for these beautiful things - I simply could not resist.
And thanks to this thread for the valuable info.

[jon_slider]

@cannga

“Parcel dispatched”

woohoo!
thanks for spending your money for my vicarious thrill

I look forward to your Day of Joy…

[stephenk]

@cannga

“Parcel dispatched” from Aliexpress - 519a and 219b both at 4500k. I hope to have fun with these bad boys.
Have same LED’s in Emisar D4V2 bodies with various Carlo optics (spot, clear, frosted) so the fun should easily double. Maybe even triple with the GT FC40. :innocent:
30 bucks each for these beautiful things - I simply could not resist.
And thanks to this thread for the valuable info.

I’ve also got a S21D 519A on order. Looking forward to it, may even rival the excellent Acebeam E70.

[swhs]

@djozz

After several weeks of absence I was in my 2m2 hobby universe once again to test a few emitters for rngwn, and while doing that I did an output test of the 519A from Simon too. So here, without proper test thread, are my results. Mind that djozz-lumens are a bit higher than maukka-lumens, on average around 7%.

Does that mean it is safe to use Simon’s copper MCPCB with the 519A in a SC31pro? I think tailcap current was measured somewhere at 6.7A (haven’t measured myself, and definitely didn’t take it apart to measure current to the LED but that is likely the same unless it’s a buck driver. Datasheet spec is 6A max on the SST40 so I suspect Sofirn won’t let it go much beyond that).

Secondly, generally the datasheet values look to be very conservative, also with the SST20, which is stated to be max 2A or 3A depending on datasheet. However, it is driven much higher in many tests. So could I take Simon’s copper MCPCB with SST20 4000K and use it in a Wurkkos FC11? (where it is driven at 3.7A as I read in one review) and will it last or will the LEDS not last long with these high currents?

I am curious about this as I don’t like the LH351D in the FC11, it is not bad, but something just feels off every time I use it. For general use I would even prefer the lower CRI XPG 5B1 and XML 4C I have in other torches. I have an IF25A with SST20 4000K and that is just perfect in light colour (for a work light), where nothing feels weird at any time (in any situation with different coloured objects), so I’d like to install that in the FC11 (along with a TIR to give a wider beam). I am curious to see the 519A in action too and compare it to the SST20.

[thefreeman]

Single 519A direct drive (SC31pro) will draw too much current, unless they use really bad springs and you use a high DCIR cell (which the case of sofirn cells actually ~60mΩ, unless they improved since last time I measured one ~2years ago)

[jon_slider]

@swhs

I have an IF25A with SST20 4000K and that is just perfect in light colour (for a work light), where nothing feels weird at any time (in any situation with different coloured objects), so I’d like to install that in the FC11 (along with a TIR to give a wider beam). I am curious to see the 519A in action too and compare it to the SST20.

good luck with your 519a testing… fwiw:

What did you mod today?

I tried to upgrade my Sofirn IF25A to use Nichia 519A’s. Too low forward voltage. I started with four of them and lost two almost instantly after turboing the light.

@Haukkeli

I tried to upgrade my Sofirn IF25A to use Nichia 519A’s. Too low forward voltage. I started with four of them and lost two almost instantly after turboing the light.

[story]

@swhs

@djozz

After several weeks of absence I was in my 2m2 hobby universe once again to test a few emitters for rngwn, and while doing that I did an output test of the 519A from Simon too. So here, without proper test thread, are my results. Mind that djozz-lumens are a bit higher than maukka-lumens, on average around 7%.

Does that mean it is safe to use Simon’s copper MCPCB with the 519A in a SC31pro? I think tailcap current was measured somewhere at 6.7A (haven’t measured myself, and definitely didn’t take it apart to measure current to the LED but that is likely the same unless it’s a buck driver. Datasheet spec is 6A max on the SST40 so I suspect Sofirn won’t let it go much beyond that).

Secondly, generally the datasheet values look to be very conservative, also with the SST20, which is stated to be max 2A or 3A depending on datasheet. However, it is driven much higher in many tests. So could I take Simon’s copper MCPCB with SST20 4000K and use it in a Wurkkos FC11? (where it is driven at 3.7A as I read in one review) and will it last or will the LEDS not last long with these high currents?

I am curious about this as I don’t like the LH351D in the FC11, it is not bad, but something just feels off every time I use it. For general use I would even prefer the lower CRI XPG 5B1 and XML 4C I have in other torches. I have an IF25A with SST20 4000K and that is just perfect in light colour (for a work light), where nothing feels weird at any time (in any situation with different coloured objects), so I’d like to install that in the FC11 (along with a TIR to give a wider beam). I am curious to see the 519A in action too and compare it to the SST20.

https://www.reddit.com/r/flashlight/comments/svw36r/wurkkos_fc11_sst20_4000k_vs_lh351d_4000k/?utm_source=share&utm_medium=ios_app&utm_name=iossmf

[Verodin]

@swhs

Does that mean it is safe to use Simon’s copper MCPCB with the 519A in a SC31pro? I think tailcap current was measured somewhere at 6.7A (haven’t measured myself, and definitely didn’t take it apart to measure current to the LED but that is likely the same unless it’s a buck driver. Datasheet spec is 6A max on the SST40 so I suspect Sofirn won’t let it go much beyond that).

Fwiw I measured the actual LED current and don’t get over 2,5A on turbo. Stock SST40. Tested with Wurkkos 3000mA and LG HE4 cells (both not fully charged though). Have another SC31pro that I modded with a LH351D and so far no issues.

see below

[thefreeman]

@Verodin

@swhs

Does that mean it is safe to use Simon’s copper MCPCB with the 519A in a SC31pro? I think tailcap current was measured somewhere at 6.7A (haven’t measured myself, and definitely didn’t take it apart to measure current to the LED but that is likely the same unless it’s a buck driver. Datasheet spec is 6A max on the SST40 so I suspect Sofirn won’t let it go much beyond that).

Fwiw I measured the actual LED current and don’t get over 2,5A on turbo. Stock SST40. Tested with Wurkkos 3000mA and LG HE4 cells. Have another SC31pro that I modded with a LH351D and so far no issues.

There is probably an error somewhere, it’s a direct drive light, the current should be much higher than 2.5A with an SST-40. Zeroair measured 6.8A.

[Verodin]

The cells were not fully charged (but 100% sure over 3,78V), maybe that’s the reason for such a low current. Got the LG charging now and I still have it lying on my workbench. Will report back once charging in done.

edit:
Redone measurement. Fully charged battery, settled for 15 mins at 4.17V. Measured 3.3A on turbo and quickly dropping. See my next post.

Started seeing smoke and found it was melting the plastic of the measuring clamp thingy… woops

[thefreeman]

DMMs cannot be used to measure flashlight current because of their high voltage burden (i.e. they have a high resistance), which will significantly reduce the current.
You need to use a current clamp, that way no resistance is added to the circuit.

[Verodin]

Oh wow, massive difference! 7,85A with cell 4V. On high 7,35A

Thank you for sharing your knowledge

Probably also time to reconsider the LH351D I have in the other sc31pro… most likely luck that I didn’t use a fully charged cell yet.

[thefreeman]

Yep, so the SC31 pro has a fairly low circuit resistance , using my DD calculator , around 20mΩ (7.7A with SST-40, 30mΩ for LG HE4, 4Vbatt).
So even with a 35E/GA (35~40mΩ) it could do 10A with a 519A and fully charged cell.

[Verodin]

Since the FC11 with LH351D was brought up above as well, I had a go at that one too. 2,8A on turbo with same cell at 4V.

Same method with the current clamp from the start

[contactcr]

@Verodin

Since the FC11 with LH351D was brought up above as well, I had a go at that one too. 2,8A on turbo with same cell at 4V.

Same method with the current clamp from the start

Unless the MCPCB is screwed down you shouldn’t go into turbo without a good heat path. In a lot of lights that means having the bezel/tir/reflector pressing down on the PCB. Most of us use a thick wire between battery negative and bare metal where the tail cap threads on.

[swhs]

@Verodin

Oh wow, massive difference! 7,85A with cell 4V. On high 7,35A

Probably also time to reconsider the LH351D I have in the other sc31pro… most likely luck that I didn’t use a fully charged cell yet.

So what ‘thefreeman’ wrote about turbo on the SC31pro being direct drive, would that mean that removing turbo (changing the firmware) would make it safe? I mean that anything you get to with ramping/stepping is current controlled, right?

I would be interested if you could measure what the highest current is that you can get to with the highest ceiling and then using just ramping to ramp to the ceiling.

If that is not too high a 519A could be installed, otherwise the ceiling or stepsize would need changing in the firmware too.