XHP50.3 HI - 8,41 mm²
LHP531 - 13,84 mm²
XHP70.3 HI - 14,6 mm²
SFT70 is about 6,5 mm²
3 Thanks
That might be a sensible choice, as the 5A driver has better mode spacing. But wait for others to chime in and see what they think. It’s possible that a high-drain 21700 might get a bit extra output out of a 8A driver.
1 Thank
But it also depends on the specific design of the driver, such as the max duty cycle, and the overall voltage drop of the circuit. The construction of my Fenix flashlight is actually a bit atypical, it’s an electronic tail switch light with single battery tube with normal battery, probably not a typical example.
It’s true that SFT25R can’t stay regulated at 8A for very long, ideally, I think for an 8A(+) SFT25, should use BUCK+turbo FET design.
2 Thanks
Just checked the T6 at 50% (2.5A) level and measure 889 lumens. The 8A driver at 35% (2.8A) measures 850 and at 50% (4A) it’s only 1100 when the T6’s 5A is hitting almost 1500.
I wonder why this 8A driver isn’t doing well paired with the SFT25R on all levels
Tried with the 5A Buck driver that had the early stepdown issue, and get a max of 1300 lumens, so I’m just going to leave it with the 8A and forget about it. Very strange though.
4 Thanks
Hi Simon, Hopthink has some interesting new emitters, any thoughts?
2 Thanks
@Simon_Mao, would LMP be interested in making LHP531 in high CRI version, or do they make other high CRI LEDs?
Also, how is the performance of SFT42R, according to your test?
Already very excited for LHP531 in 5000K and below!
3 Thanks
Hi @Simon_Mao ! I recognized that you are offering MCPCB for 5050 LEDs in 25mm in your shop.
But when I am looking in your LED offerings there are only 16mm and 20mm shown.
Would it be possible to get for example SFT40 and XHP50.3 dircetly on 25mm MCPCB?
1 Thank
Some thoughts in the last weeks. Might not be the right topic, but maybe it is worth sharing.
So at the end you earn money with my tests. It is good that you pointed this out directly. Maybe I should think about Patreon or a subscription model 
To be honest, this is something I was discussing with other flashlight enthusiasts and modders some days ago… I do the tests for the community, and I’m still very happy to do it (otherwise I wouldn’t do it anymore). For the modders, for enthusiasts, for people who want to develop a driver as a hobby, build their dream lamp privately or simply equip an old lamp with new LEDs.
I’m not a fan of stores and other companies/manufacturers taking my data and using it for their business to develop new products or avoid complaints or issues, especially without any mention of names. Of course I can’t prevent this, but nobody pays me for the time I spend testing in my free time alongside my actual work (except someone sponsoring the LEDs itself sometimes). It’s simply a question of fairness, but I think it’s generally a tricky subject.
Pretty sure I’ll think about a suitable license (CC BY-NC-SA or something like this). Sure, I can’t avoid someone taking my data and selling a product for money, but I appeal to everyone out there to be fair regarding test data, reviews, and other valuable information regardless of the type which helps to build and sell lights.
9 Thanks
I’m just after whether or not the 10A driver is fine to pair with the LHP531 since the datasheet says one thing and Simon is doing another. Am I wanting to know the max current to build and sell a hotrod, no.
I personally love your tests, but I’ve only used them in the past to confirm Simon isn’t pushing LEDs too hard and then discussing them on BLF or reddit. Lexel instructed me years ago to never sell an LED pushed over spec, so I actually don’t on anything unless Simon provides no alternative driver at its max current, or its considered safe from talk in the community (most likely stemming from your test).
I’ve tested many things for this community, including weeks sorting the mess with each Convoy batch change on drivers, reflectors, and my own custom gaskets I’ve made available free for all. Disappointed Simon never considered using them, it would have fixed Convoy’s biggest issue.
Not a bad idea having a patreon though since it is something you’re doing consistently.
7 Thanks
Not to mention that many companies already use BLF informations and ask for advices for a while to build/improve their products. Simon uses pics and graphs from BLF users in his website…
4 Thanks
1 Thank
12 Thanks
Thank you very, very much for this data! The numbers for known emitters seem realistic, possibly on the conservative side as some other reviewers report 2200lm from SFT40.
IIRC the H17Fx has FET turbo, yes? We are seeing more than +30% output, both initially and at 30s, suggesting a combination of lower Vf and greater efficiency. The +15% intensity is a very nice bonus. This emitter does seem to be substantial upgrade over SFT40.
Do you happen to have measurements for SFT25R in the same driver/host? Very curious how the throw compares.
3 Thanks
… that users have achieved through a lot of dedication, time and money (measurement technology, tools, computer software etc.).
Companies could easily buy all the equipment and knowledge and then use it. For Wurkkos (as an example, could be any other company), 2200 $ for a new spectrometer is absolutely no problem, but for me as a private individual it is. I would be happy to upgrade to much better quipment for providing better measurements, but it is simply not possible.
And I think that’s what I see as critical. I have designed, built and refined my measuring equipment over many months. I have (only in 2π Config) achieved an accuracy of ± 4 % for my sphere, which I was able to confirm at least with two different non-accredited calibration light sources. That’s the reason why I would never give guarantees of any kind for my measurements, but it doesn’t come from “build a sphere and bang, measurement is accurate”, but instead “good things will take time”.
(Accuracy was also a thing for me recently, because the accuracy of my equipment was questioned after doing some testing. Of course, I could be wrong, but a deviation of 20-30% (!!) in the specific topic - which was mentioned there - is more than just far too conspicuous.)
Of course, I understand why many enthusiasts want to have the companies on board - otherwise many interesting projects would not be realized. After all, many lamps were only created as a result. So don’t necessarily see my post as pure complaining, but rather as an approach to think about the topic in general.
8 Thanks
Hello, @Simon_Mao.
Looks like the “Ask a question” feature on your site is broken.
I’m sure I enter the captcha code correctly but every time I only get this
And I’d like to know if the SS bezel for C8 C8+ M21A suits M21E ?
I’m thinking about buying the L21A or L21B with the LHP73B or XHP70.3.
Is it likely that there will be focus issues and problems with the 9mm reflector as per this host with the SFT25, etc?
The focus issues are there, but won’t affect the XHP70.3 and LHP73B as both have huge emitting surface.
1 Thank
