Been reading this review here
Runtime looks like it stacks up good, ideally im wanting a nice thrower that can run close to 1000 lumens or more for a good 1-2 hours.
However, I was looking at getting a model with a 519a or perhaps the XHP50 but would like to see runtime/brightness data across different emitters to see what one provides the longest runtime with the highest brightness.
If anyone else can report on their data or point me to reviews/test data, that would be helpful, TIA
S21A is everything you’re not looking for.
Thrower implies big head and high intensity, and sustaining high output over time requires large thermal mass, large battery capacity, and a highly efficacious (non-high CRI) emitter. The S21A with a single 21700 and a 519A is not what you want.
Try your luck with the Convoy L21B with SFT40 or XHP50.3HI. Or one of the larger SBT90.2 offerings.
The only throwers resembling the S21A’s tube light form factor use a laser-excited phosphor (LEP) to produce an extremely narrow beam with no spill. They have low output and extremely high intensity that translates to long throw distance. LEPs are very expensive.
A few lights using the high-intensity Osram LEDs manage around 400m FL1 throw in an 18650 or 21700 tube form factor. The S21A does offer the CSLPM1 (“W2”, “white flat 2mm”) LED and probably does hit 400m+ FL1 throw with it.
This post doesn’t mention throw requirements though; it only talks about high sustained output. With Convoy lights, the type of driver used is as important for that as the LED. Some versions use a linear driver, which has an inherent loss of efficiency when the battery voltage is higher than the LED’s requirements. Models with 6-volt LEDs like the Nichia B35AM and some (but not all) XHP50 variants use a boost driver, which does not have the inherent waste of power.
So if your goal is high sustained output, not throw, color rendering, or anything else the S21B with XHP50.2 is probably the best performing 21700 tailswitch Convoy tube light. You might need to ask when ordering to make sure it’s the high-voltage XHP50.2 with the boost driver though.
According to the review, it runs at 800 lumens for a good amount of time, hence why i wanted to compare runtime with the various emitter options it comes in.
I thought that the optics determined the throw more than anything. Im wanting something more like a spotlight and less floody.
I thought there should be plenty of 21700 lights out there that can sustain close to 1000 lumens for an extended time.
For example, a ledlenser p7r can run for up to 2 hours at 1000 lumens.
I like their focusing system lots but their quality is nothing special for the price.
I dont need anything that fancy i dont think.
I just need 1-200m throw would probably be enough and ive seen this on similar sized nitecore lights for example.
I mostly will be using it as a spotlight for hunting and will mount on a rifle at times, hence why i dont want anything too big and chunky.
I do like warm colour temperature and high CRI if possible, i was wanting to see how it stacked up with the nichia 519 emitter.
S21A with SFT 40 diode will have solid throw.
Now Convoy store has is in 5000k colour.
Problem is that flashlight don’t have enought thermal mass to prevent overheating and sustain high output.
100% mode will stepdown to 35% but without cooling it will hot for 10-15 minutes that it will be uncomfort to stay in your hands.
10% mode will be cool or eventually 20% mode if you choose group which has that mode but it is a question will You be satisfied with that power.
I thought there should be plenty of 21700 lights out there that can sustain close to 1000 lumens for an extended time.
200m throw would probably be enough
I do like warm colour temperature and high CRI if possible
That combination is actually kind of hard. The Convoy S21E with Nichia 719A 3000K can do it, but only with a cooling fan blowing on it. The 719A has more throw than the 519A and B35A at the same power levels, but if you can do with a bit less than 200m in this form factor, the B35A will probably work for you.
The B35A has similar tint and CRI to the 519A (neutral and outstanding, respectively), but higher efficiency. Since it runs at 6V, it requires a boost driver, meaning it should be able to make flat runtime graphs like the 719A. It won’t generate as much heat, so it might not even thermal throttle.
For example, a ledlenser p7r can run for up to 2 hours at 1000 lumens.
No it can’t. Its output is not stable over its runtime like the S21E/719A is.
I think I now see your needs. The optic used can retain or reduce throw, but head diameter is a hard upper bound on how far a light can throw.
The difficulty you face is the fact that the three quantities you care about: throw, efficacy/output, and warm + high CRI, are in tension with each other. Optimizing any of them comes at the expense of the remaining parameters, so the best you can do is try to strike a balance or sacrifice some of them.
To get 100-200m of usable throw, you need at least the head size of a M21B, if not M21A. The SFT40 in 5000K has top-notch throw and output/efficacy with good tint, despite being a low CRI emitter. I generally find that CRI is a bit less important for long-distance (thrower) applications; CRI only matters when the light is used with high enough intensity on the target to engage photopic (color) vision; this does not happen often with a thrower.
Im not too familiar with those diodes, i cant even find the manufacturers web page and/or data sheet.
I see the review i posted was for the model using the SST40, that was running over 800 lumens for the best part of 2 hours, that would be quite acceptable to me.
I could probably add a custom mode and lower it to around 700 lumens and get even more runtime.
If its easy enough to switch between modes, i could probably get away with running it for shorter periods st higher output when i need it, that would make things a bit easier. Your right that the thermal load affects runtime significantly, thats why you see many of the higher powered models with fins machined into the body. Over winter months when its colder it probably helps in this regard too. Looks like some of the convoy models are advertised as having a copper head, i know most of them are brass that ive seen, but the copper ones probably help dissapate heat away better?
If i can get a tighter beam i would be able to run at a lower putput quite easily.
Really need to compare brightness to runtime ratio and go from there.
The B35A looks like it could be worth exploring, especially if i can get a tighter beam.
Speaking of optics, there are a ton to choose from. Whats the story with TIR optics?
They typically can give a good throw if you choose the right one.
I personally prefer warmer colour temperatures (around 4000K), but i could afford to go for a lower CRI, you make a good point that its less important at a longer distance.
I know higher CRI emitters are less efficient, but does the same thing apply with color temperature?
Yes, though the difference is not as drastic as the difference between a 70 CRI emitter compared to a 90 CRI emitter.
With everything else the same, a 6500K emitter will be a bit brighter than a 5000K emitter, which will be a bit brighter than a 4000K emitter, which will be a bit brighter than a 2700K emitter, etc.
I like really warm tints, so I go for 2700K even though it’s slightly less bright than cooler tints. 
By the way, if you want a thrower that isn’t too big, I suggest the Convoy M21 series with an SFT40 emitter.
The bigger the reflector, the more the throw, but of course the less pocketable the light will be.
The SFT40 emitter is the best practical throwy emitter, in my opinion.
You can get more throw with the SBT90.2 emitter, but that emitter is not very practical (it produces a massive amount of heat, and it has short high-output runtimes.)
With larger lights you will see overwhelming use of reflectors over optics. This is because TIR optics are filled-in solids while reflectors are surfaces, so with increasing size TIR optics become much heavier and therefore less favorable. The S2+ and S21A have nice optics, but they won’t give you the throw you want.
You can get an upper bound for the intensity at 1m of a Lambertian source in an optic/reflector: (output in lumens)(frontal area of optic)/(pi times area of light-emitting surface). Notice that the type of optic/reflector does not come into the equation; only a larger diameter (i.e., larger frontal area) can raise the upper bound.
Warm emitters are less efficacious than their cool counterparts. (Note that lumens/watt is efficacy, while efficiency is unitless.) This is because warmer emitters produce more red light, which the human eye perceives as less bright than green/blue light, which cool emitters make more of.
Nice explanation!
I doubt I’ll remember all of it for next time, though. 
Looks a good option, its tempting to grab one for the price and if i find its not suitable, I will still have a good use for it anyway.
Is there much difference between an SST40 and an SFT40? I see both options are available.
Also reading a good review with the B35A model, slightly less output but has high CRI, not too sure how throw would compare between the B35A and the SFT40, but if the SFT40 has better throw, I think I will go for that.
I also believe its best to go for an SMO reflector to give the best throw too.
That makes sense, so its not really less efficient, its just our eyes are less sensitive to red light, so it appears less bright to our eyes.
In theory it wouldnt make much difference with measurements unless the equipment is designed to compensate for the red shift for the human eye?
Huge difference between SST40 and SFT40–the SFT is better in every way, especially throw, it has almost 2x the intensity without sacrificing output. And much better tint uniformity.
In terms of throw and output, the B35AM does not compare to the SFT40.
If the equipment measures in units of lumens, then it does compensate for the luminosity function (i.e., how bright each wavelength appears) of the human eye. The lumen is not really a physical quantity and depends on how humans perceive light.
In general, sadly, warm and high CRI emitters are also less efficient, aside from being less efficacious. My guess is that red phosphors aren’t as efficient as green phosphors. Also any sort of conversion of blue light by a phosphor is lossy, and more conversion happens in a warm or high CRI emitter, so there are more losses.
OK sounds good, i expect there is not much difference in power consumption between an SFT40 and an SST40?
I dont see the option listed for an SMO reflector, so will have to ask them if they can swap it out or else I will order one separately.
