Convoy S21A runtime emitter comparason

They have essentially the same electrical characteristics, functionally the SFT40 is just a throwy version of the SST40, everything being equal.

In such a small light and large emitter OP versus SMO won’t make too much of a difference in throw, but try asking! If you’re going for an S21A with SFT40, you’ll probably get just a bit under 400m of ANSI throw, which translates to just around 200m of usable throw.

1 Thank

I would be pretty stoked with 200m of throw, that will be more than enough i think for the most part.
Will order one and see how it goes.

I’m actually looking at building my own custom unit at some stage with loneoceans driver, but it will be more general purpose and more floody.
That being said i would like to experiment around with different configurations.

I hope someday we will see more efficient LED technology that generates less heat, i know there is lots of R&D in this space.

I’m going to reiterate here that the linear drivers used for the SST40, SFT40, and other 3V LEDs in these models have a significant disadvantage for sustained high output. Linear drivers turn the difference between batter voltage and LED forward voltage into waste heat, and it can be significant.

A 6V XHP50.3 HI will probably have the best efficiency at high sustained output in combination with at least 200m throw in a Convoy S21 series.

2 Thanks

The XHP50.3 HI will definitely be a more efficient choice. The question is how much throw does the OP want, the XHP can probably make 200m of ANSI throw with no issues, but usable throw might be <100m.

An SFT40, even with a less efficient driver, should be preferred over the XHP for sustained throw–by virtue of having a smaller die, it can sustain the same throw while drawing way less power and thus producing way less heat. But if I recall correctly now there are 3V buck drivers available that might be more efficient than the linear drivers.

Im not sure what driver is used in the convoy, but if you tske a look at the review i posted at the top, it suggests it provides a good 800 lumens of sustained light for the best part of 2 hours, i would call that pretty impressive.
Linear drivers are not efficient, i agree, bit i thought that buck drivers are more common these days anyway and those use a switched mode power supply in much the same way a boost driver does.
Typically in that instance i always believed its more efficient to step down the voltage rather than boost it higher.

I need to look at they kind of driver used in the convoy, but i always thought linear drivers were not as common these days for that reason.
Linear drivers are typically found in more low power/low cost flashlights.
Those switched mode buck drivers are very efficient, so if this model ships with one of those, there would be no real advantage over a boost driver.

Looking at their parts store, it suggests they use buck drivers and the inductors on the board suggests they are the efficient switched mode buck drivers.
https://convoy.aliexpress.com/store/group/driver/330416_254401905.html?spm=a2g0o.store_pc_allProduct.0.0.19014989KFN02R

Convoy has buck drivers, but I don’t think they quite have the bugs worked out of them yet and I believe they’re not shipping by default in the S-series tube lights that could use them.

In theory, buck drivers should be more efficient than boost, but a major downside relative to what you’ve requested is that single-cell Li-ion buck lights usually can’t maintain their maximum output as the battery drains. Recall the flat runtime graph of the boost driver S21E and contrast this buck driver light:

1 Thank

I messaged them to ask about what driver tyey use but have not heard back yet.
I dont know what you think of this, but here is the runtime graph of that light with the SST40.
Seems pretty efficient to me if it can sustain over 800 lumens for the best part of 2 hours?
I dont think you would see this performance with a linear driver?

Let’s call the output in that graph 850 lumens for 2 hours or 1700 lumen-hours. That’s not quite right, but it’s going to make the math really easy when we call the battery 17 Watt-hours, which is also not quite right. 100 lm/W with an SST-40 in the 800-1000lm range is about what I’d expect from a linear driver. In this test, the emitter is making 180 lm/W at 900 lumens.

1 Thank

Yeah, looking at that data, it performs better than i thought for that LED, looks like 900 lumens is 6W.
So with a decent 5ah battery you should get the best part of 3 hours runtime with a decent driver.
Will be interesting if they can switch the driver out for me or not if it is indeed a linear driver.

Edit, while writing this, the convoy store responded and said its indeed a linear driver, and can fit with a buck driver, but warned me it would get hot, but this wont be an issue if i run at 6 watts.

The SST-40 is, indeed very efficient, though there’s a big cost in terms of green tint and very low CRI (under 70).

This review of the S21E with high-CRI XHP50.3 HI doesn’t have graphs, but does include a useful data point:

On 100% mode, maximum brightness was sustained for a fairly impressive 110 seconds, followed by a gradual decrease to 720 lumens by 4 minutes, where output stabilised, with only a small reduction in brightness during the rest of the useable runtime. With a 4000mAh Samsung 40T 21700 battery, low voltage warning (red flashing switch and brightness stepdown) kicked in at 133 minutes.

That’s roughly 110 lm/W: slightly more efficient, but with neutral tint and CRI in the 90s.

1 Thank

Thats the linear driver for sure.

Oops, missed the comment where you said you already found that out lol

Btw, the s21b has better thermals and more driver possibilities. I think it looks sharper too

1 Thank

Might take another look at that, when i was looking at reviews, the runtime didnt appear flash, but it probably had a different emitter

Youre probably looking at graphs of runtime on 100%. The 35% mode on the linear driver is known to be pretty flat. The emitter wouldnt change that. Unless theres something way off about it

Well I have ordered a unit with the SFT40 and a buck driver, will report on how well it performs.
He told me it will get hot if run at full power for extended periods, but I wont need be running too high power with this most of the time.

That’s great! I have a 6A Convoy driver in my M21A and 3 amps is sufficient… high is almost too much. The reflector in the S21A is smaller but the throw with an SST40 was impressive. The SFT40 will be better.

1 Thank

Cant wait to try my one out then.
He told me that the driver is ratedat 8A which will be more than enough i thinn!

That looks like my XRP graph😁

Light arrived yesterday and i tried it out.
Runs pretty well and as I expected. Something I noticed with this is that it shipped with a de-domed LED!
I just need to figure out how the configuration goes on this, its got a fair few modes to cycle through, but I knew that when I bought it.
I Just want to have 3 or 4 levels of brightness and leave all the flashing modes out of it if possible.

That’s what the sft40 is

I like mode #10 personally

1 Thank