Okay, I don’t hate them, but the inefficiency of the drivers that everyone uses drives me crazy.
For example, when I compare output of a Zebralight (boost driver) with an XHP35 emitter, to something like a D4 or FW3A (linear driver + FET) with a XPL emitter, the Zebralight gives me way more run-time at every output.
Just comparing the 1x7135 regulated output of 350mA to a comparable regulated output on the zebra, I get about 8.5 hours with the linear driver, but 12 hours with the boost driver. Almost 50% more run-time with the boost driver!
And it’s no wonder, when I look at the XPL spec sheet. The forward voltage at 350mA is only about 2.9v. That means the extra voltage from my 4.2v lithium-ion battery is just being thrown away as excess heat! And I can feel the D4 or FW3A light get warm, especially when the battery is fully charged and at a high voltage. The Zebra, with its boost driver, runs cool, since it’s not throwing away that excess voltage as heat.
And, when it comes to the FET driver… well, it’s even worse at just about every output except max.
Why can’t someone build a budget light with a buck/boost driver???
It can’t add so much to the cost as to make it unaffordable. The extra cost can’t be more than a few dollars. And it would be well worth the increase in price.
I know 7135 chips are cheap and easy, but can’t we finally move beyond them? Going to a buck/boost driver is like suddenly adding 50% to your battery capacity.
You’ve got to keep FET lights topped off after every use or you will see diminished output next time out. I’d rather a flat line output. I’ve got a Convoy M3 C sitting in NYC customs I had to buy after seeing the run time graph showing 2+ hours of 1500+ lumens until battery drains.
I care. I’m all about capacity. But to imply that I’m too stupid to care because I buy $14 Convoys instead of $100 Zebralights is kinda elitist, don’t you think?
The issue is it’s hard to put boost/buck and budget together since those drivers are A) more complex and B) more expensive. Plus you won’t get the highest output with a boost/buck driver since the output is current regulated to what the inductor can handle. Multi emitter lights are especially taxing on these drivers since total output would otherwise be lower than a 7135 +FET setup. Until someone develops a cheap to mass produce 17mm boost/buck driver, we’re stuck with fet+7135 for high output multi emitter lights.
Well, that’s what I’m wondering… why hasn’t anyone done that? Or, has it already been done, but nobody is using it because it’s easier to sell something for $5 less when nobody seems to care about efficiency?
I can’t believe Zebralight’s driver is that expensive. It can’t be more than a few dollars, at most. They can put out 3000 (high-ish CRI) lumens with it, in the SC700d light, which is a respectable amount of output for a small light.
I think the budget light manufacturers are stuck on the FET+1 driver, simply because they know it, and all anyone seems to care about is max lumens.
I just got a Sofirn SP33 V2 in today with boost driver and 6V XHP50.2, it can hold just over 1000lm for 2.5hrs according to Zeroair’s review and was only ~$23 when new (now discontinued). Currently waiting for a 90cri 4500K XHP50.2 to be delivered so I can swap it in. There are still good single emitter budget lights out there, like the Sofirn SP35, Sofirn SP31 V2, Wowtac A5, and the Convoy lights with buck or boost drivers for example. But like people have said before, the majority of consumers just want the must lumens for the least cost and FET based drivers excel at this. Similar reason why low cri cool white emitters are still the most common.
Convoy has a 2.5A buck driver meant for red emitters with low Vfs but I think it should also work with white emitters, although the modes are weird. I don’t know how efficient it is but for a qty of 5 they’re less than $3 each. It seems like it should be possible to have a ~3A 17mm buck driver with good efficiency for < $10.
Personally I would happily pay an extra $10-$20 for a high efficiency buck or boost driver in most of my lights, specifically my Emisar/Noctigon lights.
Well there’s ways to make a lot of lumens-a single high output led like a xhp50.2 or xhp70.2, or multiple leds. It’s easier to drive a single led with good efficiency. Like the Convoy M3 or M21 with xhp70.2 6V. You can run those for hours on a 5 Ah battery at 1400+ lumens because they aren’t being driven that hard to generate that amount of light. In that case a boost driver makes sense. You could do a buck driver, but max amps would be 8 or so, and sustained brightness on something like a sst40 would be a lot less.
Has anyone measured the efficiency difference between 7135 and FET? My MF01S 4000K is FET and I feel that it gets too hot for its output and the small SST20s are not really helping with efficiency/heat. Even 1000lm gets hot indoors (no cooling).
Dude, loneoceans has literally announced two amazing switching regulators in the last year. I don’t know what you mean that there are no 17mm options, he already developed the drivers, released the boards, and made them BLF firmware compatible to boot.
While switching converters can have advantages in efficiency and regulation, linear drivers are not as terribly inefficient as you might think.
With your 1x7135 XPL at 2.9V, the average efficiency over the discharge of an average cell is 2.9V/3.65V=79%. At the beginning when the cell is full the efficiency is worse, but it really practically only makes sense to consider the average efficiency over the entire discharge; that is what correlates with the runtime.
Now, 79% is not great. A 95% efficient driver would get you 20% more runtime, 10.3h instead of 8.6h on a typical 18650.
For higher power settings the efficiency is better. For example running a SST40 at 4A would get you an average efficiency of about 90% (3.25V/3.6V).
1. The average voltage (50% drained) is closer to 3.75v on most of my 18650 cells.
2. You can’t use amp-hours for the comparison, you have to use watt-hours. A buck/boost driver uses watts to power the LED, not amps. There’s more watt-hours in a battery at the top-half of its capacity, than the bottom half. Obviously, 1 amp at 4v gives 4 watts, whereas 1 amp at 3v only gives 3 watts. So, a light with a buck/boost driver will drain a battery more slowly (uses less amps) when it’s full, than when it’s half-drained. A linear driver just sucks the same amps all the time, and only becomes as efficient as the buck/boost when the battery is almost empty (2.9v).
Bc of the boost driver? I have had my eye on that light, and only this morning realized it must have a boost driver. That makes me really want it. Is that why it has that susta8ned lumens you mentioned?
I have a Zebralight. It’s nice, but it gets hot too. And if it is any more efficient I can’t tell that when I EDC it. I seem to change batteries in it just as often as any of my other EDC lights.
Either way, I have loads of 18650’s and recharging a battery is hardly difficult or a deal breaker.
But then again I also have a V8 Camaro (yes in the UK!!!) and a V8 Range Rover instead of a 1.2 litre hatchback. So maybe I just don’t really care about such things anyway and I’m just happy to enjoy them all for what they are.