Well I guess I mean more efficient on the emitter side of things. Does driving it with direct current produce less heat in the emitter than pwm? I know TK was describing how pwm only turns the emitter completely off when a single channel is being used, but anything higher and the emitter is never fully off. I feel like PWMing a fet produces more heat than PWMing a bank of AMCs. But I’m just stabbing in the dark here trying to understand.
Edit: I’m thinking about this because contactcr said it would be hotter on the emitter using a ta driver vs boost
You are right, PWMing a bank of AMCs is better than PWMing a FET.
But Buck-boost is better than PWMing AMCs.
When using a bank of 7135s (7 or 7+1 in this case) the total current is usually above the efficient point for the led
and that is only on the emitter side of things, the driver part needs to be added as well, heat will combine from both these parts.
7135 chips will burn excess voltage (Vbat - Vled) as long as being regulated, then they become more efficient but unregulated.
I don't think there's any difference between the 7135's and a FET for heat at the LED. For example, driving the LED from PWM's at 2 amps on a bank of 8 7135's (2.8A max) vs. a FET should see no difference in heat, even though the spikes from the FET are higher amps than the spikes from the 7135's -- I never heard of differences in respect to heat or output. Efficiency off the cell and maintaining that 2 amps while the battery depletes though is another story.
Yes, a given input power will produce roughly the same heat regardless of the driver. The difference is the amount of lumens you will get.
All that said, after years of messing with different drivers and LED’s. I tend to prefer the PWM setup over constant current for many applications. It is less efficient but not as much as some would have you think med-high levels. At very low levels is where a buck or boot driver really pulls ahead in efficiency. At higher output levels the difference is not worth talking about.
The downside to contact current drivers is that the LED tint will change depending on the output. With PWM the tint will only change minorly, with the PWM tint generally looking better. This is a big part of why I just use PWM in most of my lights, the tint is more consistent and better looking in almost every case. Worth the loss in efficiency for my uses.
I really like Boost and buck/boost drivers and they have their place but at this time we don’t have any with enough power output to compete with FET, so they are limited to low output long runtime lights for me.
Okay, I just finished 2 mods:
Amutorch X9 XHP50.2 3V
I kept the stock driver + wires.
20A from a black Liitokala, so might be in the 7000 lumens range ?
I also put one in my Emisar D1…
NOW that’s a hotrod !
15A from a 30Q, it heats faster than my SST20 D4 !
I have no idea of how many lumens it does since I don’t have an integrating sphere but it has the same ceiling bounce lux measurment than my stock SST40 Amutorch X9 !
You are really putting your marketing hat on right now to spin this to sound like something other than what it is.
Unless i’m way off base the obvious comparison we are trying to make is heat vs output. If you put the Boost driver at 500lm and the FET+N+1 driver at 500lm the FET will be hotter because it’s using more “input power”.
I’ll be the first to say most of my mods are FET based Anduril or Bistro drivers regardless of that fact. The UI, the sheer brightness, the simplicity, in some cases like you say PWM tint is better, cost. However, lets not pretend it doesn’t have any trade-off.
Excuse my lack of understanding.
But he did say there is a little bit of a difference, maybe you’re both describing the same thing, but understand it to have different degrees? How much hotter are you talking here? And are you talking about heat from the emitter? Or heat from the emitter/driver combo? Because I can imagine that a boost driver runs a little cooler, but I have no idea how much. Is the amount of extra heat we’re talking about here, negligible? Or like, way hotter?
I plan on putting this emitter in an H03 with 1+7+fet driver. My usual use case is just a couple hundred lumens for up close work, maybe 7 or 800 lumens for outdoor work, and then turbo for fun/to see something farther away for a minute. Is this xhp50 gonna cook me in my first two normal use cases? Vs the lH351d I’ve got in there now?
Edit: also, very cool X3! I’m super curious about that D1 too. Would be interesting to see what it does in a D1S too
Notice in the post I replied to he asked if the same input power would be roughly the same Heat. This is true and what I said.
I also said that the constant current driver will have a bit higher lumens, which it will but the difference will only be noticeable at low output levels. At medium to high modes the difference is not worth talking about.
This is all fact and I stand behind it.
Even at lower modes the difference in heat will not be nearly as bad as many people think. Most would be hard pressed to notice it without some kind of equipment. The most noticeable thing will be the reduced runtime. Although even this would be hard for the average user to notice if under normal circumstances of a little here and there and not a non-stop run.
Now my personal preference is that I prefer PWM drivers since the tint does not change as much between high and lower modes and the tint as a whole is generally better. Although this is still fact based, many simply will not care about that.
I used a generic butterfly XM-L2 centering gasket, which perfectly match the emitter and the reflector hole.
The beam is well focused, quite a big hot spot of course, and with the typical yellowish corona around it
TA and I are both “right” just putting more emphasis on different points, no sense going further really.
The H03 is extremely low mass host but if you are managing with a FET+N+1 driver in it now at the modes you want it will probably do fine at similar levels. On turbo and with a high drain battery it will cook you much faster, no doubt about that.
The question asked was relating input power to heat, not heat to output. so you are off base. But in his answer TA also addressed heat vs output by explaining that with the same heat the lumen output will be different, so i don’t understand why you’ve challenged his answer.
Can you tell me which bit of his answer is ‘something other than what it is’ please? And what trade-off he is pretending doesn’t exist?
Best guess, 17.2A on the 3V should be comparable to 8.6A on the 6V, and according to Djozz’s and Texas Ace’s 6V test that equates to around 4700/4800 lumens, and they tested warmer tints.
So, a lot How long before it’s too hot to hold?
Those are raw emitters that were tested. In a light you have losses from the reflector, lens and driver. Also, those tests were before djozz and TA calibrated their spheres with Maukka’s lights. So maybe closer to 3,800 to 4,000 lumen? Just a guess.
How do you plan to power the XHP35 in the S41? It runs at 12V and the S41 is a single cell light. So you would need a boost driver but I do not know of any boost driver powerful enough for that.
How long before it’s too hot to hold?