What happens if you set the ambient temp to 1 degree C?
Maybe it is a bad sensor and you can fool it.
The sensor is (probably) fine since it increases to correct itself. It could just be the rate of change.
I’ve tried low ( not that low) and high on both settings —same result—- I’ve had lights act like this with ground issues but the mainly start off low and slowly ramp to high —- we need some of these GURUS to chime in here—— 
Like I said — in the real world uses it’s not very noticeable —- From Turbo they have always dropped like a Heavy Rock
Last night I ran mine on turbo for 30 seconds.
it did not step down, I had the temp set for 70C
I read the temp from the light, it was 69C
I was tossing it from hand to hand, it was too hot to hold!
I ran my Fire Flies E01 for 90 seconds on turbo.
It did not step down.
It was hot but not horrible.
It showed a temp of 46C
And the 50E battery still checked at 4.1 V
Toykeeper should know, she wrote the code. She’s a night owl, so we have to wait a while.
30 seconds on turbo is way too much for the FW3A. You are purposely defeating the temp protection and hopefully you understand those risks.
It was a test.
I agree, turbo in this light is a few seconds only.
But no ill effects so I got away with it this time!
With certain cells the light is in an instant ramp down from the start —- I usually use Pany Bs in this light — with the XPLs it reaches around 2100 + lumens with an instant ramp down —but it still gets hot as a ———————————- in no time at all —- I find anything above 850 lumen is really pushing it in these lights
I use Efest (35A supposedly) and Orbtronics (20A).
They hold turbo as stated above.
With xpl-hi and a 30Q I get 15 seconds of Turbo (before I calibrated the ambient temp it was 12 seconds). A Panny B is weaker and should last a little longer before stepping down. It’s not instant unless the light is already very hot internally. My 15 seconds is from room temperature.
Are you saying “instant” for dramatic effect, meaning Turbo is very short or do you mean it’s literally stepping down right away (within 1 or 2 seconds)?
Does it blink 3 times then step down? That would be low voltage protection. You didn’t mention a blink, so it’s probably not that, but I wanted to mention it just in case.
I’m talking about as with any light in a sphere/tube —from the initial turn on point there’s a slow ramp down ( why most specs are at 30 sec etc ) with many lights it stops are slows considerably —- with a fet driven light — depending on the cell it continues to slowly ramp down —- with the FW3A using a Pany B or old Samsung laptop pull it’s a continuous ramping down because even though the output is lower it still gets hot therefore kicking in the thermal protection
Just to be clear at the top of the ramp your cell type should not matter much, only the voltage since it’s using 7135 and no FET.
I would think with a half charged cell my light could sustain max ramp longer but I think the code errs on being more safe than necessary. I think there’s room for improvement here. Wish more lights had a temp sensor on the MCPCB.
Pretty much so —- But there are slight differences between a Pany B and a VTC5A at the top of the ramp
I kind of like Turbo timers — I guess because I’ve fooled with them a lot more —— A S2+ running guppy I set the Turbo at 20-25 seconds it’s done —- You still have to work around Heat issues though—-But manually
Could anyone tell me how to activate manual memory? I’ve set the level where I want and clicked 5 times, but it’s not working. Thanks.
The FW3A doesn’t have manual memory by default. Could be accomplished by changing the fw.
Hmm, so either this diagram is wrong or I got an older version of the firmware I guess.
When was that diagram added? I’ve never seen that one before.
It shows a lot of things not on the FW3A, like aux lights.
Maybe this will be the newest version of the FW3A which is not out yet (iirc).
That is the chart from the D4V2 and the E07
Note the reset.
Pretty sure it just links to a picture TK changed