Back in Black - SST-90 4D Maglite "Shower Head" Aspheric build.

Old-Lumens have you measured the lux number of this light yet? I am really curious!

I even doubt that if my Sanmak SM5200 can reach over one mile… your light is genuinely insane!

I love this shot . . .

A purty chunk-o-glass for sure but of course, the engineering staff at Foybezels is most impressed with that absolutely gorgeous doll of a bezel.

Somebody is going to own one hell of a flashlight. I predict it sells within two hours of being offered, depending on time of day, of course.

wowFoy

It was around 3000 lumens… with one of the driver boards disconnected and only getting 6 amps. When it gets back, I’ll put it through its paces.

3000 lumens from a single LED?! :open_mouth:

texaspyro you are the new owner of this light? :bigsmile:

An SST 90 aspheric build at 6 amps giving out 3k lumens…

That is a bit… Hard to believe.

I would have to agree.

The N bin I used will do 850-1,000lm at 3.15A. I don't know what it would do at 6A, but now it should be back to it's max, with the solder joint fixed. Possibly it would be hard to read lux due to the light being concentrated thru the aspheric lens? I don't know.

Yes, it’s mine! Muahahahahahaha J)

I did the measurement with the head off in my integrating box… I don’t have the sphere set up yet. It does tend to over-estimate lumens at high levels. I suspect it is actually something somewhat over 2000 lumens.

Hmm… I don’t know about the 3000 lumens things, really, but I am not gonna deny it though since I do not have proof.

First of all, I do not deny that this thing is BRIGHT and has Serious Throw, but still have some doubts. As claimed by Old Lumens, this thing can throw up to about 1.376 miles, which can be rounded as about 2.2km, or 2200 meters.

Assume that we define the throw distance by ANSI standard, to only 0.25 lux.

Then the lux at 1m should be (2200/2)^2 = 1,210,000 lux @1m! In other word, 1.21 million candlepower!

As far as I know the legendary DEFT created by saabluster (Michael) has “only” 700k cp. And this is done with overdriven XR-E with 101mm aspherical lens.

No offense anyway… but just, curious…

From what I have read, an “N” bin SST-90 can produce from 2,500 to 3,000 lumens at 9 amps. I’m only from reading other threads in other forums.

As far as the 1.376 miles, well there's the photos and there's the google map stating the distance. I don't know what that means for people needing numbers for proof, but for me, it was good enough to see with my eyes, that the building, (the whole dammed building), was lit up at that distance. I don't know if possibly the numbers don't show reality or what. I've seen things that defy the numbers many times in my life and to me seeing is believing. I usually trust what I see, more than what numbers tell me, but I know, for a lot of people, numbers are the only real thing out there and I am not knocking that. I know that numbers and calculations are needed for people to be able to understand or verify that what they see is real, but since I don't have a IS, I can't give anyone those numbers and I think it would take a big IS and a very good metering system to verify 3,000 lumens, not the little 12" spheres & inexpensive meters, that many people have.

I guess what I am trying to say is, I don't think there is a way to verify satisfactorily, at this time.

at the end of the day, as long as a light (or whatever) does what you want it to do, the numbers are largely irrelevant. I prefer subjective labels such as low/mid/high or, for my bike lights, dim/bright/damn bright.

Old-Lumens, you are a great modder and willing to share your knowledge with us. I have seen your youtube videos and they are very helpful to me and I believe to anyone else, so I definitely have no doubt with your modding skill and your flashlight. It was simply out of curiosity that I asked about the lux reading, but no mean of any disrespect at all.

Maybe it’s my bad english expression that cause some misunderstanding here. Yes, it is a bada thrower exactly!

I know that. I didn’t take it badly at all. I was just saying that sometimes numbers are needed, to corroborate. I just can’t give them that’s all.

It's all good.

I got it back in today after a little rework to re-attach one of the driver leads. It now draws the full 9.3A I put it into my sphere without the lens. It starts off at around 1750 lumens. Within 60 seconds it is down to 1000 lumens. After 5 minutes it levels off at around 800 lumens. It looks like it is having trouble sloughing off 30 watts of heat. The flashlight body temp is not too bad though.

The color temperature readings are a bit low. The sensor was mounted outside the sphere, looking through the styrofoam wall. That seems to absorb some of the blue light, reducing the apparent color temp.

Looks like it'a bust, all the way around. You can send it back, if you wish and I will refund the money. I am thoroughly ashamed of it at this point. I went through the soldering twice now and still a bad solder joint? I just need to stop using drivers and limit myself to direct drive.

Justin it still draws 9.3 amps, which means there is no bad solder joint. It just has 1750 lumens, which to you might be bad but I think is alright. I think what texaspyro is trying to say is that a lot of the heat is staying inside the light rather than being delivered through the body to the outside. Because judging by the lumen outputs after 5 minutes and that the flashlight body temp is ‘not too bad’ I think it’s just the fact that it isn’t transferring heat from the LED or driver to the body well enough.

Just curious, texaspyro, what are the lux readings on this one?

don’t beat yourself up OL, you did a fantastic job. Any torch/ flashlight will struggle to dump 30W+ of heat with limited surface area. Hell, even dedicated bike lights with 2-3sq.in. of surface area per watt can’t do without 3-5mph airflow, it’s just physics that’s all. Given that it has modes, I doubt it’ll be run on full for long periods anyway, not unless he wants to get arrested for misdirecting local air traffic. Just consider Texaspyro’s testing to be an extreme case.

Funnily enough, this brought to mind a complete nutter on CPF who made a 10x XM-L light for his RC plane and then made videos of it buzzing a freight train. I wonder what the driver thought?!

No way I’m sending it back! It’s a wonderful machine and nothing to be ashamed of. The repaired driver is working just fine. A solid 9+ amps out.

I don’t think that the drop in light output is due to heat transfer from the pill to the body. The pill was only gets 5C warmer than the body. I think the issue could be Luminus’s over exaggeration of the ability of that LED to handle 9+ amps. At 9 amps, the LED package must move over 200 watts per square inch of package base (9A x 3.9 Vf / 0.17 sq/in). My 150 watt Bridgelux arrays are less than 1/5th that power density and they are not trivial to cool.

I’m going to try some forced air cooling on the pill and see what that does.

Here’s a plot of it behaving rather well on medium (2.5A / 9 watts). Note that the light output is actually increasing with the temperature! It was cranking out around 525 lumens. I wonder if a high of 5 amps might actually deliver more light than the 9 amp setting after it gets cooking for a while?

I agree that Luminus’s data sheet may be exaggerated. If you look on cpf here you can see that at 9 amps an ‘N’ binned SST-90 will pull just over 2,000 lumens. But if you look at here, from what I took from a Luminus data sheet a while back:

  • Luminus SST-90 K-bin (1620-1890 lm @ 9A)
  • Luminus SST-90 L-bin (1890-2295 lm @ 9A)
  • Luminus SST-90 M-bin (2295-2700 lm @ 9A)
  • Luminus SST-90 N-bin (2700-3240 lm @ 9A)
  • Luminus SST-90 P-bin (3240-3915 lm @ 9A)

It seems the test of the N bin was around 1000 lumens less than what it should have been. I hate to be the one to say this, but if we assume that OL somehow received an ‘L’ or ‘K’ bin LED then it’s understandable why the output was like that.
That’s just my theory though.

EDIT: This whole thing is under the assumption that there is adequate heat management. If the forced air cooling noticeably bring the output up then this whole thing just goes out the window.

The part number I received from mouser, in a sealed package, was the "N" bin. That's all I know. The heatsink is way too small for this light and I only meant it for a quick turn on wow light. It was never meant to run for more than a minute at a time. In order to cool one of these in a passive cooling environment, such as a flashlight, it would take a piece of copper/aluminum the length of about 2-D batteries, plus the area above the switch and where the switch is as well (about 200mm in length). If I had used a 6D flashlight, I would have filled the first 1/3 of the light with a heatsink, but I had a 4D and space was such an issue, that I put the drivers flat, to gain a few mm of room. I never imagined it would be used for more than a WOW light. I thought wrong... See, I'm not a flashaholic, so I don't think like one. I turned it on the first time and thought WOW, but ok, what would I ever do with it? I never ran it for more than 30 seconds at a time. You guys turn them on and immediately see how many amps you can squeeze out and for how long and make charts & graphs, LOL, I'll never understand, (I am Not putting y'all down, more power to you and good for you and the knowledge you share). Definitely, this is the second SST-90 that I have tried and failed with and this is the last SST-90 I will ever waste the time on. An SST-90 at 9 or above amps needs to be in an active cooling situation. A fan, or water cooling, if it's going to run steady for several minutes. I just wanted to make a thrower that would reach over a mile, to put all the chinese throwers to shame. That was the whole reasoning for even doing it.

Anyhow, I'm done rattling on over nothing...