Oh you have an sbt-70 aspheric? Cool.
What kind of lens did you use in your one?
It’s definitely a small spot, but don’t quote me on the exact size just yet.
Don’t have the focus dialed in right for maximum throw yet, just did a quick test getting the die roughly in focus at that range.
So I suspect this isn’t ideal for longer ranges and that may have a knock on effect on the close range spot size as well.
The Courui is actually a great host for dialing in focus, since you can fine tune the distance between the die and the lens just by unscrewing the “head” a little bit to verify if you’ve got it right.
I don’t recall exactly where I got the lens from. It was the first lens I ever bought, I was using it in a different light previously. I have bought about 3-4 lens’ since then but they were all rubbish so I’ve given up looking for another lens.
After posting that last post I was laying in bed and i thought ‘why didn’t I compare the deft-x?’ Oh well, I can do that tonight maybe.
I’d be interested in hearing a spot size description for the deft-x as well.
—
Good news and not so good news.
The good…
Driver is finished and working great, I dug through my parts box and switched to a different fet with a lower Rdson and added a second one in parallel to reduce that resistance even further. Should be around 2.5mOhm now. My lowest resistance fets, some vishay ones, unfortunately didn’t play nice with the op-amp output. They were the first that didn’t work and produced some nasty oscillations that totally killed the mode range.
Annoying, but I’m not doing anything to clamp down on oscillations on the opamp feedback circuit so it’s bound to happen with some components.
Every other fet I’ve tested on this driver has worked great and with the currently installed units it’s even compatible with 17khz fast PWM while maintaining a full mode range. No flicker or audible whine, so it’s just like a 7135 driver but without the headache of stacking 35 chips! Sweet
Current sense resistors are 2x 0.01Ohm and 2x 0.05Ohm all in parallel which gives me a regulated 13Amps to the led.
I was too caught up wrangling all those wires into place to take more pictures. But you can make out the two fets soldered to the copper heatsink piece and the black tape wrapped driver. Yellow wire is gate signal and the driver is powered through a reverse clicky.
The ground connection is through a brass bolt and copper tab screwed into the body of the light. I didn’t trust the contact board to make good ground contact, it’s a poor design aspect of the Courui that the driver board relies on an bad interference fit and pressure from the batteries to make ground contact.
-
Now for the not so great news.
The light is very impressive indoors and the lovely round hotspot with pencil thin beam is delightful. But in terms of throw figures and outdoor wow factor it’s kinda underwhelming. Of course I knew this wasn’t ever going to be a throw champ but still, the combination of very narrow beam and sorta weak maximum brightness makes it surprisingly dull outside at night.
Bearing in mind I live in a heavily light polluted spot and the moon was pretty bright last night.
But still, I had my standard reflector-ed Courui with me for comparison and subjectively it blew the mega lenser away in terms of perceived/useful throw and of course also in total output.
Even though the standard Courui is reading at less than half the candela of the Lenser I didn’t really spot any targets (100-200m range) where the Lenser illuminated it visibly better than the Courui. Of course the high CRI of the sbt-70 definitely showed much better definition of reds and browns of tree bark for example. But because the spot is so small and the area illuminated so narrow its hardly a very useful feature.
I don’t think it would need to be all that much brighter to be impressive, I was just hoping to be awed a little more.
So a pretty disappointing first field test tbh…pointing the light up into the sky is still awesome though. It’s a true white laser beam in that regard, and if there’s fog…well… J)
Initial readings at 5meters
Courui Lenser: 257,500cd (SBT70 - 13Amps) Dirty lens, focus not dialed in perfect etc but should give a good first impression
Courui Reflector: 97,000cd (XML2 - 5-6Amps) partially discharged cells, possible grounding issue/flicker Edit: Grounding ring improved and reading now shot up to 165,000cd
—
I need to take some proper measurements and verify the throw figures for both lights, and I need to verify if current draw is actually reaching 13A on the fully assembled sbt-70 light. But I think it’s pretty much doing fine at the driver level.
So I’m thinking I will have to try some light recycling to try and boost the lux to acceptable levels, the real problem with having a lens so far away from the led is that the vast majority of the light output is not hitting the lens at all.
I also have a 95mm PCX lens that I could try. It has a much shorter focal length, and while that produces a far larger spot (2.5x wider at 10 meters) it may be more impressive/useable outside at night. I will also give that a go and see if I like it better.
Frankly my favourite use of this light right now is as a true mule, so I can admire the gorgeous high cri output of the led. I’m tempted just to hang it from the ceiling like this and use it as a worklight
I’m working on an aspheric light for a friend and am finding similar boring results. This one too has the aspheric quite far from the emitter, an opposite of yours in that it’s a de-domed XP-G2. Very disappointing end results though.
Like you, I really like this as a mule. It’s just that pretty an emitter and makes up for it’s lackluster performance. I actually ordered a second one and will most likely use it in this configuration in a smaller light, just to look good.
Yeah, I guess it’s also my lukewarm feelings towards aspheric/lenser lights in general that are tainting my expectations. The circular spot fools me into thinking it’s directly comparable to a reflector light when it really isn’t.
I took the light out again tonight without the regular Courui and I had more fun, I suspect I should have gone with the cool white variant of the emitter though. Just to get that extra cool white brightness pop off distant targets, not to mention the extra output would have been welcome.
Anyone in Europe have a Wavien collar they’d like to sell?
I’d love to play around with one of those things.
I would also love to play with an Wavien collar but they are just crazy expensive
But you looks to be a handy guy & you can design stuff in 3D, maybe you can get something fabricated that works just as good. Check out borg’s post #51 in this thread
And comfychair made one also, unfortunately he seems missing from BLF for quite a while & all his pictures is off line, from this thread from post #6
Thanks for the links, I’d seen Comfy playing around with diy collars before but Borgs cool machined aluminium piece looks great.
Am I mad or would something like this be a great starting point for making a collar?
Polished up on the inside (I think it’s aluminium, hopefully it can be mirror polished) and ground top and bottom to get the right dimensions for the openings.
Hopefully they are actually nice and spherical as that would seem to be the most critical aspect to it’s effectiveness as a collar.
Yeah maybe it could work, i could be willing to take a stab at testing if i just could find some good starting point, where did you found those?
I really hope we can get some more discussions & modding on collars going here on BLF, because that looks to be an massive potential booster for aspheric lights. The problem is Wavien collars are anything but budget, it would be great if we could find a easy way to do them our self, so it is not an question if we can afford it, but it is just one more thing we always do when we want to to max out our lights.
Sorry, I forgot all about it as I’ve been busy lately.
I measured the deft at 12m. It was out of focus at such close range but measured it anyway and got around 32cm across the square die image.
As far as cheap collars, the mag reflector is hard to beat. I have a wavien on the sbt-70 aspheric but if I were to do another recycled light mod I would be using the mag reflector. It is marginally less effective than the wavien, based on my own dodgy foolings, but not enough difference to justify the price.
Also the sbt-70 light is more practical than the deft. I went camping the weekend just gone and a portion of my lights are throwers which weren’t really any good for the area we were in (hills and tall trees) a floody multi emitter would’ve been more practical. Of course, I know these things get built for the fun of it 8)
I ordered 5 to have a play. If these offer a good starting point and can be polished up properly then they might be a good cheap option.
Using a parabolic reflector for this always sounded kinda wrong to me but you can’t argue with results.
I’m also worried about getting a very clean recycled light image to avoid having a bunch of stray light bounce back and illuminate the noctigon/emitter package around the die. My initial thoughts lead me to believe a spherical reflector would produce a cleaner result but I could be wrong.
If it results in higher lux but at the expense of a messier projected hotspot then I wouldn’t find that acceptable.
I thought the Mag Incan reflectors were more spherical. Didn’t I read that somewhere? The Mag LED reflectors would not be a good choice, but I’m almost certain I saw someone say that the Incan ones work fine. Anyway, if you already have those spherical cups on the way, they should work fine as well, with a good polish and careful grinding for the right focus of all excess light back to the emitter die. If it were me, I’d aim to miss the outer edge of the lens completely. In other words, make the outlet hole of the Wavien collar small enough that the light coming out stays entirely within (slightly less than) the viewing area of the lens.
The MiniMag reflector is a little bowl, like the top cut off a ball. That would be the way to go.
I tried making my own reflector a couple of years ago and got caught up focusing on the emitter hole, by the time I realized what I was doing wrong it was too late… I cut a bowl shape that is nearly perfectly round. Of course that didn’t work with the LED.
I tried that on top of the de-domed XP-G2 in this aspheric I’m working with. It did improve output, but only about 17%. Still FAR less than expected. This is a pretty long focal length lens, and it seems to me that’s the problem… the light is getting too scattered before hitting the aspheric and losing a lot of intensity. It’s running about as hot as can be expected at 4.86A to the small die. The guy I’m trying to help out with this has spent a lot of money on it already, and it just underperforms so horribly that it’s frustrating.
The aspheric is from a name company that specializes in lenses, I don’t think it was their most expensive but it also certainly wasn’t cheap!
I’m 99% sure those incan reflectors used in old maglites are also straight parabolic reflectors, they simple have more of the base of the paraboloid visible compared with an equivalent led reflector.
It’s the same thing as I showed with this image comparison.
Of course that’s probably why they work acceptably well in this application because the base of the parabola more closely approximates a spherical reflector. Tune it right and a good portion of the light that hits the reflector will get returned.
It’s not ideal though and I would think quite a lot of light get’s bounced to areas where it shouldn’t when it hits the parts of the reflector which don’t fall in line with a spherical surface.
Ahem…You mean the “BLF Baking Tray Dust Cover” - Patent Pending*
* No association with RLT or Wavien, any similarities are purely accidental. Polished internal surface is for aesthetic purposes only. Is not designed to or intended to increase total output of any led. Any increases in visible or measurable Lux are purely accidental and an unwanted side effect of the primary aesthetic and dust repelling purpose of the baking tray appliance.
That’s a very good point though. I’ve noticed that the blue fringed corona around the central spot is a direct result of abberations that happen as light hits the very edges of the main lens. But if I use the collar edge as the “aperture” instead, especially because it’s bright aluminium, I’m sure I’ll risk producing some ring artefacts. Maybe I need to make a perfectly round black plastic ring to sit on the collar and act as the aperture.
If this works as expected I will definitely be playing around with the aperture size and seeing how it affects the total throw and image. Excited to experiment with this stuff.
Aperture! I knew there was a word! Brain too full, hard to find the right words to say what I’m thinking!
Edit: Also, I don’t think Mag Industries was quite so lazy in their design. Whether they used a parabola or not for the Incans, I don’t know for sure, but I thought for sure I read somewhere it was more spherical. As for the Mag LED reflector, they didn’t just cut out the bottom of the parabola, and leave it flat. They actually brought in the sides, so it is more conical overall, and deeper than the Incan reflectors, instead of shallower as your model would suggest.
I don’t know about the mag led reflector and I’m not saying you can’t do a bunch of stuff beyond the basic paraboloid. Two stage reflectors for controlling corona and all that.
Just that I’ve often seen people read too much into different reflector “designs” when really they’re all paraboloids of different parameters. There is nothing fundamentally different between an average reflector designed for an incandescent vs one designed for an led. That’s what the image is supposed to show.
They can however look very different to our eye and the incandescent can often be described as bowl shaped or more spherical because of this.
It’s not about being lazy, you want to collimate a beam of light with a reflector the parabola is really the best option. Maybe the minimag actually has a spherical reflector, but I doubt it. The only light I’ve seem with a properly spherical reflector was some 10 cent plastic cheapy with a horrendous beam
There’s some nice basic diagrams of parabolic vs spherical reflector stuff here if anyone is interested.
I have played with the Incan reflectors and I would not say they are good for this. They are not spherical and are not the best choice. I think something like the cake trays, or even a Christmas ornament, would be better. When I played with the wavien collar, it was more like a half sphere and I think that is what is needed. The Incan reflector is not a half sphere shape.
The christmas ornament was something I wanted to try from reading those threads ages ago, but the ones I have here are a very poor reflector on the inside. I started sanding one down a couple days ago, but didn’t go further once I saw what it was like inside.
The small enough ones I have seen are much more lumpy on the inside and have like a golden yellow coating which isn’t as reflective as it looks from the outside. I think there is a sealant or something that’s applied to stop the shiny coating from oxidizing.
Also glass is really annoying to grind/cut.
