Talk about future projects and donation topic

634 posts / 0 new
Last post
DavidEF
DavidEF's picture
Offline
Last seen: 1 day 11 hours ago
Joined: 06/05/2014 - 06:00
Posts: 7671
Location: Salisbury, North Carolina, USA

JasonWW wrote:
The Miller wrote:
Stupid question maybe but
If we use the exact same driver as the Q8 to power 1 led (XPLHi or SST40) would that be fed 20A and blow?

I’m no driver designer, but I don’t see how one Fet driver, the Q8, would function differently than another fet driver, blf A6. At full power it’s basically direct drive.

You do have less voltage sag with 4 18650 compared to a single 18650, but I don’t think it would make a huge difference.

I think the internal resistance of the led is what controls the current (in direct drive). When you run 2 emitters in parallel, or 3 or 4, etc… your just reducing the resistance which allows for more current to flow.

So an xpl might see 5-6 amps, 4 xpl in parallel might see 20 amps, but only if given enough battery power.

If you look at a quad parallel xpl light with single 18650 you don’t get 20 amps due to battery limitation and voltage sag. I think you get around 12 amps?

I’m sure there are many people here that can tell you exactly what would happen Miller.


Well, you also have the fact that the Q8 holds 4x cells, so more potential current flow. The LED will not regulate current. If there is over-current, the LED will blow.

The Cycle of Goodness: “No one prospers without rendering benefit to others”
- The YKK Philosophy

fixed it
Offline
Last seen: 3 days 22 hours ago
Joined: 12/08/2015 - 14:27
Posts: 395
Location: Canada

DavidEF wrote:
The LED will not regulate current. If there is over-current, the LED will blow.
It will, to some extent, depending on the Vf curves. Some LEDs (eg. XP-L) you can pretty much connect straight to a 4.2V power source. Others (eg. 219C) would likely blow indeed if there is not enough resistance elsewhere in the light. So it’s not unthinkable that the same driver could drive a single LED but it’s not a given either. As they say, the devil is in the details.
The Miller
The Miller's picture
Offline
Last seen: 1 year 7 months ago
Joined: 12/14/2015 - 12:08
Posts: 9630
Location: Charente France

JasonWW wrote:
The Miller wrote:
Stupid question maybe but
If we use the exact same driver as the Q8 to power 1 led (XPLHi or SST40) would that be fed 20A and blow?

I’m no driver designer, but I don’t see how one Fet driver, the Q8, would function differently than another fet driver, blf A6. At full power it’s basically direct drive.

You do have less voltage sag with 4 18650 compared to a single 18650, but I don’t think it would make a huge difference.

I think the internal resistance of the led is what controls the current (in direct drive). When you run 2 emitters in parallel, or 3 or 4, etc… your just reducing the resistance which allows for more current to flow.

So an xpl might see 5-6 amps, 4 xpl in parallel might see 20 amps, but only if given enough battery power.

If you look at a quad parallel xpl light with single 18650 you don’t get 20 amps due to battery limitation and voltage sag. I think you get around 12 amps?

I’m sure there are many people here that can tell you exactly what would happen Miller.

Yes this is my thinking
But probably better to gofor a new driver with more regulation anyways (and with single LED and driver with big pcb ample space for chips to do so.)
Thanks

patmurris
patmurris's picture
Offline
Last seen: 2 hours 53 min ago
Joined: 12/22/2014 - 15:54
Posts: 1400
Location: Nice, France

While we are throwing ideas…

One thing i’ve been thinking about is a legoable set of components that would allow for a great number of lights – the generalized idea of the Q8/ST compatibility.

A lot many lights are made of 3 parts: head, battery tube and tail cap. The head itself can be split in 3 parts: optics, led and driver. Imagine all those can be easily swapped. It’s just a matter of using the same ‘interface’ – threads and electrical connections.

If a set of interfaces is agreed upon and published, you can then focus on one part or the other to create new components and implicitly a new set of lights. For instance, instead of building a new light from scratch, you just need to create a new head, or maybe just a new driver. One could build some fancy tailcaps, or a usb charger ‘stage’ that would screw in between the battery tube and the driver, other optics, other firmware based drivers… or a purely decorative ring that could fit in different places… Different materials, different finishes, different functionalities… all legoables.

Of course there would need different interface ‘scales’ to fit the different batteries – AAA, AA, 18xxx, 20xxx, 26xxx… 3×18xxx, 4×18xxx… But still some ‘adapter’ ring could allow to mix them… sometime. Battery carriers could be designed to fit smaller batteries in larger tubes…

Not sure this is a good idea or an unwanted constraint, but it seems a good idea to keep things compatible as much as possible to save on future project design and production.

PS: are the GT and Q8 parts compatible?

The Miller
The Miller's picture
Offline
Last seen: 1 year 7 months ago
Joined: 12/14/2015 - 12:08
Posts: 9630
Location: Charente France

That seems very smart
Hard, but smart
Look at the A6 and its Astrolux counterpart, they dont even Lego.
But indeed a default threading size would be the way to go.

Fin17
Fin17's picture
Offline
Last seen: 33 min 7 sec ago
Joined: 02/05/2017 - 07:31
Posts: 2580

Let me just add one little thing off topic.

In my opinion, a good logo is needed for BLF-projects.


What do you think?

 

 

patmurris
patmurris's picture
Offline
Last seen: 2 hours 53 min ago
Joined: 12/22/2014 - 15:54
Posts: 1400
Location: Nice, France
Fin17 wrote:

Let me just add one little thing off topic.


In my opinion, a good logo is needed for BLF-projects.



What do you think?

Good idea! That would probably mandate a dedicated thread…

A quick sketch:

Agro
Agro's picture
Online
Last seen: 2 min 38 sec ago
Joined: 05/14/2017 - 11:16
Posts: 4754
Location: Ślōnsk

patmurris wrote:
While we are throwing ideas…

One thing i’ve been thinking about is a legoable set of components that would allow for a great number of lights – the generalized idea of the Q8/ST compatibility.

A lot many lights are made of 3 parts: head, battery tube and tail cap. The head itself can be split in 3 parts: optics, led and driver. Imagine all those can be easily swapped. It’s just a matter of using the same ‘interface’ – threads and electrical connections.

If a set of interfaces is agreed upon and published, you can then focus on one part or the other to create new components and implicitly a new set of lights. For instance, instead of building a new light from scratch, you just need to create a new head, or maybe just a new driver. One could build some fancy tailcaps, or a usb charger ‘stage’ that would screw in between the battery tube and the driver, other optics, other firmware based drivers… or a purely decorative ring that could fit in different places… Different materials, different finishes, different functionalities… all legoables.

Of course there would need different interface ‘scales’ to fit the different batteries – AAA, AA, 18xxx, 20xxx, 26xxx… 3×18xxx, 4×18xxx… But still some ‘adapter’ ring could allow to mix them… sometime. Battery carriers could be designed to fit smaller batteries in larger tubes…

Not sure this is a good idea or an unwanted constraint, but it seems a good idea to keep things compatible as much as possible to save on future project design and production.

PS: are the GT and Q8 parts compatible?

I totally agree that there should be standards. But at the same time I think they should not be binding, f.e. when a project targets size over all, standard compliance would not be welcome.
And, obviously, we should have different standards for the same part, depending on use. There’s place for lights that have threading smaller than Q8.

The question is: what can be standardized?

  • battery tubes, battery-tube o-rings
  • battery carriers
  • lenses, bezels, o-rings (Q8-based thrower could be compatible with Q8 f.e.)
  • switches
  • drivers
  • LED PCBs

Something else?

I would also consider standardizing a set of battery formats. I count eleven fairly popular battery sizes used in lights. And several others less popular. Cutting that to just a few would reduce the cost of entry for those of us who don’t have huge battery collections yet.

As to drivers, I’ll start with saying that I don’t believe that 17 mm is a good standard for 18xxx batteries, for all regular lights 20 mm will fit. Also, max driver height could be standardized.

The Miller
The Miller's picture
Offline
Last seen: 1 year 7 months ago
Joined: 12/14/2015 - 12:08
Posts: 9630
Location: Charente France

Single 18650 width tube, threading like Convoy S series
Single 26650 width tube (also fits 21700/20700) Convoy L series size threading
4*18650 tube, Q8 threading

This seems logical right?
We probably can get the sizes or even CADs from Simon.

JasonWW
JasonWW's picture
Offline
Last seen: 26 min 56 sec ago
Joined: 10/22/2016 - 11:41
Posts: 11784
Location: Houston Texas

The Miller wrote:
JasonWW wrote:
The Miller wrote:
Stupid question maybe but
If we use the exact same driver as the Q8 to power 1 led (XPLHi or SST40) would that be fed 20A and blow?

I’m no driver designer, but I don’t see how one Fet driver, the Q8, would function differently than another fet driver, blf A6. At full power it’s basically direct drive.

You do have less voltage sag with 4 18650 compared to a single 18650, but I don’t think it would make a huge difference.

I think the internal resistance of the led is what controls the current (in direct drive). When you run 2 emitters in parallel, or 3 or 4, etc… your just reducing the resistance which allows for more current to flow.

So an xpl might see 5-6 amps, 4 xpl in parallel might see 20 amps, but only if given enough battery power.

If you look at a quad parallel xpl light with single 18650 you don’t get 20 amps due to battery limitation and voltage sag. I think you get around 12 amps?

I’m sure there are many people here that can tell you exactly what would happen Miller.

Yes this is my thinking
But probably better to gofor a new driver with more regulation anyways (and with single LED and driver with big pcb ample space for chips to do so.)
Thanks


Why not just try it and see what happens. Take the Q8 prototype, remove the mcpcb and reflector.
Wire up a single xpl on a 20mm or so mcpcb, find something to hold it down with, maybe a small reflector you have. Hold it down with the Q8 glass. Then install a single weak battery and try it out. Ideally you want to measure the current as you ramp it up. If it survives, then replace the battery for a single high power one and test again. If it survives and the amps are not excessive, try 2 high power cells, etc… Maybe it will hold up to 4 high power cells with only 6 or 7 amps. Maybe not. Testing with an xpl-2 would probably make the amps go up higher. If you think the sst-40 is a good choice, get one of those and try the same test procedure. See what the actual amps are for the stock driver. Then you will know. Smile

Texas Ace Lumen Tube calibrated with Maukka lights

Click this to go to signature links.

The Miller
The Miller's picture
Offline
Last seen: 1 year 7 months ago
Joined: 12/14/2015 - 12:08
Posts: 9630
Location: Charente France

Oh if only days had more then 24 hours Wink

JasonWW
JasonWW's picture
Offline
Last seen: 26 min 56 sec ago
Joined: 10/22/2016 - 11:41
Posts: 11784
Location: Houston Texas

patmurris wrote:
While we are throwing ideas…

One thing i’ve been thinking about is a legoable set of components that would allow for a great number of lights – the generalized idea of the Q8/ST compatibility.

A lot many lights are made of 3 parts: head, battery tube and tail cap. The head itself can be split in 3 parts: optics, led and driver. Imagine all those can be easily swapped. It’s just a matter of using the same ‘interface’ – threads and electrical connections.

If a set of interfaces is agreed upon and published, you can then focus on one part or the other to create new components and implicitly a new set of lights. For instance, instead of building a new light from scratch, you just need to create a new head, or maybe just a new driver. One could build some fancy tailcaps, or a usb charger ‘stage’ that would screw in between the battery tube and the driver, other optics, other firmware based drivers… or a purely decorative ring that could fit in different places… Different materials, different finishes, different functionalities… all legoables.

Of course there would need different interface ‘scales’ to fit the different batteries – AAA, AA, 18xxx, 20xxx, 26xxx… 3×18xxx, 4×18xxx… But still some ‘adapter’ ring could allow to mix them… sometime. Battery carriers could be designed to fit smaller batteries in larger tubes…

Not sure this is a good idea or an unwanted constraint, but it seems a good idea to keep things compatible as much as possible to save on future project design and production.


This concept was discussed in detail in some older threads and it never really made sense. You would have to convince a flashlight company to make all these individual parts without actually knowing which parts are going to sell well and which ones may not sell at all. It’s not practical for a company to make all these different parts and keep them in stock. The average flashlight buyer does not want to buy parts and put it together himself, that is only an enthusiast type of thing which is going to greatly limit sales. The end result will be a bulky looking and feeling flashlight due to the inefficient packaging you get by Lego’ing a bunch of universal parts.

That was the general impression in those older threads.

Texas Ace Lumen Tube calibrated with Maukka lights

Click this to go to signature links.

Agro
Agro's picture
Online
Last seen: 2 min 38 sec ago
Joined: 05/14/2017 - 11:16
Posts: 4754
Location: Ślōnsk

The Miller wrote:
Single 18650 width tube, threading like Convoy S series
Single 26650 width tube (also fits 21700/20700) Convoy L series size threading
4*18650 tube, Q8 threading

This seems logical right?
We probably can get the sizes or even CADs from Simon.


About right, though frankly I’d skip on 26650 entirely for now and wait for 21700. It’s likely that 26650 will be obsolete before we make the first light with it.

Also, it would be good to specify battery compatibility even tighter:

  • min/max length (i.e. protection)
  • max diameter (protection too, but oversized batteries happen too (Shockli 26650))
  • button top?

Q8 forces many people to buy new batteries, making it ~$57 light for them. I don’t know the reasons behind that, I assume it was a well thought-up decision, but should we require all 18650 batteries to be button top for compatibility with BLF lights? Maybe we should, because we already went that way. Maybe we should re-think and go either way. Or maybe some projects will be better with button-top-only and others with either-way-fits.
Overall, it would be good if we could make sure that users don’t have to buy new cells to use a new light unless that light has way different size than anything they own already.

Agro
Agro's picture
Online
Last seen: 2 min 38 sec ago
Joined: 05/14/2017 - 11:16
Posts: 4754
Location: Ślōnsk

JasonWW wrote:
patmurris wrote:
While we are throwing ideas…

One thing i’ve been thinking about is a legoable set of components that would allow for a great number of lights – the generalized idea of the Q8/ST compatibility.

A lot many lights are made of 3 parts: head, battery tube and tail cap. The head itself can be split in 3 parts: optics, led and driver. Imagine all those can be easily swapped. It’s just a matter of using the same ‘interface’ – threads and electrical connections.

If a set of interfaces is agreed upon and published, you can then focus on one part or the other to create new components and implicitly a new set of lights. For instance, instead of building a new light from scratch, you just need to create a new head, or maybe just a new driver. One could build some fancy tailcaps, or a usb charger ‘stage’ that would screw in between the battery tube and the driver, other optics, other firmware based drivers… or a purely decorative ring that could fit in different places… Different materials, different finishes, different functionalities… all legoables.

Of course there would need different interface ‘scales’ to fit the different batteries – AAA, AA, 18xxx, 20xxx, 26xxx… 3×18xxx, 4×18xxx… But still some ‘adapter’ ring could allow to mix them… sometime. Battery carriers could be designed to fit smaller batteries in larger tubes…

Not sure this is a good idea or an unwanted constraint, but it seems a good idea to keep things compatible as much as possible to save on future project design and production.


This concept was discussed in detail in some older threads and it never really made sense. You would have to convince a flashlight company to make all these individual parts without actually knowing which parts are going to sell well and which ones may not sell at all. It’s not practical for a company to make all these different parts and keep them in stock. The average flashlight buyer does not want to buy parts and put it together himself, that is only an enthusiast type of thing which is going to greatly limit sales. The end result will be a bulky looking and feeling flashlight due to the inefficient packaging you get by Lego’ing a bunch of universal parts.

That was the general impression in those older threads.


It’s different when we are designing the lights. Keeping standards (but not procrastinating to keep compatibility) will make the flashlights every bit as good for regular users, while making them better for us. And it also reduces design time.
patmurris
patmurris's picture
Offline
Last seen: 2 hours 53 min ago
Joined: 12/22/2014 - 15:54
Posts: 1400
Location: Nice, France
Agro wrote:
Q8 forces many people to buy new batteries…

The SRK design has always mandated button top cells. And to get the most output from the Q8 you’d want unprotected high drain… but that’s just an option. Nothing different from a ‘regular’ SRK clone there.

WaylonJennings
Offline
Last seen: 1 month 2 days ago
Joined: 05/16/2017 - 21:08
Posts: 210
Location: Elk Mound, Wisconsin

ToyKeeper wrote:
I’d still like to see some significantly smaller BLF lights, like a 14500 version of the brass AA with BLF driver and high-CRI emitter, and a BLF clone of the Olight S10/S1/S-Mini. Both use small batteries and are small for their battery type.

The goal wouldn’t be to make the brightest lights of their size, but to put a lot of features in a small space, with good attention to detail.

I would love to see a nice brass AA with a high CRI emmitter too.

WaylonJennings
Offline
Last seen: 1 month 2 days ago
Joined: 05/16/2017 - 21:08
Posts: 210
Location: Elk Mound, Wisconsin

Agro wrote:
Today I got a thought about a moderate light that I’d love to have.
Not big, not small.
Not a flooder, not a thrower.
Not high output, but not a weakling either.
Not the runtime king, but nevertheless very good.
In my mind I call it a Jax of all trades because of one inspiration.

Details:

  • 3*18650
  • head just slightly larger than body
  • XHP35 HI, driven hard
  • efficient driver
  • compact package
  • side switch
  • zoomie!

Could be significantly cheaper with linear driver and SST-40, but efficiency would suffer a lot.

Would this end up being like Supbeam K40M in a smaller package? I love the super-wide beam that throws pretty well of that light. It is one of my favorite “general use” lights for sure. If BLF could build a similar wide beam/nice throw light in a smaller package I would buy two for sure.

The Miller
The Miller's picture
Offline
Last seen: 1 year 7 months ago
Joined: 12/14/2015 - 12:08
Posts: 9630
Location: Charente France

Yeah a zoomie would be cool to do for sure.
Idk how the XHP35HI fares under a lens.

Too bad there are no round DIE LEDs.

Agro
Agro's picture
Online
Last seen: 2 min 38 sec ago
Joined: 05/14/2017 - 11:16
Posts: 4754
Location: Ślōnsk

The Miller wrote:
Yeah a zoomie would be cool to do for sure.
Idk how the XHP35HI fares under a lens.

Too bad there are no round DIE LEDs.


Luminus makes some round ones, but they are neither cheap nor efficient.
There are also COBs.
Jerommel
Jerommel's picture
Offline
Last seen: 5 months 1 week ago
Joined: 01/04/2014 - 13:18
Posts: 5721
Location: the Hague, Netherlands

There’s the Nichia 319A which has a hexagonal die.

But wasn’t there talk of a plan for a dual lens zoomie?
That could be a fun challenge.

2Q19

Enderman
Enderman's picture
Offline
Last seen: 3 hours 45 min ago
Joined: 11/03/2016 - 22:42
Posts: 3965
Location: Vancouver, Canada

Agro wrote:

Luminus makes some round ones, but they are neither cheap nor efficient.
There are also COBs.

The circular CBT90 and CBT140 are not COBs, it is a single LED, and they are about as efficient as cree LEDs.
The only difference is that instead of being 6 or 12v they are 3v and much higher current.
The Miller
The Miller's picture
Offline
Last seen: 1 year 7 months ago
Joined: 12/14/2015 - 12:08
Posts: 9630
Location: Charente France

Yes Jerommel, when a zoomie is done as BLF special I have something to do with it can only be a tad strange, so double lens or some sort of wavien thing.
I have seen your Brinyte and have a Jaxman Z1, no point in making a regular, those are available in good quality already Big Smile

DBSAR
DBSAR's picture
Offline
Last seen: 58 min 50 sec ago
Joined: 02/11/2013 - 23:28
Posts: 6013
Location: Ontario, Canada

I’m not sure if the BLF Lantern project is still on your table as a future project, but many would still like to see that be added hopefully in the future.

That Canadian flashlight guy & Lantern Guru -Den / DBSARlight

Agro
Agro's picture
Online
Last seen: 2 min 38 sec ago
Joined: 05/14/2017 - 11:16
Posts: 4754
Location: Ślōnsk

SBT-70 is not a COB.
According to Luminus CBT-90 is a chip-on-board LED

CBT-140 is a COB too.

As to efficiency, if I read the datasheet correctly, the top bin of CBT-90 does 2100 lm at the rated current. That’s 9A. At this current Vf is 3.1 V. So 75 lm/W. In Texas Ace tests of XHP35 E2 bin we see the same output with c.a. 100 lm/W. There’s HI E4 available.

ADDED: that said, I’d love to see a big thrower based on CBT-90.
ADDED: though probably compact 3*18650 package is (by far) not enough to make it thermally stable at high output.

The Miller
The Miller's picture
Offline
Last seen: 1 year 7 months ago
Joined: 12/14/2015 - 12:08
Posts: 9630
Location: Charente France

DBSAR wrote:
I’m not sure if the BLF Lantern project is still on your table as a future project, but many would still like to see that be added hopefully in the future.

Yes I saw your post.
I had hoped Barry and you would have kicked it off.
Charging, interesting head
Could it be done in such a way the head Lego’s with the Q8?
Maybe some sort of jumper so we could do this one with the long tube 2S for long runtimes. Adjust jumper and use it with the normal Q8 or short ST tube?
Though a really good diffuser (with inverted shining cone surrounded by frosted white) that screws on the Q8 instead of bezel would make it a lantern but missing the smart design features you made.)

Just thinking out loud here Wink
.
.
.

Those luminous LEDs, only the 140 seems round. 21A is a good match with the Q8 driver though, are they as expensive as the SST were/are?

Enderman
Enderman's picture
Offline
Last seen: 3 hours 45 min ago
Joined: 11/03/2016 - 22:42
Posts: 3965
Location: Vancouver, Canada

Agro wrote:
SBT-70 is not a COB.
According to Luminus CBT-90 is a chip-on-board LED

CBT-140 is a COB too.

As to efficiency, if I read the datasheet correctly, the top bin of CBT-90 does 2100 lm at the rated current. That’s 9A. At this current Vf is 3.1 V. So 75 lm/W. In Texas Ace tests of XHP35 E2 bin we see the same output with c.a. 100 lm/W. There’s HI E4 available.

ADDED: that said, I’d love to see a big thrower based on CBT-90.
ADDED: though probably compact 3*18650 package is (by far) not enough to make it thermally stable at high output.


“ Chip on board package assures straightforward system assembly with the best
possible thermal performance for high power devices.”

This is referring to the fact that the LED chip comes attached to the MCPCB board.
It is not a “COB” type LED that has multiple blue LEDs underneath a phosphor layer.

This is a real COB:

This is a single die LED with a sound shape:

DBSAR
DBSAR's picture
Offline
Last seen: 58 min 50 sec ago
Joined: 02/11/2013 - 23:28
Posts: 6013
Location: Ontario, Canada
The Miller wrote:
DBSAR wrote:
I’m not sure if the BLF Lantern project is still on your table as a future project, but many would still like to see that be added hopefully in the future.
Yes I saw your post. I had hoped Barry and you would have kicked it off. Charging, interesting head Could it be done in such a way the head Lego’s with the Q8? Maybe some sort of jumper so we could do this one with the long tube 2S for long runtimes. Adjust jumper and use it with the normal Q8 or short ST tube? Though a really good diffuser (with inverted shining cone surrounded by frosted white) that screws on the Q8 instead of bezel would make it a lantern but missing the smart design features you made.) Just thinking out loud here Wink

I never hear much from Barry. (i don;t remember who that was. :P) im not sure about the high amp output of the Q8 driver would work with a continuous-running lantern head. The main concept and ability of the lantern design is to have a warmer white output that creates less eye glare, a warmer tone & higher CRI for lantern use, added with substantial long run times even on maximum mode, compactness for storage, ability to run on 1, 2 ,3 or 4 cells, have built in regulated parallel charging (that can charge from a solar panel in off grid locations) etc. If i had a Q8 sample to work with and modify, i could probably start from scratch on a re-worked second lantern prototype based on the Q8 body and its driver with all its modes, but at a reduced current to increase run times.) (and add charging ability)
Below is the current prototype still being battered hard and used & performing as the best LED lantern i ever used, here on a mountaintop expedition we did back in July in a mountain range on the Atlantic coast:

That Canadian flashlight guy & Lantern Guru -Den / DBSARlight

ZozzV6
ZozzV6's picture
Offline
Last seen: 3 hours 40 min ago
Joined: 03/24/2016 - 12:19
Posts: 2324
Location: Near to my soldering iron.

I like to do tests so I will be happy If I can help the future projects with some homework. Big Smile
The convoy and Q8 thread compatibility is a good idea.
Also a very small light I think would be better with 18350 than 16340 because now we have the Aspire 18350 as a decent high drain cell but I can not name much good 16340.
I can measure lumens, Intensity, temperature, current, voltage. I can solder and put together a driver so you can count on me in if need to do some experiment on future lights Cool

Agro
Agro's picture
Online
Last seen: 2 min 38 sec ago
Joined: 05/14/2017 - 11:16
Posts: 4754
Location: Ślōnsk

Enderman wrote:
Agro wrote:
SBT-70 is not a COB.
According to Luminus CBT-90 is a chip-on-board LED

CBT-140 is a COB too.

As to efficiency, if I read the datasheet correctly, the top bin of CBT-90 does 2100 lm at the rated current. That’s 9A. At this current Vf is 3.1 V. So 75 lm/W. In Texas Ace tests of XHP35 E2 bin we see the same output with c.a. 100 lm/W. There’s HI E4 available.

ADDED: that said, I’d love to see a big thrower based on CBT-90.
ADDED: though probably compact 3*18650 package is (by far) not enough to make it thermally stable at high output.


“ Chip on board package assures straightforward system assembly with the best
possible thermal performance for high power devices.”

This is referring to the fact that the LED chip comes attached to the MCPCB board.
It is not a “COB” type LED that has multiple blue LEDs underneath a phosphor layer.

This is a real COB:

This is a single die LED with a sound shape:
!{width:25%}http://cdn-images.av-iq.com/products/enlarge/CBT-140.jpg!


Thanks for the info.
The Miller
The Miller's picture
Offline
Last seen: 1 year 7 months ago
Joined: 12/14/2015 - 12:08
Posts: 9630
Location: Charente France

DBSAR, yes you are right, sorry for again talking about a simple diffuser.

So a Q8 tube compatible head
Driver with less amps then Q8
Single warm led
USB charging.

It seems a driver like the ST needs but with less 7135 chips and charging chip

Of course in your proto the total length is more then needed, since you build on a SRK.

You had an inverted reflector/cone planned for sideways light right?

Powerbank function?
Charge the lantern with a solar panel and at night cherfe phone and or speaker.
Ample space and more versatile.

Maybe cool to do ST and lantern at the same time aiming for early summer for finishing the lantern.
You have a Q8 ordered?
Not a red lighted switch but green acting as beacon.

Pages