fiduciary

Author Topic: 30 Amp LED Sabers & Fluorescence Enhanced Saber Blades (Part 1 - Saber)  (Read 27364 times)

0 Members and 1 Guest are viewing this topic.

Offline Photonic Bladesmith

  • No Force
  • *
  • Posts: 38
  • Um...Hello?
I wanted a pair of LED-in-hilt sabers to be used for cons, sparring demos and costuming.  Like most people on the forum it seems, I wanted the sabers to be as bright as possible so that they can look good in brightly lit environments.  After spending months reading through all of the archived technical posts it became apparent that if I wanted a pair of super bright sabers that I would have to build my own.

Inspired by Sunrider's work with high power LEDs I did a search for the highest luminosity LEDs that were commercially available for sale to experimenters.  The highest power single emitter LEDs that I found for sale were the Luminous Devices CBT-120 and CBT-140.  I purchased one each of the 30 amp rated red and blue CBT-120s from DigiKey and one of the 5,000 lumen 28 amp rated CBT-140s from Mouser.
 

CBT-120-B-C11-KM301 Luminus Devices Inc | 1214-1101-ND | DigiKey
CBT-120-R-C11-HK101 Luminus Devices Inc | CBT-120-R-C11-HK101-ND | DigiKey
CBT-140-WCS-L16-UA123 Luminus Devices | Mouser

When searching to see if there were any batteries able to power these LEDs that can fit in a saber hilt, I was surprised to find that Sony, Samsung and LG had a new line of high constant output current 18650 rechargeable Lithium batteries.   The 2,500 maH, 35 amp rated continuous output LG 18650 batteries were available for immediate shipment at a reasonable price.   A dozen of these LG 18650 batteries were ordered for this project.
 
In Stock genuine high discharge rate rechargeable cylindrical 18650 Li-Ion

The XTAR WP2 II intelligent charger that I used for conventional lithium 18650s worked perfectly to charge these high current output batteries.

After some preliminary experiments operating high powered LEDs at reduced current drive using translucent polycarbonate blade with improvised collimators and a series of hacked together current regulators, I decided that the way I wanted to drive the blades was at their maximum rated current of 30 amps at a duty cycle of 50%, at a frequency of between 20 and 30 Hz, where the blade would have a live shimmer that would break up the static plastic glow tube appearance of the blade at constant level lighting. 

I couldn't find any low dropout voltage, externally modulatable, variable current output, 30 amp current regulators that would fit in a saber hilt and be powered by a pair of 18650 batteries so I had to design and build my own.



The LED drive electronics is configured as three independently activatable, paralleled, 10 amp linear fold back current regulators.  Activated simultaneously they provided a measured 30 amps to the LED.  The linear current regulators are designed to have zero current draw when not activated and be rapidly switched on and off by low current level control inputs.  A dimmer switch disengages activation of two of the three parallel regulators when reduced light output or longer run time is desired.  Dropout voltage before loss of current regulation is about one volt.  A 555 timer is used to generate a variable frequency square wave to modulate the regulators.  The 555 timer is switched on by the LED output of the Nano Biscotte Sound Module V2 going active low.   

A separate power supply consisting of four Eneloop rechargeable AAA batteries had to be used to independently power the Nano Biscotte sound card.  The output from the lithium 18650 batteries being operated at 30 amps with constant square wave modulation turned out to be so electrically noisy that even with a regulator and filtering, the Nano Biscotte was constantly having randomly triggered sound and on/off events.

 

This saber project was very much a learning experience experiment in operating and powering very high current LEDs.  The mechanical design of the saber was dominated by: heat management for the LED and current regulator, fast tool-free battery changing (10 minute operating time per charge at peak power), ease of disassembly and reassembly to replace burned out parts during circuit debugging, and  ease of machining and re-machining the various mechanical parts.  As such, this is a zero Greeblie saber with esthetics being totally ignored in favor of engineering function.  The 4" diameter aluminum heat sink for the LED at the front of the saber though does double duty as an interesting looking and functional hand guard.

The electronics were mounted on a sled that was constructed from a section of 1/8" wall aluminum tubing with a 1 - 1/4" OD that matched the 1 - 1/4" ID of the hilt tubing.  Areas of the inner surface of the sled were milled flat or inset to facilitate good thermal contact with the high current transistors and resistors.  Arctic Silver thermal conductive epoxy was used to bond the heat generating components to the aluminum sled and LED heat sink.  There is a threaded centrally located hole in the sled that the Covertec knob with an embedded 1/4-20 screw engages through a hole drilled in the side of the hilt to pull the sled into tight thermal contact with the inner surface of the hilt main tube.

The battery holder was designed to operate at high current and to allow rapid tool-free battery changes.  Conventional conical steel spring battery contacts have too high a resistance (10 - 20 milliohms) to operate at 30 amps without overheating and collapsing the steel springs.  The battery contacts are 3/8" diameter solid copper rod, directly soldered to the 14 gauge regulator and LED power wires.  The front copper contact is set up to slide linearly and is spring loaded to tension the battery stack.  The rear contact has a retention screw that rotates into a detent in the Kevlar-epoxy composite battery tube. 




The rear cap containing the speaker and the AAA batteries is slid onto the rear of the hilt tube and is held in place by a knurl headed 1/4-20 threaded thumb screw.  Loosening the knurl headed retention screw and removing the rear cap allows tool-free access to the 18650 battery tube, the Nano Biscotte microSD card slot, the mute switch and the modulation rate setting potentiometer.  Battery swap time is about 20 seconds.  Since operating time at full brightness is about 10 minutes, the ability to rapidly swap batteries without tools was an important part of the design.




The LED light output is collimated by solid aluminum reflectors pulled from Nitecore MT2A flashlights.  As advised by Sunrider in his light focusing solutions thread, the rear surface of the reflector was cut back until the reflector, when placed in contact with the front window of the LED, showed an even reflection of the emitter of the LED over its entire area when viewed from the front.



How bright is the output of a 30 amp driven LED with a properly collimated reflector? 



 
LightSaber 30Amp BurningHoleTrashBag - YouTube

Here's a picture and a video of the blue LED saber illuminating a trash bag from the output of the blade holder and burning a blade diameter hole through the plastic.  Definitely not something you want to stare into!



Indoor test - Brightly lit room.



All of the artificial room lighting is turned on. The front door and a 4 foot by 6 foot window to the left of the door are opened to let in the outside daylight.  The room is brightly lit despite the appearance from the photo from having  the camera f-stop and shutter set such that the saber blades did not over expose. 

The sabers from right to left:

30 Amp CBT-120 Blue LED saber, Translucent polycarbonate blade

Master Replicas Anakin

30 Amp CBT-140 White LED saber, Translucent polycarbonate blade

White LED string driven at 6 amps mounted within a spinning light baffled blade.  ( "Persistence of vision" TFU style Black Blade Saber)

The blue LED, despite having the highest radiometric (absolute optical power) output of the colored LEDs in the CBT-120 series appears relatively dim at 600 lumens compared with the 5,000 lumen CBT 140 white LED driven at the same current.  There is a reason however, for selecting this Blue LED for construction into a high current saber despite its relatively low lumen output.  The blue saber's high absolute radiometric power at short wavelengths is one of the key elements for an idea of how to build a daylight usable lightsaber blade.

[  Continued on - 30 Amp LED Sabers &  Fluorescence Enhanced Saber Blades (Part 2 - Blade)  ]

« Last Edit: June 24, 2017, 07:07:50 PM by Photonic Bladesmith »

Offline EXAR KUN

  • SITH/EMPIRE Master MODERATOR
  • Master Force User
  • **
  • Posts: 3117
  • Formerly known as SpaceWindu
    • My YouTube channel: youtube.com/spacewindu
This is crazy!

I like the look of the giant heatsink... looks sort of like a JQ saber I like a lot.

I'll bet people will go crazy over that black blade! That looks crazy. The video is insane. People that like the black blade might have a heart attack when they see...



Offline BEN KENOBI

  • Crazy Old Wizard
  • FX-SABERS VIP II
  • Master Force User
  • *
  • Posts: 8490
  • The saber scientist
 :o :o

  wow I must say very nice in the day light.

Offline snilam

  • Experienced Force User
  • ****
  • Posts: 375
I live in texas and if your mid day sun is anything like ours, which im sure it is you being in Albuquerque. Thats is really impressive. Im sure its even better in person. Also nice post with lots of info.

Offline LUMINARA UNDULI

  • Master of the purple blade
  • Administrator
  • Master Force User
  • *
  • Posts: 10749
  • Formerly Known As Allaerra
    • Vader's Vault
Interesting.  Keep in mind with today's LEDs though, those particular LEDs may not be advantageous over other options because of the heat and what you need to do to manage it and the robust choices we now have available.

For example, a green CREE XPE-2 running at 1A gives around 200 lumens.  Obviously these can be overdriven so can provide even more.  So, if you compare the TriCREE XPE-2 green, you are getting 600 lumens at 3A running those in parallel.  If you pumped it up to 4A you'd get around the same 800 you get with the CBT-120 at a fraction of the cost.  In addition you also have a lens solution that optimizes the light in the blade.  So it would be of interest to see a side by side comparison of LEDs.

With the white, you are looking at needing 21A to garner between 3-4K lumens.  With the new CREE XPL you are getting 1226 at 3A so in a Tri Config you get the same lumen count at a fraction of the power and cost that you get with the CBT-140 at 21A. 

However, I look forward to watching your research with these, always very cool for someone to push the tech forward.  I'll be keeping an eye.
« Last Edit: March 20, 2015, 10:16:28 AM by LUMINARA UNDULI »



Website: Vader's vault
Facebook: VVFB

Offline LUMINARA UNDULI

  • Master of the purple blade
  • Administrator
  • Master Force User
  • *
  • Posts: 10749
  • Formerly Known As Allaerra
    • Vader's Vault
Your black core blade idea is very very cool though 8).  Not practical for how most of us use our sabers but very awesome idea.  I'd love to see a video of it in motion!

I have to say you are doing some spectacular work :D. 



Website: Vader's vault
Facebook: VVFB

Offline Photonic Bladesmith

  • No Force
  • *
  • Posts: 38
  • Um...Hello?
Interesting.  Keep in mind with today's LEDs though, those particular LEDs may not be advantageous over other options because of the heat and what you need to do to manage it and the robust choices we now have available.

For example, a green CREE XPE-2 running at 1A gives around 200 lumens.  Obviously these can be overdriven so can provide even more.  So, if you compare the TriCREE XPE-2 green, you are getting 600 lumens at 3A running those in parallel.  If you pumped it up to 4A you'd get around the same 800 you get with the CBT-120 at a fraction of the cost.  In addition you also have a lens solution that optimizes the light in the blade.  So it would be of interest to see a side by side comparison of LEDs.

With the white, you are looking at needing 21A to garner between 3-4K lumens.  With the new CREE XPL you are getting 1226 at 3A so in a Tri Config you get the same lumen count at a fraction of the power and cost that you get with the CBT-140 at 21A. 

However, I look forward to watching your research with these, always very cool for someone to push the tech forward.  I'll be keeping an eye.

Thanks for the critique.  I take very seriously the advice from individuals who have had many years of hands on saber building experience and many hundreds of sabers constructed!  Especially since my total build count is three!

I checked the specs on the Luminous Devices CBT-120 530 nm green LED though I didn't buy one.

4.2 Amps   800 Lumen  (very similar to the CREE green LED)
18 Amps   2,100 Lumens
30 Amps   3,100 Lumens (pulsed at 50% duty cycle)

The efficiency clearly falls off at high current levels.  Focusable arrays do look like they have a fundamental efficiency advantage over single element high power emitters.

I agree that the electrical to optical output efficiency of the blue and white Luminous devices LEDs that I used was disappointingly low compared to that of state of the art CREE LEDs at the same wavelengths.  For my experimental build, though, I was looking for maximum focusable LED output power via any means possible, and the Luminous Devices LEDs as far as I know were and possibly still are the most powerful focusable LEDs light sources available even though the CREEs totally out class them so far as energy conversion efficiency.
 
Just glancing over the LED reference you gave me, It would be fantastic if CREE would make CBT-120 sized single element emitters.  Their royal blue XPE2 LEDs at the same blue wavelength as the Luminous Devices CBT-120 has four times the watts/amp output as the Luminous Devices CBT-120!  Too bad that the official peak power output for the CREE blue XPE2 LED is limited to 1.7 watts at 1 amp.  Do you know how much these CREE XP-E2 royal blue LED's can be over driven?

__________

I took up your suggestion and looked up the green CREE XPE-2 LEDs that you mentioned and ran the numbers on it vs. the blue CBT-120.  Unfortunately the numbers came out disappointing so far as using it for illuminating a "sunlight challenging" fluorescent green blade.

The brightness of the fluorescent blade depends primarily on the photon flux at wavelengths within the fluorescein dye's absorption band which to first order is proportional to the radiometric power NOT the brightness in lumens.  The energy per photon of blue light is greater than green light by only 10-20% between the wavelengths of common green and blue LEDs so I am neglecting this difference for this calculation.

When I checked the reference tables, there was a huge difference in the lumens per watt between the green and blue wavelengths that the LEDs we are comparing operate at.

464 nm Blue  -- 45 lumens/watt

520 nm Green -- 484 lumens/watt


600 Lumens of 462 nm (CBT-120 blue) = 13 watts of radiometric power
Blue light at 462 nm is right in the center of the fluorescein dye absorption band.

800 Lumens at 520 nm (CREE green) = 1.7 watts radiometric power 
Green light at 520 nm is right at the edge of the fluorescein absorption band where the ability of the incoming photons to excite the dye to photon emission is falling to zero.

Predicted fluorescent blade brightness

Blue CBT-120 LED:
13 watts of radiometric power x 484/ lumens/watt  x 0.92 (fluorescent conversion efficiency) = (approx) 5,800 lumens of light centered on 520 nm (green).

Assuming that the TriCREE XPE-2 green LED can excite fluorescence in the fluorescein dye with equal quantum efficiency with the blue CBT-120:
1.7 watts of radiometric power x 484/ lumens/watt  x 0.92 (fluorescent conversion efficiency) = (approx) 757 lumens of light centered on 520 nm (green)

_____________

Lumens per watt conversion reference table:

Light Measurement Geometries - Chapter 7 - Light Measurement Tutorial | ILT


Offline BEN KENOBI

  • Crazy Old Wizard
  • FX-SABERS VIP II
  • Master Force User
  • *
  • Posts: 8490
  • The saber scientist
 :)) :)) :))

   I have to say I have not seen a light saber yet except this one that could burn a hole in anything lol.

Offline Spiritchaser

  • Force User
  • ***
  • Posts: 107
This is every kind of awesome!


I'm pretty sure that the new (really bright) rebel limes work more or less this way: blue led and fluorescent... Phosphor I think for that led

I'm wondering how efficient other colors might be?

It's probably only a matter of time before green single die LEDs are 400 lumens at one amp

Offline LUMINARA UNDULI

  • Master of the purple blade
  • Administrator
  • Master Force User
  • *
  • Posts: 10749
  • Formerly Known As Allaerra
    • Vader's Vault
Going to have to read and reread this as my head is exploding :D.  Great stuff!

One thing I would like to clarify with you though is that with the CREEs we use 3 of them on one star so make sure you are taking that into account in your calculations.  Currently the green I am using is this one: XPEBGR-L1-0000-00F01




Website: Vader's vault
Facebook: VVFB

Offline darth hondo

  • Master Force User
  • *****
  • Posts: 577
  • there can only be two
Re: 30 Amp LED Sabers & Fluorescence Enhanced Saber Blades (Part 1 - Saber)
« Reply #10 on: March 20, 2015, 08:34:24 PM »
Using the blue LED to "pump" the dye to generate a green blade is a very good idea. Looks like it is working quite well.

Instead of comparing the efficiency of the pumped green light from a Cree XPE2 green, it really should be a more direct comparison for consider the (expected) quantum efficiency of a tri-Cree XPE2 BBB or rBrBrB on the florescence of the dye relative to the CBT blue LED producing about 600 lumens at 30A.

The lumens rating is 45 lumens for a XPE2 blue driven at 350mA. I don't know of the lumens rating for XPE2 blue LED at 1A drive current. Anyone?
« Last Edit: March 21, 2015, 06:08:24 AM by darth hondo »

Offline Photonic Bladesmith

  • No Force
  • *
  • Posts: 38
  • Um...Hello?
Re: 30 Amp LED Sabers & Fluorescence Enhanced Saber Blades (Part 1 - Saber)
« Reply #11 on: March 20, 2015, 08:57:59 PM »
Your black core blade idea is very very cool though 8).  Not practical for how most of us use our sabers but very awesome idea.  I'd love to see a video of it in motion!

I have to say you are doing some spectacular work :D.

I uploaded a video of the spinning LED string Black Lightsaber but it has the same type of video time aliasing effects as when attempting to video a rapidly strobed saber blade.
Specifically, the blade looks like it is spinning in slow motion.

SpinningLEDChainBlade - YouTube

There is a collection of photos of the spinning LED string Black Blade in my public photobucket page.

photonicbladesmith's Library | Photobucket


Offline Onli-Won Kanomi

  • Master Force User
  • *****
  • Posts: 4689
  • Jedi Knight Of Faith
Re: 30 Amp LED Sabers & Fluorescence Enhanced Saber Blades (Part 1 - Saber)
« Reply #12 on: March 21, 2015, 04:31:52 AM »
Whoa someone actually took my [I thought] half-baked idea for the black blade and made it actually WORK???

OK now I'm going to have to officially eat a big helping of Crow because I honestly believed it would require far too much calculation, experimentation, trial and error for anyone to ever turn the spinning tubes black blade approach into practical functionality so BRAVO MY HAT IS OFF TO YOU SIR!

RESPECT!
To DREAM the IMPOSSIBLE DREAM. To FIGHT the unbeatable foe. To BEAR with unbearable sorrow. To RUN where the brave dare not go. To RIGHT the unrightable wrong. To LOVE, PURE AND CHASTE, FROM AFAR [-sigh-]. To TRY, when your arms are too weary; to REACH the unreachable Star!... This is my Quest; To follow that Star, no matter how hopeless, no matter how far...

Offline mikestar007

  • Force User
  • ***
  • Posts: 231
Re: 30 Amp LED Sabers & Fluorescence Enhanced Saber Blades (Part 1 - Saber)
« Reply #13 on: March 21, 2015, 10:27:33 AM »
Using the blue LED to "pump" the dye to generate a green blade is a very good idea. Looks like it is working quite well.

Instead of comparing the efficiency of the pumped green light from a Cree XPE2 green, it really should be a more direct comparison for consider the (expected) quantum efficiency of a tri-Cree XPE2 BBB or rBrBrB on the florescence of the dye relative to the CBT blue LED producing about 600 lumens at 30A.

The lumens rating is 45 lumens for a XPE2 blue driven at 350mA. I don't know of the lumens rating for XPE2 blue LED at 1A drive current. Anyone?

On the Cree XP-E2 data sheet there is a "Relative Radiant Flux (%)" graph.
At Forward current (mA) of 350 all colors are at 100%.  Royal blue at 1000 mA gets very close to 250% (probably 240%)
and Blue is just a hair over 200%(maybe 210%).

So I assumed for Blue one would multiple 45.7*2=91.4. and that would be a rough estimation of the lumen output. (Please correct me if I'm wrong). And for Royal blue in mW flux I don't know how that translates to Lumens.

I know these can be over driven. But it seems like we could only get a little past 300 lumens overdriving a BBB Cree XP-E2 :( . But other colors we could defiantly hit that 600 lumen mark! I have a RoRoRo thats 802.5 lumens not even being over driven!


Offline Photonic Bladesmith

  • No Force
  • *
  • Posts: 38
  • Um...Hello?
Re: 30 Amp LED Sabers & Fluorescence Enhanced Saber Blades (Part 1 - Saber)
« Reply #14 on: March 21, 2015, 10:47:56 AM »
Whoa someone actually took my [I thought] half-baked idea for the black blade and made it actually WORK???

OK now I'm going to have to officially eat a big helping of Crow because I honestly believed it would require far too much calculation, experimentation, trial and error for anyone to ever turn the spinning tubes black blade approach into practical functionality so BRAVO MY HAT IS OFF TO YOU SIR!

RESPECT!

I looked up your post on using optical interference effects to create a Black Blade.  I can't take credit for implementing your idea!  It is way beyond my level of maker skills!

The way that I created the illusion of a Black Blade was way more simple and brute force.

In summary:



If you want to construct a saber blade that provides the visual appearance of a dark central core with a luminous halo around it, visible simultaneously from all viewing angles, another hardware verified way (Turgon's cross polarized film method was the first) to build such a blade is by embedding an LED string and diffuser into a channel that goes along the sides and around the tip of the blade, with the LED string that traces the outline of the halo recessed deep enough in the channel that it cannot be seen unless it is directly face on to the viewer.



Spin this blade around its central axis with an electric motor, and persistence of vision combined with motion smearing creates the illusion of a bright luminous halo around a dark core.  The scattered light from the walls of the channel in which the LED string is embedded creates a simultaneous secondary less bright luminous outer halo.

The electric motor is powered up to spin the blade at just below the human vision flicker fusion threshold frequency in order to produce a live shimmering luminous halo effect. The LED string is driven at 6 amps through a slip ring and brush assembly.  The LEDs in the string are oriented sideways with their lenses facing out the channel towards the viewer.  The high drive current and the LED orientation combine to make the blade halo bright enough to provide a dramatic visual effect even in a sun lit room.

I built this a year and a half ago and it never fails to get a "What the H*** is that!?" response when I activate the blade in front of people who haven't seen it before.  Had lots of fun with it at the local Comic-cons!

Considering that Turgon's polarized film black blade and this spinning LED string blade have demonstrated two distinctly different ways to build a simultaneously-viewable-from-all-angles TFU style black blade, it is likely that there are other approaches out there that will also work.  I'm looking forward to seeing what other experimenters on this forum come up with and build. 



« Last Edit: August 20, 2015, 09:27:37 PM by Photonic Bladesmith »

 

retrousse