TOPIC: A Great Deal To Learn About LED's I Have

Hi Gang,

So I am attempting to put a Medium Blue LED in my Redeemer. I built it with Quick Connects so my trouble is not in the board but in the LED. I want to separate the Royal Blues from the Blues and that way I can have a Light Blue Episode IV font, a Medium Blue Font (will probably use an Episode V graflex font), and a Deep Blue Episode 3 Font and I can adjust the FOC for each.

My LED looks like this before I rewired it:





I thought I'll just put the Royal Blues together and the Blues together. Well, I tried that by removing the bridge shown, adding another negative lead on the soldered pad that does not have a lead and I got a Medium Blue on each Channel.

I reached out to CS and found out that the way the star is designed, the colors are b / rb / b / rb so any way I wire it it will always return Royal Blue. I am totally new at this and do not pretend to understand how the star works. I assumed each pad was for each diode with the 2 positives (one on each end).

I guess I'm going to have to request (buy) a custom X4 from shopsaberparts and then figure out what I need to resistor each. (I'll do a little homework before I start asking questions).

So, has anyone been able to get FOC to work or isolate colors by rewiring a Saberforge X4 LED? I also wanted to take a pink LED apart so my wife could get pink, or red, or blue on hers but I'm thinking I'm going to run into the same problem there as well. I just wanted to get everyone's thoughts.

Thanks in advance!

All positives are linked on the X2/X4 stars, so it doesn't matter if you use one, the other, or both positive pads.

That said, I think what you have is an X2, not an X4 star. The naming's a bit confusing, because either is capable of holding 4 LEDs, but the X2 has only two selectable channels (with negatives doubled up on each channel). If this was an X4, the star would have one positive and all four negatives soldered.

I imagine the B/rB/B/rB criss-cross orientation is for better color-mixing on a standard LED module.

If you want a selectable-blue star, you'll have to custom order a LightBlue/DeepBlue Hero LED without quick-connects. It'll likely be a B/B/rB/rB X2, but this time with BB on one channel and rBrB on the other.

Have you tested the dies invidually? I recently rewired a cyan LED from SF and found it had the same setup where colors were set up diagonally with eachother:

B G
G B

I took a battery with pos/neg wires and touched the leads to the common positive pad and each neg pad. I had each LED come on by itself. Since they're diagonally situated, I couldn't bridge the pads above the star, so I just soldered wire to each negative pad and combined the blues and the greens underneath the star after inserting it into the LED module.

Kouri wrote:
Thank you, Kouri. I appreciate your insight as always. So you could tell this is most likely an X2 simply because it only has 2 negative pads soldered? On an X4, does the circuit board design look different or is there no other way to tell the difference?

I also noticed the LED they sent me is resistored. I for some reason did not think Blues needed resistors, but I have not done the math yet so I am sure it is supposed to be there. I'll have to get those same resistors when I build my own.
Khang wrote:
I would try that, but I'm completely out of batteries to test with (the only one I have is buried in my hilt and I really don't want to take it apart with all of the electrical tape, but thank you for the suggestion. I noticed TCSS is currently out of the Panasonic 18650 3400mah. Is there another good place to get 18650 3400mah batteries with leads? I basically wire my sabers with the leads intact so I don't have to solder the battery to the board.

Solo's Hold has some wired 3200mAh 18650s (it's what I used with my Graflex).

Though for testing LED dies, you can use a 2 AAA battery holder with leads like THIS one from Lowes, and use the leads to touch the pads you want to test. You can also add a resistor to one of the leads if you are working with reds and ambers. It's a tip I picked up from one of TCSS's Madcow tutorials.

Jaden Korr wrote:
That is so brilliantly simple. I love it. Thank you so much. Time to go to Lowes tonight.

It'll be in the aisle that has all the specialty screws in the pullout drawers (of all places). At my local Lowes it's like aisle 15. There's a few "hobby" drawers, and it'll be in one of them.

Wow, that's very specific! Thanks so much! Haha

Just ordered a battery from Solo's Hold. I guess there won't be a ton of difference between the 3200mah and 3400mah.

Now I'm off to go figure out R=(Vs - Vl)/I I think the only part I don't get is how to figure out what the Vled is on the a Saberforge LED. Otherwise, I'm thinking it should be Resistance = (3.7 - some number) / 3400. I hope I have that right.

Nice thing about 1000mA LEDs is that both the resistance and wattage calculations simplify to:

[Battery Voltage] - [LED Voltage]

Battery Voltage for a lithium-ion is either going to be the nominal 3.7v (brighter) or the fresh-charge 4.2v (safer).

LED Voltage will either be the rated average at 1000mA (brighter) or the absolute maximum rating (safer).

~

But for future reference, full formulas are:

Ohms = [(Battery Voltage) - (LED Voltage)] / (LED Current in Amps)

Wattage = [(Battery Voltage) - (LED Voltage)] * (LED Current in Amps)

For a 12W+ LED, you'll want to check the spec sheets for... "Cree XQ-E" LEDs, if I remember right.

HotRod wrote:
I know right. When I went to look for them I went to electrical and checked where they keep the multimeters and asked, and the worker led me to the screw aisle and was like "duh!" Apparently I'm an idiot for not thinking to look for it next to the furniture screws...
Kouri wrote:
That's what SF says they use. Though the spec sheets I've turned up have the foreword voltages at 350mA listed but not 1000mA.

Kouri wrote:
Thank you so much for this info! So I found this:




Let's assume Royal Blue

As I understand it, since R = 3.7 [Battery Voltage] - 3.1 [LED Forward Voltage] = 0.6 (assuming the 1 A current rating)

If I did Red

R = 3.7 [Battery Voltage] - 3.1 [LED Forward Voltage] - 2.2 [LED Forward Voltage] = .9

I wasn't sure if I should go with the "typical" forward voltage or "maximum" forward voltage.


I'm also trying to take into account what you said in another thread where:

That's why I was using 3.7V instead of 4.2V. Am I on the right track so far?

Jaden Korr wrote:
Ah, and that is what I have here...so my math does not work here since we're looking at 1000mA and my chart above is at 350mA?

Maybe this helps...




I'm not sure why my pictures are coming in small. Here is where I am pulling from:

http://www.cree.com/~/media/Files/Cree/LED-Components-and-Modules/XLamp/Data-and-Binning/ds-XQE.pdf

That does help! For some reason those pages didn't show up when I was looking at that PDF originally.

Jaden Korr wrote:
So if I read that graph correctly, Red is about 2.6, Amber and Blue about 3.4, and Green in the 3.5-3.6 range @ 1000mA?

Sounds about right. XQ-Es have the same min/max ratings as XP-E2s, so it makes sense for the 1A voltages to be about the same.

Just a heads up that the ~3.3v of the XQ-E PC Amber *only* works with "PC" Amber. If using standard Amber on an old XB-D 12W or XP-E2 star, you'll use the same 2.6v as the other warm-colored LEDs.

Just so my math is documented in case anyone else comes up with this problem, here is what I understand at this point:

Battery Voltage: 3.7 (not assuming 4.2 as I am taking a "Middle of the Road Approach")
LED Current: 1A

For XQ-E LED:
Red Forward Voltage @ 1A: 2.6V
Blue / Royal Blue Forward Voltage @ 1A: 3.4V
Green Forward Voltage @ 1A: 3.6V

Voltage [V] = Resistance [Ohms] * I [Current in Amps]
Voltage = Voltage Source - LED Forward Voltage

Rework V=I*R to find R
R = V/I

I in this case is = 1 since 1000mA so we are left with V

Therefore:

Edits in Red per Kouri's suggestions below:

Red / Photo Red / Red Orange : 3.7 - 2.6 = 1.1 ohms
Blue / Royal Blue / PC Amber: 3.7 - 3.4 = 0.3 ohms
Green: 3.7 - 3.6 = 0.1 ohms

Note: Standard Amber on an old XB-D or XP-E2 would be 3.7 - 2.6 = 1.1 ohms

Does my math look good? Also, Does Saberforge not always suggest resistors on Blue and Green since the ohms are so low?

Edit - thank you for verifying the math, Kouri!

My 6W+ Medium Blue didn't have any resistors. However, at least one or two members here have had an XQ-E Blue LED burn out, so it might have become such an issue that SF thought it safer to resistor them from here on out. It's new tech, so we're going through the growing pains.

Math looks fine, though I'd say to group PC Amber with the blues, since it's basically a Phosphor Coated Blue LED.

For Red, go ahead and add Photo Red and Red Orange to the list at the same voltage. I don't see SF adding RedOrange to the lineup any time soon, but they are available in XQ-E and XP-E2.

Kouri wrote:
RichellM peeled back the heat shrink on his 12W+ deep blue LED a few weeks ago and found a diode instead of a resistor (he was getting weird readings of 1000Ω on his multimeter because of it).


Interestingly enough, the DIY 12W+ LED not only has a amber/red orange option, it lists red orange as one of the die colors they use if you want to email them about a die combo not listed, but isn't available in either the ASP plug n' play LEDs or the regular ones off the main site.

I do not want to derail the topic, but I do not want to go too far without saying THANK YOU Masters Kouri and Jaden. I am truly grateful for your help in helping me (and others) to learn this.

Okay, back to the topic...
Jaden Korr wrote:
I noticed this too when buying my LEDs today. So there's a 2xR 2xA and 2xRo 2xA. I wonder if that makes a more orangy (is that a word?) color?

Neat info on that RedOrange availability.

Something of note about the LED - aside from the obvious blood-orange hue, is that, not counting Phosphor-Converted colors, RedOrange is the *brightest* of the warm-color LEDs. I'm assuming this is because PhotoRed/Red spend a good portion of energy putting out invisible Infrared light, while standard Amber suffers from previous LED tech having trouble putting out anything between orange and green.

PC Amber (at least the Cree variant) seems to tie with Green for brightest LED in the lineup, and is more yellow-hued than previous Amber (which is my guess as to why 12W+ Yellow had to be reformulated to AAAG compared to the previous 12W AAGG). In a similar manner that Photon Green blades make bright Lime Green blades with Blue light, Phosphor-Converted LEDs basically use a fancy compound to make Blue LEDs put out different color light (this has actually been used for a while to make really bright White LEDs).

There's another PC color that's exclusive to the Luxeon lineup called Lime, sitting in between Yellow and Green (a bit closer to yellow), and beating every other colored LED from any product line in terms of brightness. Luxeon actually makes a color-changing lightbulb with a RedOrange/Lime/RoyalBlue setup that I've been meaning to replicate in a saber build. It sacrifices true reds and greens, but the end result should be a brighter blade than is capable with a standard RGrB setup.

Folks wanting to stick with SF 12W+ and color changing might want to consider some custom mixes. RedOrange/PCAmber/Green/RoyalBlue should be able to put out some really nice bright colors at the expense of never producing a deep red. PhotoRed should be added soon (if it hasn't already been) to produce some deep deep reds, at the expense of brightness. Perhaps doubling up on PhotoReds to compensate might be in order. PhotoRed/PhotoRed/Green/RoyalBlue might be an interesting color mix, though yellows might suffer a bit, and the folks at VV seem to recommend standard Red for mixing purple.