MR string minus resistors

**Jedi-Diah** · 10-21-2008, 03:05 PM

Yeah, you're right. I didn't mean it like a contest of who can get the brightest blade possible. I was just thinking the brighter the blade, the more it would look like a "real" lightsaber (as if it were a blade of pure energy), like in the movies...

But that's great that your 10 in parallel are brighter than their 18 in series. I wouldn't have a clue whether to put them in parallel or series. I am really dense when it comes to electronics. I haven't been taught well and my learning is slow. But once I "get it", I will know it. I'm glad there are people like you (and others) who know what they're doing to help guide people like me along.

One question: If I had a CF or Ultra Sound board that can manage different current output, what current setting is optimal for the led strip blade? I guess that might depend on how the led strip is wired and made.

Makoto did post on his blog how they are wired and it looks like they are "piggyback" on top of each other. I suppose that would be parallel with all the negatives together and all the positives together?

Thanks!

Originally Posted by erv

yep, the lux is really nice for that, using the other (and brighter) LED I tried, 10 are brighter than 18, for the same current used.
Me having the 10 in parallel, the hyperblade working in pairs of 2 LEDs in serie, but with twice the voltage.

I don't care about absolute brightness, just like I don't care who's pi$$ing the furthest in the snow.
My goal is to get good brightness and even blade, brighter than MR and probably luxeon, but with reasonable autonomy and battery setup. Just like good electronic systems should be designed.

**erv** · 10-21-2008, 10:12 PM

that's 100% correct, the makoto design if full parallel, means that the used current is divided equally between each led, or almost (minor differences, if the forward voltage of each LED isn't totally contant, drift over temperature etc).
Like eandori mentionned elsewhere in another thread, it's not the best configuration, but this shouldn't be taken as a generic remark.
If the forward voltage of the LED is like 3V, like the new green one I've used, then you can reasonnably put 2 LEDs in serie, then wire the pairs in //. This will divide the used current by 2 but requires 6 volt to work at an equivalent brightness.
However, with a 7.4V battery setup, 6V are low enough so that you can get autonomy and do regulation. It means that 7.4V will be "adapted" to match 6V, but even with the batteries going down the regulator will maintain the 6V on the output until the battery gets close to, let's say 6.5V.
Knowing also that the batteries will actually start around 8.4V and will remain at 7.4V for a long time (if you choose appropriate capacity versus the used current).

Just like the current CF, if I release a ledstrip version of the board, current, or at least output voltage to the strip will be configurable, of course.
For now, like eastern has tried, you can put CF to the maximum of 1.5A to drive the ledstrip. However, with all the LED in //, that not optimum, 2A will be better, but that's above the capacity of the current regulator (luxeon driver side of the board). Like he said, even with that, the LEDs are already overdriven, I've been to 31 mA per LED when I tried with the overdriven MR strip and my HB experiment.

So far, before getting in the final design of the controller, I'll build a complete blade so that I have the exact requirements and specs for the board.

**Jedi-Diah** · 10-21-2008, 10:56 PM

Would there be any difference if the led strips were wired in sections for scrolling effect or just all together in // (without sections)?

Originally Posted by erv

that's 100% correct, the makoto design if full parallel, means that the used current is divided equally between each led, or almost (minor differences, if the forward voltage of each LED isn't totally contant, drift over temperature etc).
Like eandori mentionned elsewhere in another thread, it's not the best configuration, but this shouldn't be taken as a generic remark.
If the forward voltage of the LED is like 3V, like the new green one I've used, then you can reasonnably put 2 LEDs in serie, then wire the pairs in //. This will divide the used current by 2 but requires 6 volt to work at an equivalent brightness.
However, with a 7.4V battery setup, 6V are low enough so that you can get autonomy and do regulation. It means that 7.4V will be "adapted" to match 6V, but even with the batteries going down the regulator will maintain the 6V on the output until the battery gets close to, let's say 6.5V.
Knowing also that the batteries will actually start around 8.4V and will remain at 7.4V for a long time (if you choose appropriate capacity versus the used current).

Just like the current CF, if I release a ledstrip version of the board, current, or at least output voltage to the strip will be configurable, of course.
For now, like eastern has tried, you can put CF to the maximum of 1.5A to drive the ledstrip. However, with all the LED in //, that not optimum, 2A will be better, but that's above the capacity of the current regulator (luxeon driver side of the board). Like he said, even with that, the LEDs are already overdriven, I've been to 31 mA per LED when I tried with the overdriven MR strip and my HB experiment.

So far, before getting in the final design of the controller, I'll build a complete blade so that I have the exact requirements and specs for the board.

**Jedi-Diah** · 10-28-2008, 08:34 AM

Dear Erv,

If you built a led strip version of CF, would it be compatable with the normal Luxeon LEDs as well? I mean, could you swap it between your led strip sabers and your Luxeon (Seoul/LedEngin/Cree) sabers or would it only work for led strips?

Thanks,

Originally Posted by erv

that's 100% correct, the makoto design if full parallel, means that the used current is divided equally between each led, or almost (minor differences, if the forward voltage of each LED isn't totally contant, drift over temperature etc).
Like eandori mentionned elsewhere in another thread, it's not the best configuration, but this shouldn't be taken as a generic remark.
If the forward voltage of the LED is like 3V, like the new green one I've used, then you can reasonnably put 2 LEDs in serie, then wire the pairs in //. This will divide the used current by 2 but requires 6 volt to work at an equivalent brightness.
However, with a 7.4V battery setup, 6V are low enough so that you can get autonomy and do regulation. It means that 7.4V will be "adapted" to match 6V, but even with the batteries going down the regulator will maintain the 6V on the output until the battery gets close to, let's say 6.5V.
Knowing also that the batteries will actually start around 8.4V and will remain at 7.4V for a long time (if you choose appropriate capacity versus the used current).

Just like the current CF, if I release a ledstrip version of the board, current, or at least output voltage to the strip will be configurable, of course.
For now, like eastern has tried, you can put CF to the maximum of 1.5A to drive the ledstrip. However, with all the LED in //, that not optimum, 2A will be better, but that's above the capacity of the current regulator (luxeon driver side of the board). Like he said, even with that, the LEDs are already overdriven, I've been to 31 mA per LED when I tried with the overdriven MR strip and my HB experiment.

So far, before getting in the final design of the controller, I'll build a complete blade so that I have the exact requirements and specs for the board.

**erv** · 10-28-2008, 09:01 AM

that will be a specific board, ledstrip driving differs to luxeon driver (which has a true current regulation)
Erv'

**Jedi-Diah** · 10-28-2008, 10:02 AM

Ok. Thanks for the confirmation. I guess I need to start saving so I can get both.

Originally Posted by erv

that will be a specific board, ledstrip driving differs to luxeon driver (which has a true current regulation)
Erv'

**Jedi-Diah** · 11-18-2008, 01:52 PM

I have 2 1000mA buckpucks and they each drive the led strip at exactly 1000mA (with MR battery pack - 4.8V). But when I put them together in parallel (all the same colored wires together), it measures only 1300mA and a little bit brighter (maybe 10-15% brighter) and not sure why. Do I need to increase the voltage? I don't think so. 2x1000mA buckpucks should put out 2000mA.

How can I drive it at 2000mA? These are 71 x 30mA leds all wired in parallel.

**erv** · 11-18-2008, 01:56 PM

it's a bad idea, I don't think buckpucks (and switching mode current regulators in general) are meant to be put in //, it totally fools their individual currement measurement (each one trying to maintain 1A on his own)
You need to find a 2A buckpuck.

**Jedi-Diah** · 11-18-2008, 02:08 PM

Thanks for the explanation Erv. I knew someone would be able to set me straight. I'm using those buckpucks for my kid's sabers anyway (cyan and green lux III).

Originally Posted by erv

it's a bad idea, I don't think buckpucks (and switching mode current regulators in general) are meant to be put in //, it totally fools their individual currement measurement (each one trying to maintain 1A on his own)
You need to find a 2A buckpuck.

**eastern57** · 11-27-2008, 08:12 PM

The emphasis point for this first one is that evenness is also determined by how you actually construct the blade.

This is a comparison between an MR strip, minus resistors (blue) and my latest Makoto LED string (red). The MR strip is uniformly even because of the fact that they're still uniformly mounted on the wafer board. On the red one, you can see that the LEDs are basically having a bad hair day due to sloppy construction - easy fix, but it demonstrates my point.

http://s306.photobucket.com/albums/n...t=MOV04200.flv

This next one is in the open, during the day, all shades open, and as bright as my little living room can get. Strangely, the camera shows the tip to be brighter than the base...

... still, it's dang bright.

http://s306.photobucket.com/albums/n...t=MOV04205.flv