Amber 1w Direct Drive on a 123?

I've got a nice little S2F Amber that I'd like to put into a small light. Has anyone done an Amber (or red or R/O) direct drive?

I'm worried about frying this little gem, but I also don't want to throw a resistor or buck converter into the mix unless I really have to.

Jon

I have a T2H R/O luxeon direct drive E1e mod that uses a 123. Great little flood light and super bright. It is a H bin for the Vf and yours is a very low F bin, so I don't think this helps you very much.

You'll probably be okay. The red/orange/amber Luxeons have a high dynamic resistance, which means the Vf will increase quite a lot with increasing drive current. You be might overdriving it a bit DD from 3V. One way to be certian is to measure the current (in a quick test) using the 10A current scale on a DMM.

Let me see if I can dig up my R/O tests to see how the Vf scales with current...
Nevermind, my data is at home. Let me put it this way: You definitely can't drive an amber 1W to 600mA or more on 3V. I'm sure I measured the Vf of my R/O (H Vf bin) at more than 3.6V at 600mA.

Excellent, that's encouraging. I think I'll go for it. /ubbthreads/images/graemlins/grin.gif

Jon

I have apparently misplaced my data, so here are some fresh measurements on an H Vf binned R/O:
<font class="small">Code:</font><hr /><pre>
just lighting 1.69V
30mA 1.97V
130mA 2.28V
310mA 2.71V
440mA 2.97V
620mA 3.37V
750mA 3.62V
1000mA 4.38V
</pre><hr />

I wonder if I can safely extrapolate from these numbers what my F ranked amber will do...

If H Vf is nominally 3.1 and F Vf is nominally 2.6, there should be a .5 volt difference in their forward voltages when running at spec.

If we then assume (probably incorrectly?) that 3 volts Vin to an F ranked amber luxeon is equivalent to 3.5 volts Vin to an H ranked R/O luxeon, then I will end up driving the amber somewhere between 620ma and 750ma. /ubbthreads/images/graemlins/icon3.gif

Shouldn't an H Vf luxeon be pulling somewhere between 350 and 400 ma at 3 volts? By your numbers that R/O is actually a high G?

Jon

[ QUOTE ]
StoneDog said:
I wonder if I can safely extrapolate from these numbers what my F ranked amber will do...

If H Vf is nominally 3.1 and F Vf is nominally 2.6, there should be a .5 volt difference in their forward voltages when running at spec.

If we then assume (probably incorrectly?) that 3 volts Vin to an F ranked amber luxeon is equivalent to 3.5 volts Vin to an H ranked R/O luxeon, then I will end up driving the amber somewhere between 620ma and 750ma. /ubbthreads/images/graemlins/icon3.gif

Jon

[/ QUOTE ]
Jon, this is a reasonable extrapolation if you *assume* that they both have the same dynamic resistance. There is, however, variation from LED to LED w.r.t. dynamic resistance.

[ QUOTE ]
StoneDog said:
Shouldn't an H Vf luxeon be pulling somewhere between 350 and 400 ma at 3 volts? By your numbers that R/O is actually a high G?

Jon

[/ QUOTE ]
No. The Vf bins are assigned at a Tj=25C. His test data is taken at a much higher Vf. Vf drops with rising temp.

Doug, thanks - I figured I was wrong on at least a few points in that post.

OK, so if Vf is assigned based on Tj (and not ma) that implies that the much coveted "H" ranked white lambertians may not necessarily be more efficient than a "K" or "L" with respect to light output and battery life? It seems to be a commonly held belief that you'll get better performance underdriving an R2H at say 150ma than you will an R2L @ 150ma - but is this really the case? Does the same hold true for both when they are overdriven at say 700 ma?

/ubbthreads/images/graemlins/thinking.gif

Jon

[ QUOTE ]
StoneDog said:
I wonder if I can safely extrapolate from these numbers what my F ranked amber will do...

If H Vf is nominally 3.1 and F Vf is nominally 2.6, there should be a .5 volt difference in their forward voltages when running at spec.

If we then assume (probably incorrectly?) that 3 volts Vin to an F ranked amber luxeon is equivalent to 3.5 volts Vin to an H ranked R/O luxeon, then I will end up driving the amber somewhere between 620ma and 750ma. /ubbthreads/images/graemlins/icon3.gif

Shouldn't an H Vf luxeon be pulling somewhere between 350 and 400 ma at 3 volts? By your numbers that R/O is actually a high G?

Jon

[/ QUOTE ]

Except that your battery's terminal voltage will drop with increasing current, due to the internal resistance of the battery. A lithium 123 cell has about 0.25 ohms of internal resistance. This will cause the battery voltage to drop under load, reducing the current.

Let's assume an 0.5V lower Vf (which we don't know, it's pure speculation until you actually measure the Vf at 350mA), and say it gives about 700mA. You must now see if that's possible given the battery's internal resistance.

At 700mA, the battery's terminal voltage will drop to about 2.8V. But we've assumed that the Vf was 3.0V, so that solution is no good. Vbatt must equal Vf in DD. I'll guess that the real solution gives you about 500mA current.

However, it is quite likely that the Vf of your amber is near the high end of the Vf range, so even with the lower binned Vf, I'll still say it will be around 500mA or lower. Best way to find out is to use a DMM to measure the current from a 123 cell in DD. Only test for a couple seconds just in case.

I don't have a DMM handy but I believe my father does - I'll try to test it this weekend.

I have a follow-up question about the ranking then... If Lumiled ranks forward voltage at Tj=25C, does it also rank luminous flux at this point - or do they measure that at a constant current level (350ma? or 375ma?)?

Jon

[ QUOTE ]
StoneDog said:
I don't have a DMM handy but I believe my father does - I'll try to test it this weekend.

I have a follow-up question about the ranking then... If Lumiled ranks forward voltage at Tj=25C, does it also rank luminous flux at this point - or do they measure that at a constant current level (350ma? or 375ma?)?

Jon

[/ QUOTE ]
Flux, Vf, and color bins are all under the conditions of I=350 mA and Tj=25C.

Interesting...

Thanks for your help everyone.

I just had my second P4GR red /o direct drive off 2 AA's die. I can't figure out why. They were running just fine not even getting warm. One died about 3 weeks ago and the other last week. Anybody have any clues? According to the bin codes I was right in spec.

Mutech

The 2 dead R/O worries me a little...

I still haven't had time to test the amber (still need a DMM). I'm planning on making a DD "sandwich" from 1/8" copper and I may through a 1 ohm resistor in there just to be safe.

Jon

Well, I built a module over the weekend with 1/8" copper slug over two pieces of Micarta. It's basically the same dimensions as a MM or BB sandwich. I very carefully placed the Amber emitter on it, securing it with JB Weld (didn't have any AA Epoxy).

The plan was originally to use this in a CNC-123 but my 6 year old son saw it and really liked the orange beam, so I put it in a mini-mag for him.

Unfortunately It died after about 30 seconds. I have a feeling the body or slug wasn't completely isolated. An immediate post mortem didn't reveal an overly hot copper slug, so I suspect the LS body wasn't completely isolated (thanks to either the JB weld or my bad soldering job) and some sort of short occured. I removed the emitter from the homemade module and tested it seperately - no luck, it doesn't fire up at all.

My interest in Amber emitters has passed... /ubbthreads/images/graemlins/frown.gif

Jon

Can you look at the die with a microscope or magnifier? Are you able to tell if the bond wires fused?

If you didn't properly isolate the slug, then the bond wire going from the positive tab on the emitter to the slug (so not the bond wire attached to the top of the die) will have fused open.

The closest I can come to that is the macro lens on my Minolta 7Hi. I'll take a few pics and see .

When the emitter died it dropped from bright to dim (such that I could look at the glowing die quite easily). I turned the light off, back on and it was bright for a moment, then dimmed again, and slowly faded out completely. Does this sound like a fused bond wire? If so can this sort of thing be fixed?

Jon

I took a few macro shots (quality isn't good enough to post). From what I can tell the positive bond wire arcs from the side of the emitter to a spot on the silver "dish" pretty close to the positive lead. I just can't tell if actually touches the dish or not - the wire seems to end in a small bead...

Jon

[ QUOTE ]
StoneDog said:
I took a few macro shots (quality isn't good enough to post). From what I can tell the positive bond wire arcs from the side of the emitter to a spot on the silver "dish" pretty close to the positive lead. I just can't tell if actually touches the dish or not - the wire seems to end in a small bead...

Jon

[/ QUOTE ]

emitters can be fixed? Mine all look perfect, they just don't work anymore...

Apparently they can't be fixed... /ubbthreads/images/graemlins/smile.gif

Jon