NUBM07E 465nm 2.9W Diode Test (Hitting 470nm+)

Alaskan said:

I pushed mine hard at 4 amps (or was it 4.5? Probably.) a few months ago and it died after just a few uses, but my batteries weren't firmly touching together inside the host causing rapid on-off when I wiggled the host, that killed the diode but the driver was a soft start driver, shouldn't have done that.

Click to expand...

Do you know if there's an inductor in the current path on the driver? Inductors don't like it when you turn off their current suddenly and generate lots of voltage to keep things the way they are (e.g. that's how points worked in older cars).

They "SXD" are op amp buck drivers and they use an inductor, there is a soft start cap and these drivers can be tuned for different " ramp up " lengths, but in the case of quick on/off spikes from an intermittent battery connection that could have hammered it with the full 4.5 amps square wave style.
I haven't studied the drivers, but I don't think the soft start cap is a fast bleed down, I just tested the output against a 1 ohm load......I would recommend setting max current and checking it with fully charged batteries, the cut off is 6.8 volts and if you don't use INR/IMR cells the sag driving an 07E or NUBM44 can be substantial, but the SXD will take 3 cells in series, up to 14 or 18 volts I think, but I want to test the next one I get with 3 INR cells in series and see how stable they are over a multi minute run with adequate heat sinking, my only issue was with the little pot, it is very touchy it seems and a German member speculated that the graphite in the little single turn pot that has a wide adjustment range could be effected by heat.
Battery voltage never dropped much when I tested I used fully charged INR cells " new LG's " but output had a range so I would set on fully charged cells.

The inductor ran around 165 f average with the 8 pin chip clamped against a 3/4 inch thick 4 inch by 5 inch block of aluminum, as the heat got into the pot I'm speculating the resistance may change a little bit, Lazereer explained the quarter amp variance or so I recall, but I'm not the power expert, I have a power expert in house and we could scope the driver and dissect it but they work in my 07E and 44, I would just set it south of 4 amps for the 07E.

SXD aren't boost, SXB are. Did you use the wrong term?



I don't think it is likely the driver, I will confirm later.

Edit: I just watched your video, I had a setup like that and found the long leads from the driver to the diode threw off my current setting a huge amount when I then shortened them for use in a host. DTR helped me figure out what was going on, I really didn't think a few inches of power lead to the diode would do that, but six inches of alligator clip-on leads plus the long leads from my multimeter measuring the current together with long leads to my power supply threw everything off a huge amount. I had a lot more wire in there for my setup than you do in that video, so maybe you are ok, but I'd check that.

Alaskan said:

SXD aren't boost, SXB are. Did you use the wrong term?

I don't think it is likely the driver, I will confirm later.

Click to expand...

Do check the driver. I have had a few SXDs mysteriously quit. They were heatsinked well, just like the at least 25 other ones I've built with 4A or 4.5A SXDs. No clue what happened with them.

I will, but I'm pretty sure it's the diode, the beam now has a dark stripe in the middle of its output and only puts out 175mw. Maybe that's a symptom of under current?

Yes I know it's a buck, I just said boost because my draim is bamaged.

Maybe our diodes drop resistance as they warm up and the driver is designed to mirror this but my 1 ohm resistor does not change with heat?

Maybe I just have a funny one, it's the driver I bought already set at 5 amps with my original nubm44 that wore out.

I will test the next one I buy and see if it's different.

I know adjusting the little pot is very touchy, I saw where Tater had trouble getting them to the lower range, it's just odd that it wont stay stable.


@ Alaskan: Yes, I had 6 inches of wire between my SXD and diode and it caused it to fall well short of it's pre set output, my 07E has about 2.5 inches of wire between it and the SXD, it may be only getting 3.2 amps, I don't know.

Funny thing is a little extra wire should not effect a current limiting driver, maybe the voltage is limited as well? Anybody have a schematic for the chip/driver, or a part number? I searched and can't find anything.

I believe the SXD-V3 I bought from Flamminpyro was made by Angelo, good luck getting a schematic, I believe it is considered proprietary.

Well sad news on my NUBM07E and I really liked this wavelength , I didn't even abuse it :cryyy: .
Oh well, I bet the next one with G-ball on will look cool, it will fit in the regular module, just no adjustable lens until I add something post G-ball.
It is a bright gorgeous color.

I heat sinked the heck out of it and kept it at a conservative 3.5 amps and it was strong as ever until it just started dropping output just like I have seen others do when they dim down, I checked my SXD driver and into a 1 ohm load it puts out 3.4 amps just fine.

I really do not have any where near the hours of use on this diode as I do on the typical 7875 when they finally dim down.

I think I am going to run the next one with the G-ball in place and see how it holds up.
I don't think 3.5 amps was too much as it is listed as individual max on the data sheet, and it was heavily heat sinked and never got super hot, the way it went was a lot like my decanned nubm44, I hate to say it, but maybe running these exposed to oxygen is not conducive to longevity? But I could be wrong, maybe I have introduced contamination, maybe humidity is an issue because I change lenses from time to time, still total runtime would be under 10 hours.
I wish it was a bad driver, but the SXD is working just fine.

I'd like another one, but since my last one also died without much use, I'm concerned and wondering if I should buy a diode with the ball lens still on it. Do we really know if keeping the ball lens on it will keep oxygen from reaching the emitter? Is it really filled with a inert gas?

I will keep the Gball on my next 07E, you can still press it into a module or BudgetBeams still has MX-900 heatsinks you can just screw together.

http://laserpointerforums.com/f39/fs-budgetbeams-inventory-95965.html

Supposedly the Gball is almost focused, so fire it through a cylindrical pair and add a BE if you want or use as is.

Are most diode cans filled with inert gas? I don't want an extra piece of glass with its added loss if I don't need to keep it. If leaving the ball in does keep oxygen out I could just expand the beam and a cylinder pair would sure do that, except I like my beams expanded much larger than any cylinder pair I've been able to find.

What you want to do is use the cylindrical pair to square up the beam and reduce that awful aggressive divergence on the fast axis, then your expander will make a nice square spot at distance instead of a rectangle, and it will be tighter.

But I want the best of both worlds; the squared up beam as well as decimal point mRad divergence, under .5 works for me, .25 better. But what of the oxygen concern with one of these uncapped diodes, valid? Anyone? DTR where are you?

Without the correction we are clipping the ends of the highly divergent axis in many cases, also the size of the BE has to be wider to do the job, with cyl correction you have much less divergence and a smaller BE can be used yielding a tighter near field beam as well as a tight far field beam.

Even corrected, only one side has changed, the diameter is limited to the width of the uncorrected side either way. I will take the improvent to the slow axis any day though, but that won't affect my choice of lens diameter, I will still use rediculously large lenses regardless

I believe the expanded multimode part of the beam is the slow axis. The fast axis is perpendicular to the long facet of the die regardless of the diode being a single or multimode. I'm getting very interested in this diode for a new build.