A driver normally regulates current to the LED, but in your case I think it probably just lets current through with PWM for the lower modes (hence why the LED burned out instead of the driver). In that case you have a similar situation to hooking up a cell to the LED directly, just with a bit of resistance.
The XM-L can handle currents up to around 3 amps and running semi direct drive with a li-ion should balance out at 4 amps max and probably below 3 amps at 4.2v, then dropping as the cell depletes.
The Nichia 219 has a slightly higher forward voltage, so would settle at a lower current with the same battery. However, the current is still likely to be above 2 amps, maybe around 2.5, this would be fine for an XM-L of course but will fry a Nichia 219.
Sorry, XP-E will probably pull similar current to the 219, but has an even lower max current rating. So that would be a no go as well I’m afraid.
If I’m understanding what you said, only the Nanjg one would be ok for the Nichia and/or XP-E?
Also, theoretical question: There are two 7135s on the driver I have now (the original one from the broken drop-in - picture in this thread earlier). Isn’t it like 380 mA per 7135? So, isn’t the original driver more like a 700 - 760 mA driver?
Another theoretical question: What if I desoldered one of the 7135s from the original drop-in driver? Wouldn’t that reduce the amount of current it could source?
Yeah, that NANJG 1400ma one would be a good match for a Nichia 219. I’m afraid I don’t know enough about the original driver to tell you, I’m not an electronics expert, just someone in a similar position to you but three years down the line
Are you sure they are AMC7135s? they look similar but could perhaps be something with a different function altogether…
I’m not familiar with the chip or your driver. I take it the seller didn’t give you specs on the driver? You could ask them.
Since you had a high voltage incident, you probably should measure the output yourself anyway. Hook up an emitter or put the driver in a different light. Take off the tail cap. Use the 10A (might be 20A, depending on DMM) setting and move the probe to the 10A port. Use the leads to complete the circuit between the cell and the body. You can switch modes by momentarily breaking contact with the body.
EDIT: You probably should not exceed 1.5amps with a Nichia in a P60.
I’m not too trusting of fasttech’s current readings, I would hook an xm-l up to it and test it first, you also do not know if you fried anything on the driver, your reading an open circuit reading, but the driver may have just gone direct drive ie your reading cell voltage minus any losses in circuit. Once you have some idea of its output, then worry about actually building something up with it.
So I got a Nichia 219 and an XP-E in today, and tested with a series resistor on the driver output, and both worked.
I know that I shouldn’t have done it, but I then tested both the 219 and the XP-E without the resistor, but only for really short time (< 1 sec), and they both worked. I didn’t dare test it for longer like that, because the emitters are just on a PCB and not attached to any pill or anything.
So, now the waiting game waiting for my XM-L emitters to come in :), but it does look like the driver is ok, because I could change modes, to SOS and strobe.
FYI, I went ahead and replaced the dead emitter with the Nichia 219, and, it works.
HOWEVER, I did such a lousy job with the soldering that I couldn’t get a new insulator disk seated. I was about to throw in towel (I really should have), but I went and tried to screw the reflector onto the base/pill, and it would barely thread, because the pill was so bent.
Anyway, I got it screwed in enough and then tried it in an L2P, and it worked!! Let there be light!
So, that’s the good news…
The bad news is that the LED is so horribly centered (or rather, not centered) that I get barely any light out :(…
Anyway, it was an interesting experience, but (1) I don’t think that I’ll be doing this again, and (2) my admiration for those of you who can do this kind of thing, and do it well, has increased exponentially :)!
Now that I’ve had a day to recover, I’m trying to understand what I did wrong. Is it because I was trying to use the Nichia with a 10mm PCB in a pill that originally used 16mm PCB? If I had a 16mm PCB, would it have been a lot easier?
Or, is getting the emitter centered ALWAYS so difficult?
Usually screwing down the reflector onto the LED while the Fujik is still wet will automatically center it for you. Since your reflector is for an XML, the hole is larger and will not tighten down around the LED. Also, with the smaller diameter PCB, perhaps one of the wires was too short, and kept it pulled to one side. I no longer work on PCBs while the FujuiK is still wet, it gets all over everything. I’ll tin the contacts, apply paste, screw down the reflector using a matchstick or something from the inside, to make sure the holes are aligned. Then I’ll let it dry before threading the wires and working on things.
In your case, it would have been easier if you carefully epoxied an XML centering ring to the PCB, making sure it was nicely centered, before screwing down the reflector. The problem you hade with barely getting the reflector to thread onto the pill could be, if your Nichia PCB came from IS, their NIchia PCBs are much thicker than the norm.
I have some proper parts coming SLOWLY on their way, and will try again. Plus, I didn’t have one of those 3rd hands last night, and you should have seen me try to solder the leads onto the emitter PCB as the PCB scooted around the table :)… No fun at all.
Edit: BTW, I don’t think I could’ve screwed the reflector all way way in. The rim of the pill was so bent in places that I could barely get the reflector to make one turn.
Just a note: All the problems you encountered, from centering, to screwed up threads, I also experienced with my first P60 build. You learn from your mistakes. The second build will go much better, and by the third you will have developed little tricks of your own, and it will look very professional. After your third build you will look at your first, shake your head, buy a new empty module for $2.40, and rebuild the whole damn thing.
This is pretty scary, but you must’ve been reading my mind :)…
Earlier, I felt like trashing the thing, but I was up late last night/this morning, and pulled the semi-working drop-in out to mess with it again. Wasn’t able to do much with it with the tools I have yet, so I put it back, pending another day.
So yes, I’ll hang onto it and definitely will rebuild it at some point.
one trick I used before I got a little vise and helping hands, use a pair of mole grips and adjust them to very lightly grip the pill (ie no thread damage etc) you can now rest one of your wrists on the mole grips, its now located whilst you solder. I built a few pills like this.
nice job on getting light out though, as said, just learn from the experience and move on to the next build.
