Well, I added two chips to a 3.04A driver this evening. One went good, the other…Im not sure. It seems that there’s that one chip that’s easy to get at, and the others are harder. Poor lighting, bad eyes, and just getting weary, no matter how hard I examine the second chip, I still can’t tell if the contacts are connected. Well one of them anyway. I’ll study it again tomorrow. I reflowed an old XPG onto a board and soldered wires with little alligator clips to it. Now I can hook up my drivers to make sure they work before installing them. The driver I was chipping tonight lit it up, so at least I know I didn’t destroy it.
I just thought of something. I forgot to put fresh tinning on the legs of the existing chips before starting out. I wonder if that went against me.
Pre-tinning the base chips definitely helps as you have slightly better access when the new chips is not in the way.
You are testing a 3A+ driver with an XP-G? J)
Yup. I’m not really running it. I just hook it up to the driver, and then touch the battery leads to it, so I know the driver still works after my clumsy modding attempts. It sure was bright though.
Can you tell me about this flux pen I bought? It has a felt tip that presses in, and the flux is like clear water. I don't know if its good or proper for modding. Maybe for applying on MCPCB befor tinning for LED reflow? http://canada.newark.com/kester/83-1000-0951/flux-pen-no-clean-0-33fl-oz-pen/dp/82H2047
Avoid running that bare XP-G star for very long… it will get hot fast.
I haven’t used those pens before. I use a syringe and it’s like corn syrup. Like water seems too runny and would spread out too quick. Give it a shot. Put some on a virgin PCB pad (never soldered) and apply heat and solder. The solder should quickly spread out to cover the pad. This demonstrates best on a longer pad, maybe 3-4mm. If it sits right around where the iron is and doesn’t flow across the pad, the flux isn’t doing a great job.
I usually only use additional flux for reflowing larger IC (8-pin and up). I have not used it on an LED yet, I just reflow by pre-tinning the star, heating through the tinned wire pads and placing the LED when the star is hot enough. I have flux and solder paste, but I’m too lazy to break it out for an emitter.
I tried to reflow an LED last night by heating the PCB from underneath with my iron. I think the metal from the “helping hands” was just stealing all my heat. I finally had to use a frying pan. I need to devise some sort of clamp, or vise grips with leather pads on the jaws to insulate the star. I don’t have a workshop, so everything I do has to be done on th kitchen table.
The clamp will not steal much heat, but open air does.
When heating it up, I tin the emitter pads and then of the wire solder pads, leaving the iron on it. If it’s a copper sinkpad I heat through the center pad and move to a wire pad when things are melty.
Then I take a piece of paper towel and lightly wrap the star a bit with it. Saves a lot of time and the paper will not burn if it lightly touches a 400C iron.
Once the emitter pads appear to have melted again , I get paper out of the way and use tweezers to place the emitter into position. Note that I do not have the emitter in place during the heating phase (this may be bad for the emitter, heat shock maybe, I dunno; haven’t killed an emitter yet).
Once I’m happy with the positioning, I remove the heat and wait for things to cool.
All done!
Even though this method works well for me, I can’t wait for my hot air reflow station. 
I’m not following you. Are you heating from below? I can’t make sense in my mind of what you’re doing with the paper. Isn’t it hard to hold the iron to the star, and place the emitter with tweezers all at the sme time?
The iron is held on one of the wire pads with solder on it. That’s my heat entry point. I find this heats things up quite quick. Heating from below would work too, but direct iron-to-solder-to-pad contact on the center pad or the wire pad heats quicker for me.
The paper insulates the star so the ambient air doesn’t steal it away. It heats up faster.
I can complete a reflow on aluminum in about 60 seconds using this method. Probably double that on copper.
The second time I did reflow, I did 6 on a hot plate at once. Probably works out to about 60 seconds apiece all told. Neat, no burns, no mess. I was quite proud. Only done it once more since then. I like the hot plate way.
This is a 6” round piece of 1/8” plate Stainless Steel that covers the small burner on a glass stove top. The plate keeps the burner on and keeps a uniform heat to the multiple stars. I put the solder paste on, place the emitter and line em up as the plate warms. Soon as they reflow, I remove em to a 3” cube of 6061 for the cooldown. Simple, easy, no fuss.
That’s a great way to do it. I will have to look into that. I don’t have a gas stove, maybe my barbeque side burner.
Toaster ovens are good for makeshift reflow ovens, and they can be had for cheap. They have decent temperature control, lots of space for big boards, and you might even be able to flame some ano in it too!
I wonder if a SS plate on an induction hotplate will work? They have a temperature sensor in them to limit the temperature. I’ll have to try that. I have two small induction hotplates that I rarely use.
WOW!!! That was fast!!! The solder melted in about a minute. It got up to about 150C and things got melty. Cool!!!
As a test, I took my little induction stove (actually it’s a ‘toy’ stove, shaped like a duck). Set the pot on there, put an empty star in the middle with a piece of solder on it. Turned it on high and waited a bit. After about 30 seconds and a few snaps and cracks from the pot, I switched to low. Not very long after, the solder melted into the star pads. I think I have a new way to reflow!
I had heard of the skillet method, but the hassle of going to the kitchen instead of the minute or two right there at the iron… now I don’t have an excuse. Well, I probably will still go to the iron… it’s always set up and ready to go.
Thanks for the tip! if I have to reflow a bunch of emitters at once, this will be the way I do it.
I only have an electric element kitchen stove. I use a small heavy aluminum frypan. I place the star in the pan with the LED in place and turn the stove on high. As soon as I see the emitter sink down on the star I pluck it out with tweezers.
I have to be quick, as there is no real heat control with an electric element. I was looking at these small portable single burner propane stoves for woks, at the asian grocery accross the street($12). Quite small, hook up a gas bottle, and I would have more controlled heat. I’m jhust leery about having a gas source in an apartment.
I’m still looking for something small and inexpensive to suit my reflow needs.
Glad to help Relic, I meant to say that our glass topped stove is an induction unit but could’nt think of that term “induction”.
lol
I got the stainless plate when my wife and I were trying to can some plum jelly and found the bottom of our big canning pot had a recess in it and the induction surface wouldn’t stay on. The big ol pot was made in the day of circular heating elements that are usually raised, the recess helps keep the pot centered on the burner. Made it really tough to keep that big pot boiling on a smooth glass induction burner! With this 6” circle of SS on the element, that’s no longer a problem and the heat control works as it’s supposed to. I turn mine up to about 6, then off when the flux shows up. When the solder melts seconds after that I move the plate off that burner to the cool one above it and give it a few more seconds to start cooling before removing stars to the aluminum heat sink. That SS plate stays hot for quite a while!
The plate also keeps me out of trouble concerning the wife’s cook pans and solder with lead in it. 
Ouchyfoot can’t you just not turn the burner on Hi? Use Med heat? Or, heat up the pan first, then lay the star in the pan with the solder paste and emitter already on it. As the pan is cooling down, it’ll heat up the star and reflow the emitter without excess building of heat that might scorch your star or worse, fry your emitter.
Hey Dale, induction is a good heat sourse for this. You get good heat control, and there’s no flame or scorching hot element when you’re done.
I posted this on another thread, but since we're talking about it, I'm reposting it here as well.
Okay. I think I've got it. Just so you know, I have to work on my dining table, so everything has to be small and simple. The last time I tried, I couldn't get the star hot enough because the metal in the "Helping Hands" was stealing my heat. Today I bought a small vise grips style c-clamp, and glued some leather to the "clamps". This worked perfectly, almost too good. It was so fast, you can see where the topside of the star scorched brown...it scorched the leather too. Next time a little less heat. Usually, I can see the LED settle down, but I missed it. I used a flux pen on the contacts before I tinned them. This pen has a felt tip, and it paints on flux like a marker pen. The flux is clear, like water, or clear marker ink. When I dabbed on the tinning, it went on so evenly, that the Led sat so flat, there was no noticable settling for me to see. Anyway, scorched or not, I tested the LED and it works fine.
Looks like smores; take two stars and put a mini marshmallow in between… Nice and toasty!
As long as it works, right? You’ll get the temperature right and now you know to look for a very subtle LED settling.
I didn’t think an alligator clip would steal that much heat, I didn’t notice much and I use helping hands.
The leather is a good idea, insulates well and resists burning for a while.
Actually, this was faster than the frying pan on an electric stove, and I don’t have to be in the kitchen, hunched over with my face close to a hot frying pan, let alone risk dropping the whole thing upside down in the pan while I’m trying to snatch it out with tweezers.
Just tried adding 7135 chips to a Nanjg 105c for the first time. Surprisingly easy with no difficulties and all chips appear to work.
I added 3x 380 mah chips to a 2.8 amp driver. So output should be around 4 amps.
Technique I used:
- Bend all 3 pins on the 7135 to be added so they point straight down.
- Super-glue the new 7135 on top of the old 7135.
- Apply flux to all 3 pins and the ground pad for both the lower and upper 7135.
- Using large soldering iron soldered ground tab using the techniques described in this thead (loaded iron up with generous amount of solder, then placed solder blob at bottom 7135’s tab and then dragged iron upwards towards top tab.
- Using a very fine tip low wattage soldering iron, soldered all 3 pins of the new to the pins of the lower 7135. Because I’d bent down the new 7135’s pins, they were actually touching the pins of the lower 7135 even before I soldered, so soldering was very easy and required very little solder. I kept the driver board in a small vise to keep it immobile while I did the soldering.
- Used a small file to file off excess solder from around the driver tabs so the driver would fit in my light’s pill.
To help make it easier to see what I was doing I held an Olight S10 modded with an XML2 neutral on medium power in my mouth while soldering.
Result is pretty nice. I now have a 1xAA sized Sipik 58 with copper heatsink running on the new 4 amp driver and powering a triple Nichia 219. Runs on an AW IMR 14500. At max power this light heats up fast so can’t be run more than a few minutes at a time. Also runtime on a full IMR 14500 should only be around 9 minutes. But for short bursts it gives a wide blast of beautiful Nichia tint light.
I also tried this driver powering a triple XPG2 5,000k neutral. Output was much brighter than with the Nichia, but had a much worse lemon yellow tint. I suspect the XPG2 needs more than 1.3 amps per emitter before it starts outputting a better tint. I think I might buy some cool white XPG2s to try out. Perhaps they will have a whiter tint. Or perhaps I could try a triple with 1-2 coolwhite XPG2s and the rest neutral tint.
What’s the proper way to attach 7350 chips to a bare board. If I had a driver with only five or six chips on it and a couple of spots that are empty.
Firelight2, the XP-G2’s in my EDC+ Triple are a very nice white tint, running all 3 at 2A (1.93A measured at the tail) They probably will change tint driven harder, but the 3 in my drop-in are a perfect white…no tint anywhere.