I figure the odds of damaging the lens with any method is high on these lights.
sorry to keep going a little off topic about sc700, but has anyone actually gotten one open before? or seen a report of someone succeeding with this specific series?
while observing with a microscope, i pressed as hard as i could near the bezel ring using a plastic rod but didn't see the lens/oring giving at all. doesn't necessarily mean anything, but it makes me wonder if the usual method doesn't work on this. i'm leaning towards trying to make a small hole in the bezel ring as a spot to start prying.
Haha, thanks - I was on mobile when I asked, so I wasn’t able to find and link this thread.
i just bought a new sc64c le to put a 519A 3500K dedome. first time taking one apart.
i ground down the handle of some metal spudger tool i had on my bench. you could probably make a tool out of almost any suitable piece of steel. just needs to be rigid enough to not flex in use. i worked it on an extra coarse whetstone and put a convex grind on it. the edge needs to be thin enough that you can fully insert the tool under the bezel. the edge radius should also roughly match the bezel radius so you get a decent contact area.
the red line in the image below shows how i flattened the edge profile on the right side (increased the profile radius). you can see the original radius to the left of it which was too narrow and didn't make enough contact.
after dialing in the blade geometry, the bezel came off very easily.
choil shot of the blade:
here's the driver removed. i added some test leads to test the circuit ex situ. spoiler: i didn't break it :)
i think it's interesting that the potting all came out with the pcb. i haven't seen a pic of this online before, so idk how common that is.
album link: https://imgur.com/a/XAKnZ2G
questions...
what are you replacing the TIM with? i have some Noctua NT-H2 paste, but i'm not sure if i want to use it because i don't think it's designed for long term application. i think performance degrades after a few years, and it's meant to be re-applied. depending on how i reflow, maybe i can just re-use the existing paste (but probably not).
how are you guys reflowing? my IR preheater is in storage, so i probably won't pull it out just for this. also, the potting is kind of in the way on the bottom. maybe i'll just do classic hot air from above (maybe with some gentle preheating with a chem lab style hotplate, but those are a little hard to get dialed in). or maybe i'll try pressing a flat style iron tip onto the brass block from below, but i don't have a good board vise rn.
Take a large tip for your iron and apply it to the copper block from below. This works much better than hot air.
all's well that ends well. if i ever have to do another one of these for some reason, i'm sure it will go much faster and smoother. but it wasn't bad honestly.
i used a knipex pliers-wrench with a rubberband as a board clamp because i have nothing else available:
i wasn't getting fast enough heat transfer with a dry (but tinned ofc) iron tip (used a 45° flat chisel type for good contact area). i didn't want to use solder because i wanted to leave the nice gold plating alone, and i wanted it to stay as flat as possible.
instead, i used hot air on the block from below. i left the potting on the underside of the board intact and used it as thermal insulation. aimed the air away from the switch and towards the potted components. worked fine, but i didn't perfectly center the emitter. only noticed after reassembling and seeing that it wasn't perfectly centered in the reflector. oh well. not worth it to redo it.
i dedomed the 519A sm353 (3500K) with a double edge safety razor blade. just did it freehand after reflowing. i gently sliced it off in about 3 slices. i did slice cutting instead of push cutting. tried to minimize stress on the emitter package while doing so, particularly normal stress. tried to make it mostly shear stress. i got the emitter from @azhu btw if anyone's looking for some.
i used noctua nt-h2 paste because i was impatient and didn't want to wait for shipping, but arctic mx-4 or Innovation Cooling Diamond might be be better choices (might last longer without drying out).
i removed the solder on the VCC pad and GND pads (by the screws). i diluted it with some eutectic Pb-Sn to lower the melting point and make cleanup easier, and then i pulled it with wick. then i added fresh Pb-Sn. this made it a little easier to re-solder the supply (especially the screws) at the very end.
new emitter installed and dedomed:
before reassembly, i chamfered the edges of the potting so it would slip into the milled recess more easily since it has to rotate into place.
i used a parallel clamp to press the bezel. worked well and gave me decent control:
beam shot:
left to right:
- sc64w hi (high cri 4000K 90+ XHP35 HI, bob)
- sc64c le (Nichia 219B R9080 sw35 3500K, bob)
- sc64c le (Nichia 519A R9080 sm353 3500K M400 L5 dedomed, me)
- d4v2 (Nichia 519A R9080 sm353 3500K dedomed, stock from hank)
1 Thank
Good job!
Looks fantastic!