Truthfully, I get nervous every time I have to go near something like that. I hate soldering those things! I have a table top light that has a big magnifying lens in it. I can pull it over top of the work area. I also have to position the boards in a small vise, where they cannot move and where I can rest both hands on it. Otherwise I can’t hold still enough. It’s hit or miss and y’all don’t get to see the mistakes, LOL, but mistakes, they are a plenty, believe me.
I would love to buy a real micro soldering gun or station, but the good ones, with the really thin tips, are way too much money. I usually file my tips to a finer point, even though it ends up ruining the tip much quicker, but I can't solder that stuff with a regular tip. I know one thing, you want a good, hot iron. At least 40 watt or more. It is so much easier to just hit the spot for a second, rather than to have to hold in place, waiting for things to get hot. I never knew that, till I read it from another member. It works!
I use 0.032" solder, 60/40 rosin core. I get it from Radio Shack in spools and wrap a few yards at a time on a 3/8 rod, so I only have to hold the small coil when I’m soldering.
I have to say I am very impressed. Even at a glance the level of quality you are able to produce without the benefit of expensive equipment is clear. The amount of patience shown in your work is admirable.
I think the upgrade to the battery carrier is probly the most creative part, it has a professional look about it too.
This is not a criticism, merely an observation based on my experience. The addition of those extra chips being soldered on top of each other is probly going give you the opposite effect of what you are looking for.
Heat is a very serious issue with these 7135 circuits ( as you probly know). Piggy backing those chips like that will compound the issue. I have found without exception when these boards are slaved it also causes a heat issue.
My personal SST-90 mglite is powered by a set of 3 drivers like that ( not as pretty as yours ), but in order to keep the current stable I had to heat sink both sides of the boards to fairly thick aluminum plates that touched the outer light wall. Without this the current will fall like a rock around the 1 minute mark till it bottoms out at like only a couple of amps.
My admittedly limited experience has been that the thermal issues can sometimes be attributed to excess heat from the LEDs. If the led sink is capable of maintaining a constant temperature then you could be okay. If the sink gets hot then the chips will get hot even if they are not being overdriven.
Heat sinks arrived today. I bought these from Speedy Metals.
They measure 1.316-1.317" OD and a typical Maglite ID is from 1.340-1.344", so I will need anywhere from .010-.012" shim/sleeve, to make it tight in the tube.
They are 1-5/16" aluminum rod stock. It's as close as I am going to get for a Maglite heat sink, unless I have some machined. I will tell you why I do not have some machined. These are about $2.00 each, shipping included. I can live with having to sleeve them with 0.010" copper or aluminum sheet, to make them work. It's close enough for me and it works for the budget. That's what it's all about here isn't it? Trying to make something that works, with what you can afford. Even if it is just a shelf queen.
This week-end, hopefully, I will get to work on finishing this build up.
I would say that I’m surprised the couple of extra chips would do so much to overheat. I haven’t experienced that yet, but there’s always a first time. The drivers will have a heatsink, but I wasn’t planning on a heatsink sandwich. I guess I will find out fairly quickly. Thanks for the heads up.
I also plan to include 3 dummy AA NiMHs, just in case I need to do the pack as 3S/3P, instead of 4S/3P.
Very interesting idea regarding the LED on copper: reminds me of a Led Lenser H7 knockoff I pulled apart a couple of days ago, the leads were wired on the bottom (XR-E) and the center of the LED was pressed (no solder) onto a central plateau.
If you could find a way to electrically isolate 3 pieces of copper while still allowing them to shed heat-easily you could potentially have a quick and easy way of making direct copper bonded LEDs… something like this would work well for isolation and heat transfer: http://www.e6cvd.com/cvd/page.jsp?pageid=309&prod=16 might get a pit expensive though
Are there any real benefits to in increasing heat conductivity to the cathode and anode of the LED?
Its looking really promising. You will be in trouble when the people who were talking about the bright light find out who it is. Looking forward to your night shots.
ha, loved the comment about the neighbours commenting about the light - can’t get a better complement than that! Some of my neighbours told me they thought aliens had landed in my garden one night when I was doing beamshots. They seemed disappointed when I told them it was just some of my lights
Depends on the led as far as electrically isolating. The crees and the SST don't have to be isolated, at least on the center strip that the die is on. That's where the majority of the heat comes from. I think I remember that the SSC leds had to be isolated on the die plate. As far as the +/-lead strips, I don't worry about heat dissipation from them. Maybe I should, but I don't think it's needed.
There are a few cree XML leds out there, on a copper star, but not too many.
LOL! I love that last bit about shining it on the garage door! I once did something similar with a SST-50 in a homemade reflector driven at 7 amps.
My house is painted with chalk - I think you call that whitewashed in the states - and I just walked 10 meters from the wall and pointed the light directly at the wall, think 90° directly, and turned that thing on! I nearly dropped the thing. Ha ha. Still makes me laugh when i think about it!
Anyway I just wanted to say that so far it sounds really impressive. I've never had me one of those SST-90's to work with so I've no idea about how nice they are. But I sure do look forward to some beamshots.
), but in order to keep the current stable I had to heat sink both sides of the boards to fairly thick aluminum plates that touched the outer light wall. Without this the current will fall like a rock around the 1 minute mark till it bottoms out at like only a couple of amps.
