In case anyone’s wondering why I drilled the outer two holes in the base of the copper heat sink when the center hole is used, it’s to give torsional strength to the glued joint. I align the ~4mm deep holes with the original wires holes in the emitter shelf of the head so the glue has “posts” to it, really limits the ability for the whole thing to slip and twist wires.
I love using the Arctic Alumina Thermal Adhesive as it sets up in minutes, it’s stuck in about 5 minutes and pretty much cured in 12. Another thing I like about it is that if an mcpcb just has to be removed you can heat up the board with the soldering iron and break the epoxy joint loose. Can’t do that with Arctic Silver or JB Weld, once in with those it’s just gonna stay there pretty much.
So you’re saying I can remove mpcb’s that I have laid out on an aluminum strip with enough heat? I have a bunch of Nichias on Noctigons that I would love to recover if possible.
—
If you don’t like what it looks like on the wall, take it outside.
With Arctic Alumina Thermal Adhesive I have found that to be true. I typically put the soldering iron on the edge of the copper MCPCB to heat it up, the adhesive will emit an odor when it gets hot, I usually use a flat blade jewelers screwdriver and twist it to apply lateral pressure or if one of the wire holes shows the bottom of the MCPCB I will put the screwdriver against the bottom of the board and give the screwdriver a sharp rap… I actually keep a 1” wood chisel on my bench to use as a hammer for this type of situation.
I have removed glued boards many times this way.
Using Arctic Silver, or JB Weld, not gonna happen. That stuff is resilient!
Ok that’s great news, thank you Dale. I just checked and have used Arctic Alumina two part adhesive. I’ll try your method and see what I can do. Do you think the noctigons will still be useful after?
—
If you don’t like what it looks like on the wall, take it outside.
Convoy s2+ w/ 2 channel red / white triple. LEDs: 2x Nichia 219C 5000k, 1x Osram Olsen SSL80 deep red (657nm). Do note this is a 3030 LED and is VERYDIFFICULT to get it to float centered on XP pads during reflow)
Driver: hacked 105c
FW: BLF-A6, surprisingly [to me] not that modified at all.
Cut one trace on thr MTNMCPCB is all it takes to make the one individually addressable
Isolate a 7135
That soldering was easy, you can’t see the tiny wire from pin 5 to the single 7135 leg.
It needs some work still, I forgot to make the red channel come on during hidden turbo and I want to mod the levels some, maybe add a red/white strobe back in. Definitely gotta
make the 6 mode group have a little higher R1 than it does now then crank R2 to max for a low/high red mode.
—
RIP SPC Joey Riley, KIA 11/24/14. Now I am become death, the destroyer of worlds.
Thanks guys.
Wasn’t gonna post this one but gotta live and learn. Plus if at first you don’t succeed; spend more, spend more money. Or something like that I think.
Thankfully the 2 nichia bit it and the SSL80 was fine, it was the only one, the other Olsen I had was standard red and the deep red was one of my main wants in this light, otherwise I’d have just used a XP-E2 and not dealt with the crazy soldering.
—
RIP SPC Joey Riley, KIA 11/24/14. Now I am become death, the destroyer of worlds.
Modded my cheapo Rayovac 3 AAA headlamp. I don’t use a headlamp that often, but when I do, it’s to work on something around the house in order to keep both hands free. Although I would love to get a nice ZebraLight floody headlamp, I can’t justify it since I wouldn’t use it much. The tint of the Rayovac headlamp isn’t bad for a cool white LED, and I like that it just has high and medium modes (no strobes).
The headlamp is also plenty bright, but I don’t like the narrowness of the beam. I’m constantly moving the headlamp up and down in order to get it in the position that I need. So I decided to replace the lens.
Taking it apart was very easy, just 3 screws holding it together. There are 2 O-rings, 1 for the battery cover and the other for the plastic lens. There was some type of thermal compound/adhesive on the back of the aluminum MCPCB, but my guess it was to hold the LED in place for assembly rather than for thermodynamics.
I thought about trying to make the Rayovac lens frosted, but it had a raised rim on both sides. So instead, I created a lens from some semi-transparent plastic I had. I used sandpaper to make 1 side frosted.
The final result turned out great. The headlamp now produces a much more useful flooded light.
—
I’d rather use my flashlight around the house than turn on the lights.
I solved the last of the problems with my first flashlight rebuild.
S2+ with TIR, shaved SST-40 and BLF-A6 driver.
The problem was soldering the driver to that damn pill.
I used a blowtorch this time and it worked.
Though I overheated the spring so it stopped being springy. And fell off.
Oh well, doesn’t matter for now, I soldered it back on.
I turn it on….nothing happens.
If I bypass the driver and connect a battery directly to the leads – ouch, my eyes hurt. So it must be the driver.
I tried to remove the pill and connect battery to it directly. I touched the pill and the driver with wires, saw a blink. Weird…
Back to the host, it’s still dead.
Again, checking the pill. When I connected it I saw some smoke from the point on the spring where I touched it with the wire.
You might have heated the driver so hot with the torch that other things flowed and shorted it out, or you may have just damaged components with the sheer amount of heat applied. Blowtorch and the fine components of a flashlight driver are a risky and tricky combination.
You might have heated the driver so hot with the torch that other things flowed and shorted it out, or you may have just damaged components with the sheer amount of heat applied. Blowtorch and the fine components of a flashlight driver are a risky and tricky combination.
I have demolished several drivers trying to solder them to a pill with a blowtorch. After I bought a 80W (Antex) solder iron (also good for soldering wires to DTP boards inside cavities) my problems were over.
Been working on the TN42 XHP70.2 P2-1A. (crash test dummy) I think I got most of the resistance out of it. Tested on a T/A’s Lumen Tube and calibrated with maukka’s lights.
4xVTC5D’s at 4.07volts. 8,840lms @ turn on.
4xVTC5D’s right off the charger, 4.21volts
9,230 lumen at turn on. I need more/better/ bigger heat sinking!!!
I SWEAR I seen just over 11,000lm Blip off the meter for a Nano Second!
Nice KawiBoy. Looks like you are losing some lumens in your measurement around the bezel. If you can lower the light a bit deeper into the tube, you will probably get a better measurement. I think I read TA once saying that he lowers the light until he gets the highest reading. I do the same thing now.
Omg, with the current FW offerings this old board is my new favorite 7135 driver!
Can be populated from a 105c and a otc.
Dual channel 1x / 7× 7135. Runs popular FW including star moonlight special and (wait for it) BLF-A6 without mods.
Fits in Convoy S2+ pills, with non filed retaining ring!
—
RIP SPC Joey Riley, KIA 11/24/14. Now I am become death, the destroyer of worlds.
Nice KawiBoy. Looks like you are losing some lumens in your measurement around the bezel. If you can lower the light a bit deeper into the tube, you will probably get a better measurement. I think I read TA once saying that he lowers the light until he gets the highest reading. I do the same thing now.
Yes, sitting on top of the ring is not ideal, and even with a tinfoil hat around it, and set inside I gained 400lm. It was mostly and experiment. I used some heavy solid core copper wire we use on our machines that I found laying around in the shop, and hand pounded and bent many pieces to connect the mcpcb to the Lexel T/A Infineon FETDD TN42 driver, to see if/what I would get. Well the pipe fitters have a term/word for it “Bones” many bones were made till I got the right length and bends in the leads to be able to solder the leads to the board or driver with out breaking the solder joints or ripping the contacts off either, and bolt the TN42 style driver up to the light. This heavy solid core wire is not manageable to say the least. I wasted a good brand new massive SD75 XHP70 board in the process. I just wished I would of done this in a light that had better sinking and fins on it. The TN42 is not the best choice, as in seconds this thing gets amazingly hot, the GT or SD75 would have been a better host, really stupid in the end on my choice with all the effort but it was a rainy weekend?.
The carrier springs had 18awg Normal wire (silicone) bypass’s and the center positive contact was soldered directly to the carrier mcpcb, brass screws connecting the stanchions to the boards, with 4xVTC5D’s 2S2P it’s a potent power plant.
I am building an integrated sphere and have a problem.
My Luxmeter has a maximum value of 200 kLux, but my Imalent DX80 emitts more light.
Does anyone know how to mod a luxmeter, that it shows lower values. Can I add a resistor? How does it work? Has the sensor AC or DC Current?
Diffuse the light. That will take some balancing to get it right but it will be easier and more reliable than tinkering with your meter.
Thank you, that was my second idea, but my favorite was to change something in my meter.
Now I added a DN25 electric installation tube with 6cm length, to put the sensor to the outside of the sphere. It lowers the readings by factor 10.
No 30.000 Lumen eqals around 50.000 Lux. This means my sphere can read around 100-120 kLumen. That may rest around 2 years to get such a flashlight, I think.
Measure a lower mode in the sphere and calculate the relative difference between that and higher modes in some other way (ceiling bounce for example). Or put a diffuser/neutral density filter in front of the lux meter. Or what you just did
Measure a lower mode in the sphere and calculate the relative difference between that and higher modes in some other way (ceiling bounce for example). Or put a diffuser/neutral density filter in front of the lux meter. Or what you just did
I have alread a Lumen-Box, but I noticed that the readings for over 10-15 k Lumen are too low. So Now I make further readings in the Toilette. But I wanted to get better readings for all output levels of a flashlight. For my Imalent DX80 and others like Olight X9R, Acebeam X80 and upcoming flashlights. Thats why I build the new integrated sphere.
And with diffusing the light before it hits the luxmeter, I will get good readings, I hope.
Measure a lower mode in the sphere and calculate the relative difference between that and higher modes in some other way (ceiling bounce for example). Or put a diffuser/neutral density filter in front of the lux meter. Or what you just did
I have alread a Lumen-Box, but I noticed that the readings for over 10-15 k Lumen are too low. So Now I make further readings in the Toilette. But I wanted to get better readings for all output levels of a flashlight. For my Imalent DX80 and others like Olight X9R, Acebeam X80 and upcoming flashlights. Thats why I build the new integrated sphere.
And with diffusing the light before it hits the luxmeter, I will get good readings, I hope.
If you want a very neat way to diffuse the light you can use a satellite sphere (which is indeed called a diffuser in the literature) where the luxmeter sensor used to be, and put the sensor in there (you even gain some integration). The sensitivity decrease can be set by the size of the connection hole: the inner surface area of the satellite sphere divided by the surface area of the connection hole gives you the decrease of the sensitivity.
So you’re saying I can remove mpcb’s that I have laid out on an aluminum strip with enough heat? I have a bunch of Nichias on Noctigons that I would love to recover if possible.
If you don’t like what it looks like on the wall, take it outside.
With Arctic Alumina Thermal Adhesive I have found that to be true. I typically put the soldering iron on the edge of the copper MCPCB to heat it up, the adhesive will emit an odor when it gets hot, I usually use a flat blade jewelers screwdriver and twist it to apply lateral pressure or if one of the wire holes shows the bottom of the MCPCB I will put the screwdriver against the bottom of the board and give the screwdriver a sharp rap… I actually keep a 1” wood chisel on my bench to use as a hammer for this type of situation.
I have removed glued boards many times this way.
Using Arctic Silver, or JB Weld, not gonna happen. That stuff is resilient!
Ok that’s great news, thank you Dale. I just checked and have used Arctic Alumina two part adhesive. I’ll try your method and see what I can do. Do you think the noctigons will still be useful after?
If you don’t like what it looks like on the wall, take it outside.
Finally got a day off and put it to good use!
Convoy s2+ w/ 2 channel red / white triple.
LEDs: 2x Nichia 219C 5000k, 1x Osram Olsen SSL80 deep red (657nm). Do note this is a 3030 LED and is VERY DIFFICULT to get it to float centered on XP pads during reflow)
Driver: hacked 105c
FW: BLF-A6, surprisingly [to me] not that modified at all.
Cut one trace on thr MTN MCPCB is all it takes to make the one individually addressable
Isolate a 7135
That soldering was easy, you can’t see the tiny wire from pin 5 to the single 7135 leg.
It needs some work still, I forgot to make the red channel come on during hidden turbo and I want to mod the levels some, maybe add a red/white strobe back in. Definitely gotta make the 6 mode group have a little higher R1 than it does now then crank R2 to max for a low/high red mode.
RIP SPC Joey Riley, KIA 11/24/14. Now I am become death, the destroyer of worlds.
Yep.
djozz quotes, "it came with chinese lettering that is chinese to me".
"My man mousehole needs one too"
old4570 said "I'm not an expert , so don't suffer from any such technical restrictions".
Old-Lumens. Highly admired and cherished member of Budget Light Forum. 11.5.2011 - 20.12.16. RIP.
Very nice CK
Head only build -- Vape conversion #4 -- UF2100 Makeover -- Ti-mo -- 6th O-L comp entry -- Titanium Build -- Hunting Light -- 21700 Tube Light -- Stripey mtg2 Light -- UV Light -- Colour V4A Light -- Copper 26650 Light -- Quad xhp35 Light -- Finger Sizzler -- Mechanical Mod Light
Thanks guys.
Wasn’t gonna post this one but gotta live and learn. Plus if at first you don’t succeed; spend more, spend more money. Or something like that I think.
Thankfully the 2 nichia bit it and the SSL80 was fine, it was the only one, the other Olsen I had was standard red and the deep red was one of my main wants in this light, otherwise I’d have just used a XP-E2 and not dealt with the crazy soldering.
RIP SPC Joey Riley, KIA 11/24/14. Now I am become death, the destroyer of worlds.
Just for fun
[REVIEWS] AMUTORCH: S3 / S3 vs 219c / AM30 / AX1 / VG10 /// SOFIRN: SF14 + SP10A / SP32A / SP10B /// NITEFOX: UT20 / ES10K / K3 /// ODEPRO: KL52 / B108 /// ACEBEAM: H20 / TK16 /// BLITZWOLF: BW-ET1 /// DQG: AA Slim Ti /// HC-LIGHTS: SS AAA /// XTAR: PB2 Charger /// OLIGHT: M2R Warrior /// WUBEN: TO10R / E05 / T70 / E10 /// ON THE ROAD: M1 / i3 / M3 Pro /// ROVYVON: A2 + A5R / E300S / A8 /// KLARUS: XT1C /// LUMINTOP: Tool AA V2.0 + Tool 25 /// LIVARNOLUX: 314791 /// SKILHUNT: M150
Tricks: 1 / 2 / 3 / 4 / 5 / 6 / 7 / 8 /// TIR Lenses: 1 / 2 /// Others: Biscotti 3 + 1*7135 / Triple TIR w/ XP-G2 ///// My Collection ///// My Review's Blog (PT)
GIVEAWAY: 1
Modded my cheapo Rayovac 3 AAA headlamp. I don’t use a headlamp that often, but when I do, it’s to work on something around the house in order to keep both hands free. Although I would love to get a nice ZebraLight floody headlamp, I can’t justify it since I wouldn’t use it much. The tint of the Rayovac headlamp isn’t bad for a cool white LED, and I like that it just has high and medium modes (no strobes).
The headlamp is also plenty bright, but I don’t like the narrowness of the beam. I’m constantly moving the headlamp up and down in order to get it in the position that I need. So I decided to replace the lens.
Taking it apart was very easy, just 3 screws holding it together. There are 2 O-rings, 1 for the battery cover and the other for the plastic lens. There was some type of thermal compound/adhesive on the back of the aluminum MCPCB, but my guess it was to hold the LED in place for assembly rather than for thermodynamics.
I thought about trying to make the Rayovac lens frosted, but it had a raised rim on both sides. So instead, I created a lens from some semi-transparent plastic I had. I used sandpaper to make 1 side frosted.
The final result turned out great. The headlamp now produces a much more useful flooded light.
I’d rather use my flashlight around the house than turn on the lights.
I solved the last of the problems with my first flashlight rebuild.
S2+ with TIR, shaved SST-40 and BLF-A6 driver.
The problem was soldering the driver to that damn pill.
I used a blowtorch this time and it worked.
Though I overheated the spring so it stopped being springy. And fell off.
Oh well, doesn’t matter for now, I soldered it back on.
I turn it on….nothing happens.
If I bypass the driver and connect a battery directly to the leads – ouch, my eyes hurt. So it must be the driver.
I tried to remove the pill and connect battery to it directly. I touched the pill and the driver with wires, saw a blink. Weird…
Back to the host, it’s still dead.
Again, checking the pill. When I connected it I saw some smoke from the point on the spring where I touched it with the wire.
Did I reverse polarity?
No.
OK, how should I troubleshoot it further?
You might have heated the driver so hot with the torch that other things flowed and shorted it out, or you may have just damaged components with the sheer amount of heat applied. Blowtorch and the fine components of a flashlight driver are a risky and tricky combination.
I have demolished several drivers trying to solder them to a pill with a blowtorch. After I bought a 80W (Antex) solder iron (also good for soldering wires to DTP boards inside cavities) my problems were over.
link to djozz tests
OK, so the best thing to do now is cut the driver off the pill, unsolder the leads and do a visual inspection. Right?
Perhaps a photo of this driver/pill assembly would help, maybe a few photo’s from different angles?
Next time, before you put in the driver, tin the rim of the pill with solder. It’ll be a lot easier to solder the driver to it.
.
Been working on the TN42 XHP70.2 P2-1A. (crash test dummy) I think I got most of the resistance out of it.
Tested on a T/A’s Lumen Tube and calibrated with maukka’s lights.
4xVTC5D’s at 4.07volts. 8,840lms @ turn on.4xVTC5D’s right off the charger, 4.21volts
9,230 lumen at turn on. I need more/better/ bigger heat sinking!!!
I SWEAR I seen just over 11,000lm Blip off the meter for a Nano Second!
KB1428 “Live Life WOT”
^
Nice KawiBoy. Looks like you are losing some lumens in your measurement around the bezel. If you can lower the light a bit deeper into the tube, you will probably get a better measurement. I think I read TA once saying that he lowers the light until he gets the highest reading. I do the same thing now.
Fun soldering tonight!
That’s how much slack I was working with. All 22AWG silicone
RIP SPC Joey Riley, KIA 11/24/14. Now I am become death, the destroyer of worlds.
Pretty cool! What driver is this?
That looks downright scary Sk. Nice work.
djozz quotes, "it came with chinese lettering that is chinese to me".
"My man mousehole needs one too"
old4570 said "I'm not an expert , so don't suffer from any such technical restrictions".
Old-Lumens. Highly admired and cherished member of Budget Light Forum. 11.5.2011 - 20.12.16. RIP.
That’d be mine. If you happened to know that and you’re asking for specifics it’s my v3.0 constant current 4Ch driver in 17mm.
RIP SPC Joey Riley, KIA 11/24/14. Now I am become death, the destroyer of worlds.
Omg, with the current FW offerings this old board is my new favorite 7135 driver!
Can be populated from a 105c and a otc.
Dual channel 1x / 7× 7135.
Runs popular FW including star moonlight special and (wait for it) BLF-A6 without mods.
Fits in Convoy S2+ pills, with non filed retaining ring!
RIP SPC Joey Riley, KIA 11/24/14. Now I am become death, the destroyer of worlds.
Awesome!
Yes, sitting on top of the ring is not ideal, and even with a tinfoil hat around it, and set inside I gained 400lm. It was mostly and experiment. I used some heavy solid core copper wire we use on our machines that I found laying around in the shop, and hand pounded and bent many pieces to connect the mcpcb to the Lexel T/A Infineon FETDD TN42 driver, to see if/what I would get. Well the pipe fitters have a term/word for it “Bones” many bones were made till I got the right length and bends in the leads to be able to solder the leads to the board or driver with out breaking the solder joints or ripping the contacts off either, and bolt the TN42 style driver up to the light. This heavy solid core wire is not manageable to say the least. I wasted a good brand new massive SD75 XHP70 board in the process. I just wished I would of done this in a light that had better sinking and fins on it. The TN42 is not the best choice, as in seconds this thing gets amazingly hot, the GT or SD75 would have been a better host, really stupid in the end on my choice with all the effort but it was a rainy weekend?.
The carrier springs had 18awg Normal wire (silicone) bypass’s and the center positive contact was soldered directly to the carrier mcpcb, brass screws connecting the stanchions to the boards, with 4xVTC5D’s 2S2P it’s a potent power plant.
KB1428 “Live Life WOT”
I am building an integrated sphere and have a problem.
My Luxmeter has a maximum value of 200 kLux, but my Imalent DX80 emitts more light.
Does anyone know how to mod a luxmeter, that it shows lower values. Can I add a resistor? How does it work? Has the sensor AC or DC Current?
Its a UNI-T UT383 BT.
Diffuse the light. That will take some balancing to get it right but it will be easier and more reliable than tinkering with your meter.
Thank you, that was my second idea, but my favorite was to change something in my meter.
Now I added a DN25 electric installation tube with 6cm length, to put the sensor to the outside of the sphere. It lowers the readings by factor 10.
No 30.000 Lumen eqals around 50.000 Lux. This means my sphere can read around 100-120 kLumen. That may rest around 2 years to get such a flashlight, I think.
Measure a lower mode in the sphere and calculate the relative difference between that and higher modes in some other way (ceiling bounce for example). Or put a diffuser/neutral density filter in front of the lux meter. Or what you just did
I have alread a Lumen-Box, but I noticed that the readings for over 10-15 k Lumen are too low. So Now I make further readings in the Toilette. But I wanted to get better readings for all output levels of a flashlight. For my Imalent DX80 and others like Olight X9R, Acebeam X80 and upcoming flashlights. Thats why I build the new integrated sphere.
And with diffusing the light before it hits the luxmeter, I will get good readings, I hope.
If you want a very neat way to diffuse the light you can use a satellite sphere (which is indeed called a diffuser in the literature) where the luxmeter sensor used to be, and put the sensor in there (you even gain some integration). The sensitivity decrease can be set by the size of the connection hole: the inner surface area of the satellite sphere divided by the surface area of the connection hole gives you the decrease of the sensitivity.
(I made one sphere with diffusing satellite sphere: http://budgetlightforum.com/node/41217)
link to djozz tests
Pages