Nichia 3.5w oops :( *pics*

[IWIRE]

Not sure what happened, I thought you could parallel boost drives. I took two SXB's set each one to 2.2a. put them together and was getting 4.4a. Assembled everything it powered right up. Was checking the output on my lpm and when it hit 5w it shut down. I thought the clicky switch had died.
Switch tested good, battery was good. The top driver smells smoked. Disassembled everything and the diode lived
Oh well back to the drawing board. I even post pics of my failures.











The copper hosts I ordered from sinner showed up today. I was wishing I had saved the diode for one of them. Guess I get to try again

 

[ryansoh3]

Close call, I'm glad you didn't fry the diode! :beer:

 

[Seoul_lasers]

Wow, indeed a close call.

I wonder what smoked your driver? Some of those Thermal pastes actually can be slightly conductive/ or even slightly capacitive especially the ones with Silver, and Carbon allotropes added. If some of that gets on the pins for the current regulator it could have caused the driver to malfunction and short out.

I experienced a similar situation with a Die4Drivers v1.3 a few years ago (2010) with some arctic 2MX. Fortunately the driver was fixable. The diode also survived.

What did you use as your TIM?


Thank you for sharing your experience!

 

[IWIRE]

Thanks. I sent Angelos a PM. I'll post a pic of the thermal adhesive I used. I've used it a lot with no problems. I even ohm everything out just to check before I hook it up. I did something wrong just don't know what.
I share my experiences good or bad. Not the first or last time I'll look like an idiot. Here hold my beer and watch this.
Thanks for the ideas, I appreciate the help.
Dave

 

[petters]

Don't toy with our emotions like that, haha thought from your title the diode died.

 

[trencheel303]

Nice photography IWIRE, sorry to hear it popped. At least the diode lives though, that'd be a real sore loss.

 

[COMMANDR]

At least the diode survived to fight (shine) another day.
It looks like you have used a electrically conductive thermal paste / adhesive.???

This stuff works good and is not to hard on the wallet.

Sales price::$7.95

Arctic Alumina Ceramic Thermal Adhesive Epoxy w/ Application Wand - 5g

Product Description

Arctic Alumina Epoxy
Two syringe tubes
2.5 grams of Part A and 2.5 grams of Part B
Adhesive allows you to bond two materials together
Layered composite of aluminum oxide and boron nitride.
Pure electrical insulator, neither electrically conductive nor capacitive.
Plastic reusable

Arctic Alumina Ceramic Thermal Adhesive Epoxy w/ Application Wand - 5g - Thermal Compound / Grease

 

[IWIRE]

Your emotions ! You should have been holding it when it hit 5w and shut off !
I'm hopeful I didn't damage the diode, time will tell, I know it still outputs the same wattage at the same current as it did before. Fingers crossed.
That's what I used Arctic ceramic adhesive. I've never found it to be conductive. Now I think the Arctic silver I used on the module may be conductive.
I've found info where boost drives have been paralleled before, just not on the specific SXB. I have two more and I'm crazy enough to do it again. I`m going to look real close at it tonight, see if I find the cause. I'm hoping to hear that SXB can be paralleled and that it was something I did that caused it.
It's weird, during testing the drives were not getting hot at all. I dunno. I did something wrong. I just have to find what that something is.

I separated the drivers and inspected all the components. I saw nothing wrong with any of the components. I took the top drive which is the one that smelled fried and hooked it up to my test load. It works and was still set at 2.2a. I checked the second drive, same result. Both drives still work. I checked my battery again, still tests good. Checked my clicky switch again. still tests good. WTF ! All that's left is the battery contact board. I put my test lead down in the housing on the spring and put the other one on the back side of the board. Over 500 ohms resistance. It's the only thing I have found wrong. Gotta go buy snap-ring pliers to get it out.
Lets see, it was pulling 6a. 500 ohms resistance through battery contact. 6X6=36 36x500=18000 . 18000 watts of heat. No wonder it smelled fried. I sure hope it is really this simple. Sure would explain why the top driver smelled fried, it's the closest to the battery contact board.

 

[Moe Ron]

Greetings,

Glad to hear the diode survived and was just a driver. I have been running mine with 2.5a, at least till I get the 4a ones am expecting, and getting 3.5w. Now for some unknown reason it only outputs 2.7w. This is of course using a 3 element lens. Have an idea what it is, and may have to decan it at some point.
Just means I'll have to work a little more overtime I guess.


cheers,
Moe

 

[IWIRE]

Thanks Moe Ron ! Appreciate it. You sound like me, my sinner host's showed up so I'm working this weekend so I can order diodes from DTR. I updated the post above yours, not completely sure at this point but it looks like it's not a driver problem. I'll update what I find. I'm gonna figure it out.

 

[djQUAN]

500 ohms resistance through battery contact. 6X6=36 36x500=18000 . 18000 watts of heat.

I'm not sure if you're just joking but that's not how it works.

6 amps into 500ohms would require 3000volts. I don't know where you found a 3000V 18650 battery.

4.2V (fully charged Li-ion) into 500ohms will just do a measly 35mW. I'm thinking more on the high current draw burned off a part of the contact board and made it high resistance afterwards.

 

[Seoul_lasers]

Your emotions ! You should have been holding it when it hit 5w and shut off !
I'm hopeful I didn't damage the diode, time will tell, I know it still outputs the same wattage at the same current as it did before. Fingers crossed.
That's what I used Arctic ceramic adhesive. I've never found it to be conductive. Now I think the Arctic silver I used on the module may be conductive.
I've found info where boost drives have been paralleled before, just not on the specific SXB. I have two more and I'm crazy enough to do it again. I`m going to look real close at it tonight, see if I find the cause. I'm hoping to hear that SXB can be paralleled and that it was something I did that caused it.
It's weird, during testing the drives were not getting hot at all. I dunno. I did something wrong. I just have to find what that something is.

I separated the drivers and inspected all the components. I saw nothing wrong with any of the components. I took the top drive which is the one that smelled fried and hooked it up to my test load. It works and was still set at 2.2a. I checked the second drive, same result. Both drives still work. I checked my battery again, still tests good. Checked my clicky switch again. still tests good. WTF ! All that's left is the battery contact board. I put my test lead down in the housing on the spring and put the other one on the back side of the board. Over 500 ohms resistance. It's the only thing I have found wrong. Gotta go buy snap-ring pliers to get it out.
Lets see, it was pulling 6a. 500 ohms resistance through battery contact. 6X6=36 36x500=18000 . 18000 watts of heat. No wonder it smelled fried. I sure hope it is really this simple. Sure would explain why the top driver smelled fried, it's the closest to the battery contact board.

hmm... well then it's not the TIM that's the problem as that would have blown the driver's IC chip (it's happened before).
Your drivers all checked out.

Now that you've explained a bit ... sounds like you have a dead short across the case and the positive contact.
Please, please.... check the contact board to make sure you've soldered to it correctly!
This is quite a dangerous situation with Li-ion cells. They can get hot in the situation enough to rupture explosively. Sort of like (similarly to) what occurred in the link below!
Check link below...

CandlePowerForums

 

[IWIRE]

I'm thinking more on the high current draw burned off a part of the contact board and made it high resistance afterwards.

Ding ding ding ! We have a winner ! While I was out getting snap ring pliers to take it apart, you already figured out what happened ! Rep +1 for you :beer:

I just made my guesstement of heat from this chart. I know ever how many watts it takes to make smoke, that's what it was It didn't work no more after the magic smoke escaped

I have no idea what this crap in here is. But it's caused me problems so it is gone ! What is this crap and what is it doing in my contact ! It's nice to find what it was. I think the diode should be fine.



If your going to use this host for anything besides a paperweight. I'd recommend getting rid of the cheap wire and whatever the rest of that crap is.
Now to put it back together. I get twice the fun out of one build ! :crackup:
Thanks everybody for your brainstorming and helping me figure this out ! I appreciate it a lot ! :beer:

 

[Seoul_lasers]

hmm... well then it's not the TIM that's the problem as that would have blown the driver's IC chip (it's happened before).
Your drivers all checked out.

Now that you've explained a bit ... sounds like you have a dead short across the case and the positive contact.
Please, please.... check the contact board to make sure you've soldered to it correctly!
This is quite a dangerous situation with Li-ion cells. They can get hot in the situation enough to rupture explosively. Sort of like (similarly to) what occurred in the link below!
Check link below...

CandlePowerForums

( edit)

The pic of the contact looks fine. The wiring doesn't look like it was able to handle the high current and became a heating element.

 

[IWIRE]

( edit)

The pic of the contact looks fine. The wiring doesn't look like it was able to handle the high current and became a heating element.

I agree the contact boards and spring are great. The wire just melted down. The wire I built it with I got from Flaminpyro didn't melt down. I can't figure out the purpose of the funky little black box that has leads from negative and positive both going to it. I'll dissect it tomorrow night and find out whats in it. I'm going to drill a small hole in the contact boards and run my wires straight to the battery spring to driver and straight from ground to driver. Screw this contact board/crap/contact board setup. The fewer connections the less chance for problems.
My Grandfather told my Father to tell his son "Keep it simple stupid"

 

[Seoul_lasers]

I agree the contact boards and spring are great. The wire just melted down. The wire I built it with I got from Flaminpyro didn't melt down. I can't figure out the purpose of the funky little black box that has leads from negative and positive both going to it. I'll dissect it tomorrow night and find out whats in it. I'm going to drill a small hole in the contact boards and run my wires straight to the battery spring to driver and straight from ground to driver. Screw this contact board/crap/contact board setup. The fewer connections the less chance for problems.
My Grandfather told my Father to tell his son "Keep it simple stupid"

Actually I'll be honest, the black box looks a lot like a voltage input for a wall wart.
I wonder if it also could be a reverse polarity protection circuit? Mysterious.
Also, it does appear you have had a dead short situation. The two wires from your contact board also appear to be PVC. :tsk:
Always... especially for high current builds use Silicone and larger dia ga wiring. 20ga might be safer as it is rated for 12A -- a little overkill but higher overhead is better.
The heat from the 2 drivers could get hot enough to weaken PVC wire to the point of failure.. hence why Silicone wire is used in RC cars...planes..
Anyways you seem to be on a roll to fix the situation. Good thing nothing further happened.


Cheers... keep us posted..

Thanks for sharing.