Sale Thread: New Inexpensive 18650/cr123a Host Kit - $25 Shipped!

[midias]

Awesome looks like I am still on the list let me know when you need pay.

 

[Helios]

Aaaargh! The postman dropped a notice for a package from some dude named "Brad".
Can't wait to pick it up tomorrow!:drool:

Wow international shipping takes awhile! Glad you wanted to pick up that damaged heatsink before it and my time invested in it ended up in the trash!

 

[anselm]

Picked it up today!
Yeah, I've got my work cut out for me, the driver tray is only about 1/3 finished.
But I knew that and I got it at a discount. Which is incredible because the original
price was very low to begin with.

It's a very nice and sturdy host, perfect fit for a 1x18650 or 2x16340, and the heatsink's
size suggests its good for any kind of high powered diode.

I'm a happy camper and lucky to have gotten the last one, apparently!:wave:

 

[rhd]

Got mine yesterday - awesome heatsink! This is just about the DIY driver-lover's dream heatsink.

A little arctic silver, and my 1084 has a perfect point to adhere right to the heatsink assembly. Built the entire driver on top of that, and enclose it with kapton tape mostly to avoid any potential contact with the host walls. (it's tidier than it looks in my photos).

It's wired for 1,800 mA, and can deliver around 1,500 with two cheap TrustFire 16340s. Once my AW's arrive, I would imagine the full 1,800 will be no problem. Even as-is, I'm seeing 1.5W from a normal aixiz glass lens

 

[midias]

^ Awesome looking personally I think mine will become a 12x 405nm

 

[rhd]

Alright... this is happening... this is the host that I'm going to murder a 445 in - in the name of testing the 1084's stability
* For anyone reading this after the fact - this is NOT how you should treat your lasers - this is an educational exercise for me

At the 3 minute mark, and counting...
*warm to the touch*

At the 6 minute mark.
- Host is getting warmer. Still quite comfortable to touch.
- LPM graph is scrolling by in a slow gradual healthy decline from the peak. Nothing unstable at this point. Diode, driver, and host all doing well.

At the 8 minute mark.
- Host is getting warm-hot. At the point that you might want to stop thinking about holding it if you were (I'm not).

At the 10 minute mark.
- Host is hot towards the head, and warm-hot all throughout. The host is either distributing the heat well, or my batteries are getting toasty themselves (or both).

At the 12 minute mark.
- Host is hot to the point that I want to circulate some air over it. Interestingly, the power has dropped at a perfectly gradual rate. Though it is half its peak, no sharp changes at all - nothing to suggest sharp failures.

At the 15 minute mark.
- I'm saving this diode's life Lasted far longer than I expected, so it gets spared. The power had dropped to less than half peak. I'm interested now to see whether a cool-down with fresh batteries will allow the diode to hit its peak again. Publishing LPM details shortly!

Very impressed with this host. Admittedly, I've never tried to murder a diode intentionally, but I didn't expect this to hold up for over 15 minutes.

And after 5 minutes to cool down - second run hits a peak of slightly lower (1,550mW or so) which is surely the result of discharging half the battery capacity.

Second Run - just to test peak

So Helios - impressive host

 

[anselm]

this is happening... this is the host that I'm going to murder a 445 in
....yaddayadda....
I'm saving this diode's life

Awww you disappoint me. You promised drama, I was expecting to see blood and guts.

So what do you think brought on the power decrease, the batteries would be my guess.
Or maye driver and/or diode overheating.
Anxious to see you try with better batteries.

Maybe you could rig some wires to read the batteries voltage during the run?
Voltage sensing wires could be really tiny, maybe through an open tailcap, the batterie negative somehow bridged to the host....

 

[Helios]

haha thats awesome! Ive left mine on for 10 minutes before running at 1500mA and it was hot but did not damage the diode output after recharging the bats and another lpm run.

I cant say I designed them for that kind of abuse but it looks like they can take it!


Also those DIY drivers tend to run a tad hotter than the microflex as far as I have experienced.....I wonder if that might have shortened the runtime a bit.

 

[rhd]

Also those DIY drivers tend to run a tad hotter than the microflex as far as I have experienced.....I wonder if that might have shortened the runtime a bit.

Perhaps. Although it's hard to think of that as a short runtime!

I'm going to grab an AC power source and reply the test in the future.

But if you think about it, two batteries supplying ~7.5 to 8 V is a constant 3 to 3.5 extra volts for the IC. That's ~5W for the IC to dissipate as heat. Almost as much as the diode itself has to dissipate!

 

[Helios]

But if you think about it, two batteries supplying ~7.5 to 8 V is a constant 3 to 3.5 extra volts for the IC. That's ~5W for the IC to dissipate as heat. Almost as much as the diode itself has to dissipate!

Thats what I was referring to. Still though not a short run time by any means! I would just be curious to see if the flex would run significantly longer and cooler. In reality there is no practical application for a handheld laser with a 10 minute constant duty :crackup: so its more than enough.

 

[Helios]

Wait rhd how did you get the LM317 to supply 1800mA? I thought they were 1500mA max

 

[rhd]

It's an LM1084 - can handle up to 5A, with a total circuit dropout of somewhere around 2V

 

[steviedezie]

So Helios when are you thinking of supplying more of these.

 

[rhd]

Helios - if nothing else, you should design heatsinks. I absolutely love that heatsink design. It is bar-none my favorite build style. I wish more people integrated driver trays like that. It's just - brilliant - especially for those of us who have a love affair for building with inefficient, heat-producing, DIY driver ICs

 

[steviedezie]

hey RHD the lm1084 REQUIRE 10mF caps on both input AND output correct? I remember looking at the datasheet and it required at least 10mF on both sides to even work at all.

 

[rhd]

Yep, you've got it