Promising Looking Buck Driver (17mm and 22mm)

[ImA4Wheelr]

I don't know if this driver meets the demands needed for some of the bigger (not biggest) diodes these days, but it seems promising to me. The designer is among the most knowledgeable members on BLF. It's been used to drive LED's in the mid 6 amp territory. He hasn't tried to determine if the driver is laser diode friendly. I've been sitting on a couple diodes (9mm 445 and a 1W 405) for some time now and I'm playing with the idea of using this driver for the builds. Especially since it's only 17mm and the Attiny MCU is easy to flash.

17mm MTN-MAXlp HP 4A-5.5A Low-Profile Buck Driver - 5V-15V

He also sells just the PCB's and/or components. He has designed a new 4 layer version of the 17mm that enhances the thermal path to pull heat away from the driver. He's an open sources kind of guy and lists all the components needed should one wish to acquire and build these drivers on their own. He has PCB's shared on OSH Park.

Here is a discussion thread on the driver:

17mm/22mm MTN-MAXlp - Low-Profile 1A-4A Programmable Buck Driver + 22mm MTN-MAX | BudgetLightForum.com

The below pics are examples of a completed driver done by RMM.

 

[ImA4Wheelr]

Ordered a set of three 17mm 4-layer boards because I should already have most of the components to build them. I may need to order the inductors and fly-back diodes. Need to scrounge though my junk, um, I mean spare parts inventory for those 2 items first.

 

[Benm]

Seems interesting, but the 10% output ripple would mean that you should probably not run the laser diode at absolute maximum power. Failure mode for LEDs is usually thermal, while for laser diodes it is optical (though thermal is obviously possible if not properly heatsinked).

If you were to chose something with modes, note that this is a PWM system and i'm not really sure how fast the PWM is (could probably be adjusted in software). If it uses the builtin PWM of the attiny it will be very slow (<1 kHz) and you will see a on-off pattern if you move the dot along a wall or something similar.

For the size you can't get much better than this though - i wouldn't try to hand solder it

 

[ImA4Wheelr]

Thank you Benm. Really appreciate that input.

I was wondering how much I should compensate for the 10% (5% from center both ways) current ripple. Sounds like I should be cautious and back down more than 5% from what I think the applicable diode can take.

Plan on going 19kHz using fast phase (as opposed to 9.6kHz phase correct with fuses 0x75/0xff). I don't know if it matters, but fast seems like it would be gentler.

The small size is definitely the primary factor attracting me to this buck driver and I agree about reflowing. Not sure you could solder the MMU package. Never tried.

Edit: fixed error

 

[Benm]

I've found it pretty much impossible to hand solder such packages with an iron. You can do it for test setups with the chip upside down connecting thin wires between the pads and the board, that that would give no thermal bonding at all.

Solder paste and an oven are the only realistic option to do this yourself, but even that can be rather tricky.

Then again you could use 50 mil spaced components in the space provided if you give up on the microcontroller which i doubt you'd use that much driving a laser.

 

[ImA4Wheelr]

I'm with you on hand soldering, but I use cheap hot air guns for all my reflowing.

My favorite is the Harbor Freight hot air gun that goes on sale for $9 periodically. It is very capable of reflowing everything from drivers to large copper parts such as DTP emitter bases to integrated copper shelves in large flashlight heads. I have also used mine to roll fenders and do other auto body work.

Posted pictures of a completed driver (removed the blank driver board pics) in the OP>

 

[Benm]

Looks like a good soldering result to me!

Did you glue the parts down or just let the solder do its work? I wondering since the board is populated on both sides - guess everything would just cling on if you did the side with the inductor last.

 

[RMM]

Do you guys actually want or need modes? Are most laser diode drivers constant current, or constant voltage?

 

[RMM]

Looks like a good soldering result to me!

Did you glue the parts down or just let the solder do its work? I wondering since the board is populated on both sides - guess everything would just cling on if you did the side with the inductor last.

I do the inductor side first then reflow the battery side last, but you could do it the other way also.

 

[Pi R Squared]

Do you guys actually want or need modes? Are most laser diode drivers constant current, or constant voltage?

All laser diode drivers are constant current.

Alan

 

[ImA4Wheelr]

The pictures in the OP are not mine. Those are RMM's that I grabbed from the thread I linked to in the OP. I wish I could make my drivers look like that. I'll add a note to the OP.

Welcome to the forum RMM

 

[Benm]

I do the inductor side first then reflow the battery side last, but you could do it the other way also.

Wonder how far surface tension in the solder goes there - that inductor looks heavy enough to simple drop off when reflowing with that pointing down. Then again, i may be surprised

 

[ImA4Wheelr]

^ Same here. That inductor looks pretty heavy in the pics, but then that is only a 17mm driver and the inductor pads are large. I would want to reflow the bottom first, but RMM probably has had weeks that he has reflowed more than I have in my whole life.

I've reflowed a lot of double sided stuff with no problems as long as the board is level and I don't bump or jar it. Reflowing 3 dimensional drivers is a real pain. With mixed success, I've had to use aluminum foil to try to keep the vertical board cool enough to not liquefy the solder.

I've gone through all my stuff. I only need to acquire a few components (Inductor, Flyback Diode, and Buck Converter). I ordered all but the inductors yesterday. Trying to decide if I want to go Coil Craft or external (not smd). I'm not concerned about the driver being low profile for my purposes, but I am concerned about ripple and that particular shielded inductor sounds like the best way to go. I don't see through holes in the PCB for an external inductor. Those large pads, however, could be easy to rig a stable mount.

 

[RMM]

As long as you don't heat the other side too long you won't melt the solder holding the inductor since it has a lot of thermal mass.

I've gone through all my stuff. I only need to acquire a few components (Inductor, Flyback Diode, and Buck Converter). I ordered all but the inductors yesterday. Trying to decide if I want to go Coil Craft or external (not smd). I'm not concerned about the driver being low profile for my purposes, but I am concerned about ripple and that particular shielded inductor sounds like the best way to go. I don't see through holes in the PCB for an external inductor. Those large pads, however, could be easy to rig a stable mount.

The pads should be heavy enough to support an external SMD as long as you aren't too rough with it. They're pretty big. If you're going external, go for a 10uH+ value--it will be more efficient.

 

[Benm]

Yeah, those pads are big enough to hand solder an inductor to after getting the reflow work done.

Through hole connections are much more shock resistant (you could rip the traces right off the board otherwise) but i suppose this was not an option here as they use the inner layers as heat conductors on this board.

 

[ImA4Wheelr]

Thanks for the advice gentlemen. I was thinking about soldering copper sheet each pad. The sheet would then mechanically bonded to the external inductor. Actually, the order would be reversed. Like Benm recommends, I would had solder the copper sheet to the PCB as it would take way too much heat to reflow.

I've been second guessing going with all my semi compatible components. I've sent you a message RMM inquiring if you have a parts kit available. I don't see one on you website.