BLF17DD Info Thread - Reference

I’m sure that Cereal_killer’s intention is still for you to use 0805 sized components. They will sit on those smaller pads and can be soldered to them. I would use 0805 sized pads for those components. (An alternative would be simply transitioning to 0603 sized components to place on those 0603 sized pads, but I don’t see why I’d bother. I’d rather just use 0805 on 0805 pads.)

It’s not really clear what is going on with your malfunctioning driver. What do you mean 50, is the LED about 50 as bright as normal? Do you no longer have modes? Is there continuity between LED- and GND while the driver is turned off?

I’m not certain what you mean by “bombed”. That resistor is in place to protect the MCU from FET, not the other way around. What spike are you referring to? Please link me and provide a post number.

I have a thread on it lemme get… BLF17DD Troubleshooting

It’s the FET but I haven’t tried measuring LED- to GND when off… I tried a few other things though including a new driver wired to the FET to eliminate the mcu

The pads are for 0603’s, but as I was saying to fellfromtree I’m sure the intention is to continue using 0805’s. This does work and is done by others here as well. I do not enjoy soldering 0805 on 0603 pads: there is no exposed pad at all with the component in place so hand soldering is more difficult and inspecting the joint is impossible. There’s also significantly less solder present, so less surface tension keeping things in place while the solder is molten, regardless of the scenario (initial reflow, hand soldering, rework, etc). The upsides (as I see them) are that you can achieve a much higher component density and ensure that there is less wiggle-room for components.

RMM’s board was never really described very much, the only mentions of it I know of are from the Oshpark projects thread in the past week or so. Over there RMM said that it was never really a big deal so he never mentioned it - just a couple of tweaks he asked Mattaus to make vs the current versions at the time.

I’ve just skimmed your thread and I don’t see Tom E in there mentioning a spike. I’ll chime in with my thoughts in that thread though.

Wight I think pretty much answered all the questions (I’m really sorry I dont have more time to post, I have literally been so busy the last month now the only time I have to answer emails is sitting on the can lol).

The 0603 pad’s are the correct size for 0805 parts, I use those common part’s from the sparkfun library and they are extremely oversized. I do all my boards like this and use all 0805 parts. As for the D2PAK (the FET) part, thats my own .lbr part, it is the correct width and only slightly taller (to allow space to solder wire’s above the tab.

About the “stars”, using the negative circle of mask is a new way for me, I had been using a 1mm thick line to expose 2 edges to jumper, I dont have any of these yet (and I dont use stars anyway, I just change my code as desired) but I’m willing to bet they work. As previously explained (and then mentioned by wight) the way oshpark render’s the mask is by adding another layer of gold color, it’s is the exact opposite of what really happens in fab- anywhere you see gold color in the render there will actually be nothing, mask is applied everywhere EXCEPT those spots.

The SW+ via is on pin3 (star3) cause thats what I change my code to use, I had forgot I changed my code when I did that [sorry], its a simple pin assignment swap to change your’s (or you can continue to solder direct to the MCU’s leg for pin 2 or whatever stock STAR FW uses). I prefer the via cause it makes for a physically stronger connection as well as allowing hookup from the bottom in a piggyback application.

Again sorry I cant sit around and answer questions and “hang out” with you guy’s all day (you know I love ya all) but right now flashlight’s are literally supporting my family of 5, I’m working on them about 10-12 hours every day (and its been like that since school started back up for my kid’s in August), I just dont have much of any free time and when I do I dont want to spend it sitting in front of my computer talking about flashlights lol…

Hehe good deal Cereal :)… I might shoot some guys over your way if I can’t figure out the RGB thing

Wight it’s mentioned in the op of this thread. I didn’t know what it means either but he mentioned a spike, maybe to the mcu I’m thinking, but that could in turn maybe end up at the gate from the pwm pin? I dunno I’m still wet behind the ears

Thanks fellfromtree, I see it now. I’m ready to be corrected, but I think that was simply Tom E misunderstanding the exact function of that resistor.

What RGB stuff are you working on? Tho I wont be releasing my code open source I do have a much more simple open source RGB project I’m working on (does not allow ramping of the individual colors, that will be staying proprietary, I am selling my driver’s however, I’ve just created a sales thread for them the other day) It will simply be a few white modes, then a fixed brightness level for the individual colors (can be changed in the code, but not changed from the working light) then a RGB (fixed brightness) mode (and maybe a strobe, if I feel like it)

Probably best to PM me, already a lot going on in this thread (also note, the open source RGB stuff is very low on my priority list, probably be a few months)

Oops, updated the OP with: reduce a voltage spike flatten the peak. Think that makes betters sense? It's what I meant Smile, at least that what the o-scope graphs look like.

Tom E, which specific scope images are you thinking of? As I insinuated it’s possible that I’m missing something here.

On the v1.0 boards, is there any other place other than the labelled pads that can be used for emitter +?

Edit: I mean on the emitter side of the driver…

On the “BLF17DD V1.0” board there is a copper pour with 4 vias which connects directly to BAT. The middle half of that pour is exposed, but the entire pour can be used if you scrape off the purple resist. Besides this small copper pour there is no other area which may be used for LED.

If you’ve torn off the copper I suppose you could drill a hole clear through the board. You’d then scrape an area on the bottom side and solder your wire there. :frowning:

I have done what wight suggests before, drilled a hole through the board and ran a wire from the top of the spring and directly to the emitter.

Yeah cereal lemme try my ham fist methods first. I’d hate to take up your time with my issues. I was actually thinking like 2 dd boards stacked and each works with each other through programming the tiny13a to get the different modes which will be red and uv. The pill in the Y3 will be big enough to do this. Thanks for looking out though I really appreciate it. If I have questions, trust me I’ll be in here asking! :slight_smile:

Thanks for all your help Dan… appreciate it

EDIT Hey I was thinking today… What if you just stacked one of these on top of the yezl y3 buck driver for modes? You could cut the switch pin to the mcu on the buck, bypass the fet so it’s always high while still having the voltage regulation intact. Then run the LED + on the buck to the emitter. Run B+ and GND to both boards. Run the LED - on the BLFDD to the emitter. Programmable! 2 cells… I know the buck can run 3 but not sure if the zener mod can handle that.

Also let’s say I wanted to run multi emitters. Can I hook a fet up to PB0 OCR0A and connect a free floating FET there for a second pwm out. I noticed it’s not doing anything on the board itself. So I could in theory run a multi colored emitter as a programmable buck up to 6a

Thanks, DBCstm. I’ll keep that in mind next time.

Jim

Hey Cereal_killer, I just had a chance to check my 22mm 7135 driver and I noticed that both your boards and mine use 1.1mm vias. According to me (post #38) I was unable to get 20AWG to fit through my 1.1mm vias, much less 18AWG. What gives? Do I just suck or do you have a secret technique? :slight_smile:

I was trying to upload a pic but apparently photobucket mobile doesnt work with IOS 8.1 (yet another reason I wish I wouldnt have updated).

Anyway I’m able to get 18AWG in threw 1.1’s, the way I do it is- strip it so its nice and clean, take the stripped end in my fingers and twist it as tightly as possible, flux it, touch it to a hot iron so just the smallest bit of solder goes into it to hold the strands in place (do not let even the smallest blob build up). Then I get the board, press it straight down on the via, re-apply the iron (from the top of the board) and it takes about a second to heat up and slide into the via, I usually have a handful of straggler strands that dont make it (usually under a dozen if you kept it nice and tight when you fluxed) and I simply cut those of with an exacto knife.

Edit:

Ok so here it is inserted threw the via but before the loose strands are trimmed off, you can see on this one I only had a couple. After I trim them I touch the iron back to it to make it perfect. Note that “20” you see on the wire is where its marked “200*”, this is 18AWG silicone 120 strand.

Interesting stuff, thanks C_k! I’ll definitely give the big wire another try.

Love the new design…the copper pour using the polygon is VERY nice!

Revisions just kept getting better and better!

Great job CK!!!

Mine were just shipped today… I like the design too CK- seems to make a lot better use of the copper