17mm & 20/26/27mm single-sided DD/FET driver release: A17DD-SO8 / A20DD-SO8 / etc

I’ve been thinking of making an updated landing pad for D1/R3 with a little more guidance for the SOD-323 diode’s leads. That won’t help with C1 though, I think that’s the only other component it looked like you had noticeable trouble with. (and I mean that only in a perfectionist sense - clearly the ‘trouble’ is only a mild aesthetic problem!!) I think a little less paste + improved heat gun technique is probably the key to that one.

I did not do more than two driver buil-ups with it yet, but I certainly can recommend using a heat block, like my converted 12V solder iron, it worked perfectly, and no components blown away.

( https://budgetlightforum.com/t/-/16170#comment-334252 -post#27- for how I made it , and oh, do not watch the shakey led reflow vid in this post, it was almost two years ago, I'm way better at it now :-) )

I use a 6” diameter 1/8” thick SS plate on my stove top. I just did 2 drivers still joined together and it worked out beautifully! Finished them up with wires and springs, complete with spring bypass and they’re still joined together. :slight_smile: Those 2 are going to bugsy to help him on his inspections…

On parts getting blown off… I found that it helped to make sure my paste, PCB, and parts all started at a reasonably warm temperature when I assembled them. I generally give all the components a little poke down into the paste too, but sometimes I don’t like where that puts me. Then I have to remove the component and re-do that paste.

@ djozz - I don’t think I could handle having hot chunk of aluminum like that plugged in on my workbench. I’d be burned in no time. :wink: [Also I’m too impatient for a solution of that type, I’d have to build a temp controlled one w/ high wattage for a fast heatup.]

I'm loving the layouts on these. To me it's the 3rd generation and each was a step up: BLF (Mattaus/comfy/warhark), 2nd: C_K, 3rd: the "wight" one Smile.

I think recent posts by Mattaus and C_K imply those designs may be done - won't be updated anymore, but not sure if they were serious, or meant no planned changes for now...

I love the short traces, everything right there, the continuous ground rings with good width. The pads seem huge but don't think it makes much of a difference - I know I'm over-doing the paste, but my fault. I haven't powered one up yet, so I know I'm speaking early - by the weekend I should know a lot more... I bought qty 12 of the FET's, so if the newer FET's are preferred, I'm not over committed.

If you are thinking of board tweaks soon, I'll hold off buying for a little while - I bougt only 6 of the 17's, 3 of the 20's.

Update: I use a coffee mug warmer. Tip from a friend who's done a ton of hobby (ham radio, rocket stuff, etc.) and side job prototype/1 off electronics (an EE). I'll sit the boards on the warmer for like a 1 minute or so to warm them up. My friend says it's about the perfect temp.

Matt is burned out, stepping out completely. So he won’t be redesigning anything for a while, if ever. I think he’s got a lot on his plate and all this was taking up too much time.
I feel like he’ll be back, maybe soon even, but he’s got to step out for a while….like many of us he’s got the bug though…he’ll be back. :wink:

Cool Dale. I've said the same sort of things - stupid forums makes your words into a permanent record, not reality at times, but just records your thoughts of the moment... I get into trouble at times, sure many of us do... All good here though - why I love BLF! Smile

Sorry Dale :_( but you should run through those in a week or so right? I’ve lost track of a few threads but was reading about single cell FET problems somewhere and thought I’d look around. There were a number of ultra low Rds on chips listed on the nxp site but non were lower in the 3V range and this one even had a different scale for resistance(ran off the graph at only 6mohms/2.5Vgs). Like I said above though, it would make a lot more sense in this case to do a GB which I’d be willing to do for something this straight forward but only if the gains were realized. To that end would you like me to order some for you to do a head to head comparison? My modding time is as always minimal but I can order things easily enough(a bit too easily).

I can understand the disappointment Tom E with your above comments but in my mind you have done an amazing job. I'd be only to happy to take these of your hands so you no longer have to look at them if you like. Thanks for showing us silly buggers your work.

I have been using the same heat gun you have after an earlier post of yours and love it for putting leds on stars.

Ohh, so cool! Yes, agree think it works pretty good, considering. The coffee cup warmer makes a big difference though - the heat gun does it's thing quicker, and think you get better solder flow.

Good tip. Do you just leave the boards on the warmer when you apply the hot air, or do you pick them up and move them? Do you add your solder paste and components before or after putting the PCB on the warmer?

I do the solder paste and parts first, then sit them on the warmer, give them maybe a minute to warm, then use the hot air with them sitting on the mug warmer. My friend said he doesn't use the mug warmers anymore because pretty much all the boards he does now are bigger. He uses an electric skillet now for everything - can handle bigger boards easily. I did the 3 at one time, but think I'll only do one on the warmer at a time - easier to manage. I should be doing my LED reflows the same way -- mostly still doing the frying pan, but totally forgot bout the mug warmer on the last round.

I should take a live action shot, or video Smile.

I built a Courui with the A17DD-S08 this evening. Toy Keepers ramping firmware. I raised the XM-L2 U2 1D on a copper pedestal for ease of focusing. Also bolted in a 1” long x 1.25” diameter aluminum heat sink under the emitter shelf. Put a short 18ga wire on the pcb in the tail with 22ga wire bypass in the springs. With freshly charged Samsung 25R cells it does 1718 lumens with a 166.5Kcd lux and 816M throw. Very nice light!

This driver rocks Wight! :bigsmile:

NICE!!!

you should have a go boosting the coffeewarmer a bit one way or another, so that it gets hot enough to do perfect reflows all by itself!

I went to the coffee mug warmer mod forum - some usefull info Smile

Anyone look into values for LVP yet with these? I tried the same values I used on the C_K boards with 22K R1's and seems like it's tripping lower. I'll need to raise the values. I think I used 115/105, so will go for 123/115 - hoping that will be about 3.0v/2.8v, maybe slightly less...

Best I could figure is the first level tripped at about 2.75v, critical trip was maybe 2.45v or so.

I used my bench power supply and double-checked the voltage across the LED with the DMM.

BTW, on that note… if you have suggestions for LVP values to use on various drivers, I’d like to include those as presets in the firmware.

I’m hoping for the attiny13 internal voltage value for 4.2V, 3.9V, 3.6V, 3.3V, and 3.0V. It’s a pain for me to calibrate voltages since I don’t have an adjustable power supply.

I tried 124 (low) and 115 (critical) last night and they are better, but just not sure of my bench power supply setup or measurements. Sometimes the supply seems to be close to the DMM, sometimes not. Bottom line is I have no clue how accurate the reading is and not sure if reading the voltage drop across the LED is an accurate way to determine the equivalent of a battery's output. I know there are voltage drops typically - I've seen other posts about that. I thought the 124 was close to 3.0v, but really not sure - not confident with my setup and readings... This is why I'm asking for other input.

Led Vf is most but not all of the voltage drop. Most of the rest is across the driver so to get a more accurate reading go from led+ to ground. I would think it safest to measure voltage on low and compare that to resting cell voltage adjusting from there.

Thanx Scott - I'll check that way - LED+ to ground. I was using high mode but at lower amps of 0.5 to 1.5A or so -- No PWM's. Hhmm, maybe that's way my readings were all over the place because I tested at different amp levels.