The Texas Buck driver series, Q8 / Skyray King 2S/4S buck driver RELEASED!

Send a working link to that mosfet and diode you found if you get a chance.

remember these were just the first things I found and I didn’t peruse them much at all.

The P-channel version of the SIR800: http://www.digikey.com/short/39nh5h

This diode was literally the first thing I clicked on that had specs that should work, simply to see the form factor: http://www.digikey.com/short/39nh50

Great.

I still want to look over diode choice more soon. There might be things in more standard packaging without losing much.

Haven't searched for MOSFETs but on first glance that one looks pretty good. Still need to run reasonable numbers on switching losses when I get a minute for it.

Ok, I updated post 96 with a bunch of diode research. Basically there are good options in POWERDI5060-8, and in TO-263 (D2PAK). These can probably fit on the same pads if done right. Tiny heat sinks exist for the TO-263 which also need footprint and space consideration if used.

TO-220 through hole also has many options, in 2 or 3 pin. Those would certainly require some kind of heat sink, but are also easy to attach to one. If the surface mounts can handle the heat, it's probably better to use those instead. I see your point about the thermal pad being on the wrong side of the diode.

I think your SO-8 mosfet is almost certainly a fine footprint. So I think we're about all set for major components.

Great, I am real busy over the next few days but I will get around to updating the PCB before too long.

Now, about your circuit, I don't know about how C1in and R1Sense are arranged in the top right.

Edited: Never mind... I see Ti is actually intentionally reading ripple current, so it makes sense as is.

I'll start reading the manual more now. The constant off time and peak current control are interesting. It doesn't seem they change any of the fundamentals, just changes how frequency and current get controlled, so for instance in my table a particular frequency at one voltage output should probably imply a different particular frequency at another. I'll probably try to get familiar with it and re-arrange my tables instead of by frequency as by offtime RC constant. That will give a little better picture, but shouldn't impact parts selection I think.

Learning all the dirty laundry about this IC.

So if we use full input range for current adjust, 1.24V max on IADJ corresponds to 248mV of sense voltage. At 15A, VI is nearly 4 watts across the sense resistor. We could use half the range for control (lose resolution a little). Or add an amplifier (seems a bit drastic to save 5% power maybe).

However there is also this funny business where the peak current to turn off the switch jumps up and down because they keep flipping the voltage comparator every two cycles. See figure 24 page 15 and the discussion. To me this seems nuts. They claim it makes a more accurate average current, but accurate compared to what? It seems keeping the fixed target would be just as consistent but with less ripple and without all the annoying caveats that come after it. Actual inductor ripple will be higher than what I calculate because of this effect.

Enough grumbling, the point is they claim you should maintain a minimum ripple current (see equation 11), bigger than the variation in their sensing setpoint (that they caused), which they esitmate as maybe 24mV. Using full scale 248mV sense range that's already 10% at full current, so going to a much smaller sense resistor and smaller voltages is maybe not a great idea. Note this is not output ripple voltage though. Mostly we care about this because if we try to keep ripple high at full current, it's even higher at low current, and goes into discontinuous mode earlier. We'd rather not have high ripple current.

You have to read details there to understand it. Actually though I think it's ok to have ripple lower than what they say, at high power and I'm not sure I'd worry about the degraded current control they say it causes. Peak current still won't ever go above the max cuttoff(s) and the net effect seems to be you'll just get an effectively double off-time followed by twice as long of an on time, in other words, higher ripple current anyway (kind of by way of effectively lower frequency), but no need to target higher ripple and damage the low current performance at the same time in my opinion. Basically I would ignore the constraint in equation 11. It could make ramping uneven as you cross the threshold for double or tripple off-time, creating jumps in ripple current, since it's peak current that's controlled, not average and higher ripple for same peak lowers the average. But any given mode should still stay pretty steady.

I suspect when all is said and done, there's about 5 more places with 1W losses that are just too detailed to find, and all this brings 7V output down to 80% efficient at high power. It's a good reason to fight a percent or two here and there where possible though, like the input caps.

for footprint, 5W, 7 to 15mohm resistors are available in D2Pak's (like the diodes I listed) or in sizes from 2818 up to 4527. Bigger pads, lose more heat anyway, so. Could also make it through hole and try to sink it to the case. The D2PAK's are about $2.00 and the standard flat resistors are about $1.35 for the cheapest. This driver won't be cheap. The D2Pak's are rated at very high power, like 35W so might stay cooler.

In a 4 by xp-l setup this is only heading toward about 1.5W at 6A per led if tuned for 248mV sense at max current give or take (1S or 4S doesn't matter, so long as you tuned the resistor that way for either, duty factor cancels the current change). It's just driving 4 x 12V xph35's that pushes it close to 4W.

Ok, I had a few mins to knock out some updates.

The D2pak footprint is HUGE, almost as large as the inductor! It will not fit easily at all. I was able to fit 3x 2512 resistors in just over half the space of a D2pak.

I added more input and output caps in 1210 sizes. Obviously traces are still way undersized, I will fix that last. On the input caps, should they be before the sense reisstor like the datasheet shows or after so it is closer to the mosfet?

very neat:

http://www.nxp.com/documents/data_sheet/BUK7105-40AIE.pdf
current sensing mosfet. It puts out a tiny proportional current that you pass through a sense resistor. The method is described here among other places:

Very unfortunately, I haven't found one in a p-type. There might be some options for using it in n-type, constructing one etc, but probably not one-stop solutions :(.

Nice,

The one in the data sheet is right. I was confused on this briefly, but this sense resistor can't be shielded from the switching ripple. It needs to sense the ripple. Looks like you have it as compact as it can get there.

This sense resistor thing is a pretty big deal too. Looks like you followed that based on those three big sense resistors. That's another 4 watts and another big heat load to clear. Those d2paks are big for a reason (somehow I was confused they are bigger than I though). Big heat in small space though, we'll need some clever sinking.

Maybe you can make the inductor pads super huge at some point with some through holes in the middle to test unshielded inductors. I guess that's a 17mm inductor yes? I'll have a look at it all more closely soon.

The input caps right now are before the sense resistors, it seems like that would shield the resistors from the ripple?

Yeah, I thought about some through hole for the inductor (yes, 17mm, with all the new parts 22mm ain’t gonna fit) but got to remember that in a SRK the other side of the board is the battery positive so can’t really have anything sticking though the PCB for fear of shorts or interference.

I wondered about the battery interference,ok. Well would just have to bend the pins and solder it on the surface.

The resistor is supposed to be between the cap+ and FET input as in your drawing in OP. That will put it directly in series with the FET with nothing between the two. I think you've about got it. I'd really like to see the sense resistor gone even more than the diode, but it might not be practical with this IC. This seems like a pretty valid baseline implementation of this IC for crazy high power.

It's going to be a bit tricky to layout part selection rules for people. This constant off time means frequency changes with voltage for a given set of parts, which isn't necessarily a good thing as far as I can tell from my numbers, so may want different coff RC for different voltage output application. Also will want different sense resitor values depending on max current capability of your setup and priorities (use full sensing range for best control, but go lower to reduce heat and aim for extended xph35 turbo).

Getting these things wrong in hardware and/or software will destroy led's. This isn't like a 4.2 V direct driver, it's easy for it to kill pretty much anything we hook it up to (that should be a selling point around here :) ), possibly even xhps. This is something to keep in mind testing it. Having some good high power diodes (zener or whatever) for testing would be useful. It needs testing at the right voltage output, not just current.

oh, do you have a wire from the MCU to the EN pin? I think it it's good to throw it it for PWM. It's funny the doc talks about using an extr fet for PWM without an output capacitor on the buck, for higher pwm speed than EN allows, had me worried, then I realized they're talking about 20khz using EN and 100khz using the FET. So we're ok with just the EN pin.

Another thing, I'd rather have 3 input caps and one output, but it looks like a bit of a rearrangement. Looks like you had room for the output caps anyway, and two on input is still ok.

I forgot about the EN pin but that will be easy, I already see how I will do that.

The sense resistor I was not worried about, it was the input caps that I am not too sure on. Right now they are before the sense resistors per the datasheet but it sounded like you wanted them after the sense resistors as close to the FET as possible.

As it is, it should shield the sense resistors from the ripple a fair amount. It is easy to swap around, simply change the schematic in diptrace but was not sure what it was supposed to be.

I think it's right the way it is. The caps shield them from ripple voltage (it shields everything from ripple voltage) but not from the FET current. When the fet is closed, there's no current in that resistor. When it's open, there is, so it's getting that pulsing, and the sloping on current, which it should. It's in series with FET with nothing in between so it's got FET current. The battery side of the cap has current that's maybe more constant, as it recharges the cap during off time. Don't let me steer you off the baseline design too easily. I like making improvement, but changes from the baseline need a couple of days to digest. In this case it was just a momentary confusion as I didn't realize how this IC does current sensing.

I do want to review all my cap math in one sitting at some point. It's been strung out with parts selection in between and everything and a some changes, so yeah. Actually I don't think I've really thought through the input cap math at all yet, mostly just refined the output cap math. If you want to pull the trigger on a board order once you get the PWM wire in, it will probably work, but I will probably ponder it all one more time and generate a new set of numbers including sense losses, and higher power scenarios too.

Yeah, I won’t be ordering it for a little bit yet. I plan to get one when I make an order from a china house as it is WAY cheaper for large boards but I won’t have the funds for that for a little while (yeah, $10 is hard to come by lol).

I am just trying to get done what I can, when I can.

You could post it up though as highly experimental anytime. Some wildcatter might take an early stab at frying some leds with it.

I guess the right thing to test it on at first is 4S xp-l. I don't know if that's a reasonable use case, but they should handle full voltage without dying... maybe.

Yeah, once we are happy with the design and we do not think there is anything else we can do without actually trying it out I will release everything like normal.

The cost is a turn off for me, I just don’t see me building these for lights that cost as much as the driver itself. I like modding but I am also practical.

That said I want to see this come full circle, I hate unfinished projects.