The spring-side 7135s on TA’s drivers have their tabs all on the ground ring. That is a nice big ‘pad’ and there is thermal contact with the host as a bonus. (The PCB artwork may be misleading, but it is only solder resist defining the pads for 7135 tabs - there is one big ring of copper underneath connecting all those 7135s. Please correct me if I am wrong TA…). The internal side 7135 similarly has a good copper area as well.
For the pin 2, I would solder them down in principle. But electrically it is the same as the thermal tab and the thermal tab has priority if it is not feasible to solder both.
I’m preparing the parts to mount one Texas Avenger 20mm board to my Courui D01, replacing the existing FET+1 driver that I conected to the Courui striped original driver board. I intend to use the TA driver with the attiny85 (e-switch) and narsil. My question is: is the original narsil compatible with triple channel boards or there is a triple channel ready version? Help please.
Tom E wow, that was an instant reply.
As TA boards could became an universal design to be adapted to different setups, an triple channel narsil firmware would be of great value. Hope that you can develop a version asap. Thanks.
I’ve been trying to follow this but I simply can’t keep up with it all.
I don’t know enough to even know how much I don’t know on the driver end of all this.
That said, I LOVE what I am seeing here with a killer new driver design from Texas Ace! Help from DEL and Tom E working to bring Narsil to the table just brings it all home. :student:
This is what BLF is all about. :+1: Well done on all sides.
Yeah, I saw that. As I said over there I simply streched out the normal TA driver to 30mm in order to have a driver large enough for basically any light made and allowing it to be sanded down to fit anything smaller. I just picked 30mm because it was a round number and happened to fit the L6.
I was not aware that Narsil supported dual switches so I didn’t really consider changing anything for the L6 in particular. Now that I know it does I could make up a set of drivers with an LDO option as well. What is your preferred LDO?
The real issue with the LDO that I ran into when designing the 4s SRK driver is how would it fit into the new parts layout from DEL? I could not figure out where it should go in the layout. I am guessing the smallest driver that it would be able to fit on would be ~22mm and that could be tight.
Yea, gonna have to git it dun . Gonna go downstairs and reflow one now, actually...
I'll check what LDO's I got - weird, just checked MtnE and he doesn't sell them, least now... Thought I got them there - could be Mouser, where a part # was posted somewhere...
The 105C as a slave is what your problem would be it sounds like. A floating 7135 will have overheating issues, particularly in a 2s setup. Since these drivers are inherently designed to not need to be used as slaves this should not be an issue in theory based on results I have seen from others.
I could be wrong though, Once I get some screws to screw down the LED (my host didn’t come with them oddly and don’t want to run it too long without them), I will do a proper test to see if they are reducing output. I was watching the lumen drop over ~5 mins with the 7135’s though and it only dropped what I expected from the heat.
Yep, you are exactly correct, the 7135’s are all directly connected to the ground ring that is as wide as possible to heat sink them directly to the flashlight body.
I also would like to solder all the pins, I tried at first but it just wasn’t happening on the small drivers and the large ones are a lot cleaner if I don’t, not even sure how I would to be honest.
Do you know off hand where in the schematic I would add in an LDO BTW? I can add it but not sure how to keep everything working optimally with it being an optional component.
I look forward to seeing narsil made compatible with triple channel drivers but after all the work on the Q8 you deserve a break. Take your time (just don’t forget us entirely lol).
Yep, I was a bit surprised just how small the whole boards look when they showed up. Gets disorientating when you build them and they are 8”+ the whole time :nerd_face: .
They are a bit oversized for the actual pad and the pads are part of the overall ground ring to maximize both current flow and mostly heat sinking.
For the LDO… 4.7 ohm, then C1, then the LDO and then C2.
LDOs are very particular regarding the output C used (C2 in our arrangement). Check the specific data sheet, but typically a good X7R, 1 uF or more, is required.
That is kinda what I was thinking, although the specifics of how to lay that out so that the LDO is an optional component is confusing.
C2 being used as both C2 and for the LDO was another issue I was not sure about. So it sounds like using C2 for both would be fine? Possibly just a larger cap?
I will try adding it to the schematic later and see what you think.
Think'n, I should combine ramping and mode sets into one -- single click for next mode, press&hold to ramp. You really want to hit full 1 7135, full 8x7135. Under pure ramping, you will not be able to find those non-PWM levels so easy. could do it another way with a pause in ramping, but that UI may not work well - hard to recognize the pause, take too long, etc.
The issue is more so how to wire the LDO into the rest of the circuit so that it will function with it in place or not. Bridging the pads of the LDO itself to bypass it is an option I was considering, the issue is seeing if a resistor is the correct size to do that or if I can adjust the pads to allow that to work.
I am thinking that since you don’t need the diode with the LDO, I might be able to use that as a bit of a “switch”. Gonna have to play with it some.
. Gonna go downstairs and reflow one now, actually...
