Can the MCU pin#5 be used for PWM output also, with the ATTINY13A?
That’s how the FET plus 7135 driver works. Uses pin 5 for one and pin 6 for the other. Somewhere in the code the mcu is programmed to send the signal to one pin or the other you would just need to change the pin that gets the signal.
Ohaya wrote:
Hi,
I can’t see images where I am now, but is it the same problem that was pointed out in post #5? If so, I’ve already eliminated that short.
No, its close to that spot though. The edge of your red photo shopped oval is touching the ground pad. The PWM pad is next to it and is soldered to your FET's Gate pin.
Those center ground pads are isolated and not connected to any traces so it shouldn’t be an issue. Only the outer ground tab is actually connected to the ground ring. In the future it might help FET clearance if you solder the gate to the Q4 pwm pad as this would shift the FET over the board more. Any pwm pin pad can be used but this particular driver might short led- to ground if the FET tab were to contact any metal in the head. It wouldn’t take much for it to scratch through some ano.
^ Ah, I see what you're saying. Yeah, the "short" I spoke of is a non-issue.
Maybe i’m totally wrong, but isn’t the FET mounted the wrong way?
I mean upside down is correct but shouldn’t it be rotated 180° horizontally with the connectors on the other side?
Not according to This data sheet
What is important is that the FET gate pin be soldered to one of the pwm pads and the source pin to ground and he has that. Not so clear from the pic is whether the center drain pin is also connected to the pwm pad with a hair wire short.
Sorry for the ugly drawing, but no, it is correct. It looks like this:
I’ll re-check, but FET drain pin is cut off (there is no leg/pin.
Thanks, i stand corrected. It was just a thought…
Yeah, I saw that it was cut but couldn’t see if it was clear of the solder to the gate pin.
I’ve found the problem!!
For reference, see the pic in post #18. The top end of the resistor labelled “4701” should be connected to the ground ring, but I had sanded this driver to get its diameter to fit a pill, and basically removed the ground ring in that area, so the resistor wasn’t connected to the ground ring anymore. I soldered a small wire from the end of the resistor to the one of the ground pads for one of the 7135s and it seems to work now.
Pic 6 in post 4 shows it best, the other pics look shiny but that one shows the fiberglass underneath. Good job.
Nice work guys. I did not understand anything but was following with interest.
Offtopic - has anyone nodded a light w a SSC p7 emitter?
Would a Get driver have problems driving an p7?
I put this driver in a light w a p7 and it is dim, but if I use a bench supply it is bright.
Weird!
You might look up the thermal resistance of the P7 and compare it to Cree LEDs. I don’t know offhand how hard it can be driven but I’m guessing not as hard.
It’s really very strange. I can pull the head off and power it with my bench supply and it’s bright, and then I put the head back on and using a battery (and I’ve tried several including a Sony V4(?) and it’s dim.
Weird!
EDIT: BTW, this is with the FET-modified NANJG that originally had the problems. I’m starting to think that that driver is still borked in some subtle way. I’ll try to put another driver into the P7 light tonight. It originally was pulling only about 1.4 amps at the tailcap and I really want to push more amps through the emitter to see how it does.
What current and voltage on the ps? Fwiw, the only figure I found was 3C/W which is quite high and indicates to me that high current isn’t an option. Add to that the large quad die image and you get a “meh” led, too outdated to compete. Good for a rebuild of an older camp light or such. Lots of old 3A builds but I don’t think it would do to well above ~4-4.5A.
I don’t think current was very high with the power supply, maybe just less than 1 amp. That’s another thing that’s puzzling, I mean I was using good batteries but still got only dim.
Hmm. I wonder if there’s some conductivity problems with the light, like maybe the body wasn’t conducting well or something. Will have to do more checking…
It should handle 3A without issue if mounted to a heat sink. I wouldn’t use it in a P60 even if it could be made to fit.