.:madworm:.'s photos
My makeshift office
LT1618 - step-up LED driver
| |
|
I need a step-up LED driver for the 12V LED bar I got at seeedstudio. I'm planning to put 2 of these on my bike and power them with 4 AAA cells.
The LT1618 supports constant current drive and a max. Vout setting. It can be used as a variable voltage step-up power supply with over current limiting too.
SMD rework - ChipQuik test set
| |
|
Moving on to new adventures: SMD desoldering of big fat chips. Getting a hot air rework station is too expensive now, so I'll try this first. The first victim will be an ATmega64 sitting on my Jyetech oscilloscope DIY KIT.
www.chipquik.com
Jyetech - Arduino speed measurement
| |
|
www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1248812473
about 8µs for:
while(1) {
digitalWrite(13,HIGH);
digitalWrite(13,LOW);
}
Please note that a new hardware version of this device has been released, therefore these images will most likely not apply to it.
Arduino speed measurement - 2µs/div
| |
|
www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1248812473
about 7.6µs for:
while(1){
digitalWrite(13,HIGH);
digitalWrite(13,LOW);
}
Tektronix 314: Arduino speed measurement - 0.1µs/d…
| |
|
www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1226290298
using direct port manipulation
about 0.36µs for:
while(1) {
PORTB |= 0x01;
PORTB &= ~0x01;
}
LED - uniformity issues
LED - uniformity issues
LED - uniformity issues
| |
|
Quite a few LEDs emit too much red. I will need some sort of dot correction to compensate ;-(
8 out of 10 boards have issues.
LED - uniformity issues
| |
|
Quite a few LEDs emit too much red. I will need some sort of dot correction to compensate ;-(
8 out of 10 boards have issues.
LED - uniformity issues
| |
|
Quite a few LEDs emit too much red. I will need some sort of dot correction to compensate ;-(
8 out of 10 boards have issues.
LED - uniformity issues
| |
|
Quite a few LEDs emit too much red. I will need some sort of dot correction to compensate ;-(
8 out of 10 boards have issues.
LED - uniformity issues
| |
|
Quite a few LEDs emit too much red. I will need some sort of dot correction to compensate ;-(
8 out of 10 boards have issues.
LED - uniformity issues
| |
|
Quite a few LEDs emit too much red. I will need some sort of dot correction to compensate ;-(
8 out of 10 boards have issues.
LED - uniformity issues
| |
|
Quite a few LEDs emit too much red. I will need some sort of dot correction to compensate ;-(
8 out of 10 boards have issues.
LED - uniformity issues
| |
|
Quite a few LEDs emit too much red. I will need some sort of dot correction to compensate ;-(
8 out of 10 boards have issues.
The manufacturer recommended using drivers that support dot correction. Also they don't check/sort the chips with respect to emittance before assembly. DUH! I wonder why all the other boards I got from them before were perfectly OK.
I implemented dot correction to the latest firmware of my controller, but it costs some speed. The dot correction is stored in PROGMEM and the reads seem to be somewhat slow. I'm also facing a few casts back and forth between int8_t and unit8_t which also takes time.
I think I'll have to rewrite the framebuffer code to allow gamma correction as well. Currently a single run of the ISR refreshes the full display and blocks other stuff for about 50% of cpu time. A rewrite to just do one intensity level per ISR run would spread the load more evenly and may be better for other jobs like I2C and serial.
For dot correction on driver level I'd have to switch some chips. No time for that right now. Getting flawless LED boards would be much better.
Seeedstudio 3W LED bar
| |
|
I took this image from seeedstudio's shop to use it on my blog. Don't want to add to their bandwidth. Also I hope they won't sue me for this, as I'm a good customer promoting this brilliant piece of magic for free.
I need more of these, I've seen the light. Incredibly bright, even at 50mA. Maximum current is 240mA. These will go on my bike. It'll look like in TRON.
RGB Matrix V3.04 - assembled back side
| |
|

















