Notifications
Clear all

Pixel Magic

9 Posts
7 Users
0 Reactions
2,993 Views
Posts: 1249
Topic starter
(@jstjohnz)
Noble Member
Joined: 17 years ago
[#25756]

At the ACDC show I was testing some of the 12V Technicolor pixel samples from Ray. My E6804 was running on 5 volts, so I knew it wouldn't be very bright, but I hooked them up just to see if they would work. Well, they did work, and brighter than I expected from a 12V pixel running on 5 volts.

The next step was to connect the pixels to a 12V power supply. So, to test with 12V, I first disconnected the +V pixel lead from the E6804, leaving only ground and data connected to the pixel strings (intention being to hook to a 12V supply). And, here's the magic: the 1st pixel in the string continued to work, with ONLY ground and data connected. It was considerably dimmer, but responding properly to all color changes.

I had to think about that one for a while....


8 Replies
Posts: 853
(@roterhead)
Prominent Member
Joined: 15 years ago

Perhaps there is, at the Pixel input;

A) a smoothing cap from pixel+ to gnd.
B) a pull-up R divider network from Gnd to Pixel+ with a tap at DIN
C) DIN is fed by 5V data signal from E6804

Then depending upon the level of the RGB signal 1s/ 0s the data (for example 255/255/255 RGB) the data could act as a charge pump for Pixel power. But Pixel+ is not high enough current to drive Pixels and signal reshaping circuit so no DOUT is generated? Guess that works for me!


Reply
Posts: 6728
(@p-short)
Illustrious Member
Joined: 19 years ago

Probably a protection diode between the signal input and the power on the IC is providing power to the chip and LEDs. It's common for signal inputs to IC's to have reverse biased diodes to both power and ground to clamp input noise/transients to the power and ground rails. If the power rail is left unconnected, the normally reverse-biased diode(s) between the signal input(s) and VDD bcomes forward-biased when the signal is 'high', so that power is provided from the output of the driving signal through this clamp diode to the VDD input of the chip, providing a small amount of power for both the chip and (in this case) the LED(s), also taking advantage of the bypass caps on the VDD rail.

My $.02...


Reply
chelmuth
Posts: 1077
(@chelmuth)
Noble Member
Joined: 16 years ago

I was waiting for the observations on the brightness of the pixels when they were hooked up to the 12v. The story just kind of ended. Maybe they were actually bright enough to burn ones retinas and he had to quit typing?


Reply
Posts: 2168
(@djulien)
Noble Member
Joined: 19 years ago
jstjohnz wrote:And, here's the magic: the 1st pixel in the string continued to work, with ONLY ground and data connected. It was considerably dimmer, but responding properly to all color changes.

I've had that happen with PICs when the serial port was still connected. It was actually rather annoying, because the PIC would not do a powerup reset unless I disconnected the serial line as well as power.

don


Reply
Page 1 / 2
Share: