Has anyone heard of the SK9822 Pixel,

Archive FastLED Archive
,
1

Has anyone heard of the SK9822 Pixel, I’m getting wind from my manufacturers that its an APA102c Clone, that it has a 30Mhz data rate, and something about “SYNC” although I’m not able to get much more info. My supplier is being tight-lipped for now… The only thing I found online is this http://www.normandled.com/index.php/Product/view/id/800.html
But it has a badly translated description that looks exactly like the APA102c description… Anyone?

2

I had some contact with Alan from http://www.szledcolor.com/
They have them on stock. Next month I want to order a custom LED panel with 32x32 LEDs on PCB.
In the past I already ordered some panels with APA102c LEDs (32x32 Pixel) on 100x100cm
Good service and fast

3

@Jurgen_Skrotzky Do you have a data sheet for these SK9822 Pixels or do you know where I could find them?

4

Partial success I contacted Alan and he sent me a Chineese datasheet http://www.elec-tron.org/wp-content/uploads/2016/04/SK9822_REV01_CN.pdf … Anyone speak chineese well enough to have a look?

5

Hi @Leon_Yuhanov :

Just copy and paste the PDF datasheet into google translate (https://translate.google.com/). It works with the datasheet as seen below:

  1. Product Overview:
    SK9822 is a dual three-channel transmission (RGB) drive control circuit and lighting circuit in one of the smart external control LED light source. Within the product contains signal decoding mode
    Block data buffer, built-in constant current circuit and RC oscillator; CMOS process, low voltage, low power consumption; 256 PWM gray scale adjustment and brightness adjustment 32; mining
    With two-output, DATA and CLK signals to synchronize data, the operation of each wafer so that the output series of synchronization;
6

It looks like, from a data timing/protocol standpoint, it’s a drop in replacement for the apa102 - how it compares to the apa102 when it comes to things like refresh rate and such, who knows…

7

Cheers @Ken_White Ill give that a go… @Daniel_Garcia I have ordered a 5M strip and will post results when it arrives. Considering how hard it has been to get APA102 if this is the same or better I will be very happy. Pretty much all my suppliers from China keep saying “The APA102 will PROBABLY be available by the end of April”

8

Just curious… Has anyone heard the reason for the APA102 shortage?

9

@Justin_Onimus I have been trying to find out from my suppliers for months, but they wont say why

10

Pure speculation on my part, but the market oversupply among LED manufactures has led to imminent bankruptcies and cutbacks.

Nan Ya Photonics Incorporation (NYPI), is ending it’s wafer manufacturing business, and transferring 10 - 13 machines to Epistar in the near future. (Not sure, but this may well be the source of Epistar APA102c, since Epistar owns a 50% stake in NYPI).

The “Maybe end of April” info we’ve been getting fits the timeline of the wafer machine transfer.

Again, just a guess.

11

Interesting info, thanks @Justin_Onimus .

12

One seller told me that there is only one company in Taiwan building the APA102c LEDs and they have some issues with the chips - so they decided to improve them and are still in testing - The new ones arrived end of April was the information of the seller. He already have some improved LEDs for testing.

This is the reason why the decided to build there own data/clock chip based on the APA102 (= SK9822)

13

Would be nice if they improve yields and eventually the price drops a bit.

Thanks for the info @Jurgen_Skrotzky

14

I’ve been trying to buy APA102C in large quantity for months now and haven’t been able to source any since Chinese New Year. It’s driving me nuts. I was sent some “compatible” chips to test about a month ago, and they are probably the ones you mention although I was given no name or spec sheet.

UNFORTUNATELY THEY ARE NOT 100% COMPATIBLE.

I haven’t take the time to understand why, but the display is always one frame behind. For instance, if you fill your strip with blue and send Fastled.show(), the strip will show nothing. Then, if you fill your strip with red and send Fastled.show(), it will show blue! On, and on. This isn’t really a problem if you’re running show() frequently and inside loop, but I’m more judicious about when it’s run.

If @Daniel_Garcia wants the LEDs to take a look, I’d be happy to send him my test board of 256 (16x16 grid zig zag).

15

It sounds like there might be more of an end frame needed to lock the previous frame in place? I’d be happy to test - I’m going to be in Boston until april 26th - drop me an email - danielgarcia at gmail

16

@Jeremy_Williams I with @Daniel_Garcia probably an end frame issue. My test strips are HOPEFULLY arriving next week and will test them out and report back

17

Alan from szledcolor send me an updated datasheet:
http://www.szledcolor.com/download/SK9822%20LED.pdf

18

@Jurgen_Skrotzky legend thanks. ive ordred a test batch from Alan already…should arrive Monday

19

Awesome - please report your testing results.

20

Hey guys - just got a mail of Nita from iPixel LED with their sheet about the SK9822 (I drop it in my drive for public):
https://drive.google.com/file/d/0B30DjPkusBqkY1Q2LTFsdUpTZzNjODlIcjdCeHFEX1J2anlr/view?usp=sharing