GS8208 LED Strip

BradsXmasLights

BradsXmasLights

WiFi Interactive
Joined
Dec 23, 2010
Messages
605
Location
Brisbane, Australia
There's new LED pixel strip out with GS8206 and GS8208 chipsets! So with just a datasheet and very limited info online about it, I ordered 75m of GS8208 and am hoping for the best. This thread will demo my findings with it.

This GS stuff is made by Genesis Systech and is protocol-compatible with the WS2811 variants by WorldSemi

First off some basic specs:
* GS8208 is 12V based and has an integrated IC - 1 LED per pixel. This would be the closet equivalent to WS2815.
* GS8206 has a separate IC chip - multiple LEDs per pixel. Voltages & LEDs per pixel vary; eg 24V, 7 LED/pixel.
* GS820x is 4 wire based, positive, negative, data and data-backup lines. Backup data (also on WS2813/WS2815) is wired in a leap-frog style configuration. If data isn't received from the previous pixel on the (primary) data line; signaling is then sourced from the backup line. This is wired back to the output of the node two pixels back - so you can effectively bypass a dodgy pixel entirely.
* GS8208 strip is currently $4.80USD/meter from RAY (double the price of WS2811)
* GS820x is 8bit - but has 8bit to 12bit gamma correction
* GS820x has an automatic test mode
* GS820x R/G/B LED elements are wired internally in series
* Datasheet with more electronic engineering specs: https://www.deskontrol.net/descargas/datasheets/GS8208.pdf
GS8208_wiring.JPG

So my question #1 - is it really protocol-compatible? Answer is Yes. Tested with a (now old) J1SYS P2, it worked perfectly first go.

As soon as power is connected - without data signal being generated - the lights will enter a self-test mode. This started off as a simple red, green, blue, yellow,etc colour cycle, then switched to rather presentable rainbow patterns. In fact it did so many different patterns you could just connect power and sell it to people with static displays.

Now the catch with this self test is it starts instantly upon power on & it took about 4 seconds for my J1SYS P2 controller to 'boot up' and kill-off this automatic self test. (The J1SYS must've been sending out blank timing data; because i wasn't sending it E131 at that stage.) Switching power off to the controller then made the auto-test patterns restart. This has the potential to be an "annoying" feature!

Pumping some existing sequences to it; fades looked super smooth thanks to the single LED per pixel control. Not sure how much of this 12bit gamma correction is going on - but everything looked very fluid. However I did notice that Pink colour rendering looked rather whitish.

Next was power consumption measurements, with the entire 5M roll (150 LEDs), powered via both ends connected to a 12V Lab Power Supply. I've done this test with Ws2811 too and other lights so I can formulate PSU estimates.

GS8208_power_V2.JPG
Now WTF is going on here! Notice how single R/G/B measurements are off the charts compared to 2811. Mixing colours, which on 2811 would traditionally increase current makes very little difference, Infact mixing colours reduces consumption! I don't believe this is voltage drop related, powering from one end was only marginally lower draw. What probably is at play is the internal series-wiring of the R,G,B, LED elements and their PWM drive circuitry.

Feeding 12V to each LED too won't be helping either - but it's a price to pay compared to dealing the hassle of 5V based voltage drop over longer distances. In reality, it'll mean higher average power draw compared to 2811, just not higher total/peak power draw. This may be a surprise factor for some people upgrading if their cabling gauge wasn't up to spec to begin with.

More testing to come; including video of side by side runs of GS8208 Vs WS2811
 
Last edited:
That connection diagram is weird as all hell. It can be done electronically but it's a lot more complicated than what's used in most pixels. I can't really see why they would go to that. Page 8 of the data sheet kinda explains it and it also kinda explains why the R or G or B currents pretty much will always use the same current when any combination of colours is on. If I understand it correctly a failure of 1 coloured led will result in all 3 being turned off. Very puzzling bit of gear.
 
Hi, the data timing is interesting. Whilst it works with the WS2811 waveform at 800KHz it looks as if they have made it easy to run on a 4MHz SPI with 4 bits making up a single bit at 1MHz. 12 SPI bytes would make up a full set of 24 bit data.
 
BradsXmasLights said:
Video on the Self-Test "Feature"

View: https://www.youtube.com/watch?v=TYxH6L55cfI
Click to expand...

I can see this strip and the test mode being problematic for those who are setting up their systems for 30% or similar very low maximum brightnesses. At power up I should imagine that the strip immediately jumps into test mode where the current/load will immediately hit 100% and will stay there until data from a controller gets sent out.
 
I wonder what would happen if you put a string of WS2811 onto the end of a string of GS8208, would the self test pattern continue onto the WS2811 or would it stop? I guess because the GS8208 is generating the self test per pixel there would be no data for the attached string but could be an interesting test, similarly would adding some WS2811 in front of the GS8208 stop the self test, or would the self test still occur, what about a null pixel in front? If adding a WS2811 or a null pixel in front stops the self test when power is received this might be a work around for the side effect created by the self test.
 
Good question Kitman. I will give that a go when I test daisy chaining WS2811 & GS8208 on the same output.
I have a feeling the first pixel is generating the entire test sequence, as it all looked synchronized when it did chase patterns.
 
I love the idea of the built in IC and individual LED control from 12v in this, even the dual data may be handy but the self test mode really intrigues me....

What happens when your sequences end for the night and your controller goes into ‘idle’ mode? I am assuming you are going to have to make an ‘off’ sequence and have that come on at the end of your night to stop your lights blaring for the next 18hours til your show starts again... Seems like quite the pain in the ass!

My guess is that when u try daisy chaining the WS string after it will follow the test patterns.. And with the WS before, well my guess is that the GS will still go into self test mode...
 
I expect you can put a single WS2811 null pixel at the start of your GS8208 strings. I look forward to hearing Brad's tests.
 
Roosta said:
What happens when your sequences end for the night and your controller goes into ‘idle’ mode? I am assuming you are going to have to make an ‘off’ sequence and have that come on at the end of your night to stop your lights blaring for the next 18hours til your show starts again... Seems like quite the pain in the ass!
Click to expand...

In the case of the J1SYS P2, with it connected - but not receiving E.131, the test-mode did *not* start. (I guess the P2 was holding the data line low - or sending idle packets?)
But as soon as the P2 was powered off, test-mode started.
 
Actually on the topic of Null Pixels, this dodgy test setup with a 10M CAT6 cable to the first pixel worked!
GS8208_10m.JPG
That's with just just Pixel data connected to 1 wire of the CAT6 cable.

UPDATE: Actually still worked with ~27M of cable !!
(2 x 10M CAT6 with a joiner, and another DIY ~7M CAT5 lead on the end)
 
Last edited:
I love the idea of the built in IC and individual LED control from 12v in this, even the dual data may be handy but the self test mode really intrigues me....
 
You must log in or register to reply here.
Back
Top