Closed b4nter closed 4 years ago
Gadgetoid
commented
4 years ago I'm unfamiliar with RGB LED HAT, but I confirmed our - very similar - Unicorn HAT product as working on the Pi 4 and haven't seen any reports otherwise. Have you managed to get it up and running since?
b4nter
commented
4 years ago Unfortunately no, tried with the most recent update too.
2E0PGS
commented
4 years ago I know @b4nter IRL.
We both have been looking at this issue.
I have had @b4nter waveshare neopixel (WS2812B) hat working on my PI 2B with the C test program and Python examples.
We couldn't get it to work on his Raspberry Pi 4 Model B Rev 1.1. Even after updates to the code for hardware detection of that model.
Oddly enough I have quite a lot of SMART strip WS2182B LEDs and I tried them with the C test program and they work on both PI's fine. However I only use a test strip of 4 pixels. Perhaps there is a timing issue.
We tested different forced timings in the code. We also tested only lighting one pixel. None of this worked for the PI4 and that hat.
I will try measure the PWM rates of each board with the test program and a scope. However I need to aquire or borrow a PI4 for that.
PI4 is using official PSU.
jgarff
commented
4 years ago Are you using proper voltage level translators? Running 3v3 signals are not reliable and will cause intermittent problems and/or hardware and temperature variations. Working fine on some hardware and not on others is exactly what I would expect if you're not translating the voltage properly. Please see the wiki pages for more information.
On Sun, Jan 26, 2020 at 7:29 AM Peter Stevenson notifications@github.com wrote:
A bit more information.
I know @b4nter https://github.com/b4nter IRL.
We both have been looking at this issue.
I have had @b4nter https://github.com/b4nter waveshare neopixel (WS2812B) hat working on my PI 2B with the C test program and Python examples.
We couldn't get it to work on his Raspberry Pi 4 Model B Rev 1.1. Even after updates to the code for hardware detection of that model.
Oddly enough I have quite a lot of SMART strip WS2182B LEDs and I tried them with the C test program and they work on both PI's fine. However I only use a test strip of 4 pixels. Perhaps there is a timing issue.
We tested different forced timings in the code. We also tested only lighting one pixel. None of this worked for the PI4 and that hat.
I will try measure the PWM rates of each board with the test program and a scope. However I need to aquire or borrow a PI4 for that.
PI4 is using official PSU. Pictures of the test strip that did work on PI4
[image: image] https://user-images.githubusercontent.com/15659301/73136651-d9411c00-4047-11ea-8947-77fa64eabf32.png
[image: image] https://user-images.githubusercontent.com/15659301/73136661-f118a000-4047-11ea-92f1-8ab7b05c7a42.png
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/jgarff/rpi_ws281x/issues/372?email_source=notifications&email_token=ACB55G5ERJYRKPEAD6SQINDQ7WM3VA5CNFSM4JBAHKZ2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEJ5VB2Y#issuecomment-578506987, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACB55G6KWC6SJ6SIL32H7UDQ7WM3VANCNFSM4JBAHKZQ .
2E0PGS
commented
4 years ago The board is designed for PI use. It even has examples using this library.
I am not sure if it has logic level converters on board. It looks like does not on my physical inspection.
I just did a bit of googling and found its schematic: https://www.waveshare.com/w/upload/b/bf/RGB_LED_HAT_Schematic_.pdf
No transister logic level converters. I can check the logic levels on a scope later. Perhaps the PI2 is a slightly higher 3.3V due to less power consumption of onboard components.
The LED tapes I am using are not using logic level converters however they work fine on both devices so ignore those. We know they work.
jgarff
commented
4 years ago Yeah, that board clearly needs a level shifter and was designed without looking at the specs for either the ws281x or Pi. The highest output of a Pi I/O is ~3.6, and nominal is 3.3, but can be lower. The ws2812 is Vdd(5v) - 0.7, or 4.3V for a logic level high. So this board is clearly out of spec. Most people are lucky in that it works, but chip to chip variation, temperature, etc., is a problem.
See the wiki page on the project for a example Pi hat that I designed which includes a level shifter.
On Sun, Jan 26, 2020 at 11:06 AM Peter Stevenson notifications@github.com wrote:
The board is designed for PI use. It even has examples using this library.
I am not sure if it has logic level converters on board. It looks like does not on my physical inspection.
I just did a bit of googling and found its schematic: https://www.waveshare.com/w/upload/b/bf/RGB_LED_HAT_Schematic_.pdf
No transister logic level converters. I can check the logic levels on a scope later. Perhaps the PI2 is a slightly higher 3.3V due to less power consumption of onboard components.
The LED tapes I am using are not using logic level converters however they work fine on both devices so ignore those. We know they work.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/jgarff/rpi_ws281x/issues/372?email_source=notifications&email_token=ACB55GYT2IBM2PBVAEEQRSLQ7XGKVA5CNFSM4JBAHKZ2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEJ52BSY#issuecomment-578527435, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACB55G6PFB3FYOPKYMJKGYLQ7XGKVANCNFSM4JBAHKZQ .
2E0PGS
commented
4 years ago Yes originally the issue was missing support for his specific PI4 hardware revision detection. So there was a software issue there. That has since been merged.
But now we got this far it does indeed seem it's of poor design now I found the shematic which I didn't find before.
I can build a logic level converter and/or still check we have no timing issues since that was a problem at one point in time. Then we can put this to bed.
I think @b4nter said the board is no longer listed for sale? Is that correct @b4nter ?
2E0PGS
commented
4 years ago Anyway,
Thank you for the reply and comment's.
Logic level is something I checked loosely with a meter before, however I haven't had time to hook up a scope to it. PWM frequency would effect my testing previously given the hardware detection I mentioned before.
Logic level probably is the cause now we gotten past previous issues. The issue is trying to make it work. Either he buys a PI2 or we try a different PSU and hope the tolerances fit.
Or he buys a new board as you suggested.
I'll get it working one way or another and report back just for closure.
Cheers,
-- Peter Stevenson (2E0PGS) http://www.m3pgs.co.uk
On 26 Jan 2020, 18:25, at 18:25, Jeremy Garff notifications@github.com wrote:
Yeah, that board clearly needs a level shifter and was designed without looking at the specs for either the ws281x or Pi. The highest output of a Pi I/O is ~3.6, and nominal is 3.3, but can be lower. The ws2812 is Vdd(5v) - 0.7, or 4.3V for a logic level high. So this board is clearly out of spec. Most people are lucky in that it works, but chip to chip variation, temperature, etc., is a problem.
See the wiki page on the project for a example Pi hat that I designed which includes a level shifter.
On Sun, Jan 26, 2020 at 11:06 AM Peter Stevenson notifications@github.com wrote:
The board is designed for PI use. It even has examples using this library.
I am not sure if it has logic level converters on board. It looks like does not on my physical inspection.
I just did a bit of googling and found its schematic: https://www.waveshare.com/w/upload/b/bf/RGB_LED_HAT_Schematic_.pdf
No transister logic level converters. I can check the logic levels on a scope later. Perhaps the PI2 is a slightly higher 3.3V due to less power consumption of onboard components.
The LED tapes I am using are not using logic level converters however they work fine on both devices so ignore those. We know they work.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub
https://github.com/notifications/unsubscribe-auth/ACB55G6PFB3FYOPKYMJKGYLQ7XGKVANCNFSM4JBAHKZQ .
-- You are receiving this because you commented. Reply to this email directly or view it on GitHub: https://github.com/jgarff/rpi_ws281x/issues/372#issuecomment-578528951
jgarff
commented
4 years ago Sounds good. I wouldn't rely on getting a Pi2 to solve the problem though. It may work, but may not. Chances are probably it will just based on the number of people who find success. Just remember, it's not the Pi4 at fault. A Pi2 could also not product 3v6 (or 3v3) on the I/O and still be within spec of the Pi. Any use of any Pi with any WS281X without a level shifter is out of spec. Getting another Pi to solve the problem is just rolling the dice. Also keep in mind that even if a Pi works today, it might not tomorrow. Remember temp/voltage/power supply/etc., also are a factor.
On Sun, Jan 26, 2020 at 1:01 PM Peter Stevenson notifications@github.com wrote:
Anyway,
Thank you for the reply and comment's.
Logic level is something I checked loosely with a meter before, however I haven't had time to hook up a scope to it. PWM frequency would effect my testing previously given the hardware detection I mentioned before.
Logic level probably is the cause now we gotten past previous issues. The issue is trying to make it work. Either he buys a PI2 or we try a different PSU and hope the tolerances fit.
Or he buys a new board as you suggested.
I'll get it working one way or another and report back just for closure.
Cheers,
-- Peter Stevenson (2E0PGS) http://www.m3pgs.co.uk
On 26 Jan 2020, 18:25, at 18:25, Jeremy Garff notifications@github.com wrote:
Yeah, that board clearly needs a level shifter and was designed without looking at the specs for either the ws281x or Pi. The highest output of a Pi I/O is ~3.6, and nominal is 3.3, but can be lower. The ws2812 is Vdd(5v) - 0.7, or 4.3V for a logic level high. So this board is clearly out of spec. Most people are lucky in that it works, but chip to chip variation, temperature, etc., is a problem.
See the wiki page on the project for a example Pi hat that I designed which includes a level shifter.
On Sun, Jan 26, 2020 at 11:06 AM Peter Stevenson notifications@github.com wrote:
The board is designed for PI use. It even has examples using this library.
I am not sure if it has logic level converters on board. It looks like does not on my physical inspection.
I just did a bit of googling and found its schematic: https://www.waveshare.com/w/upload/b/bf/RGB_LED_HAT_Schematic_.pdf
No transister logic level converters. I can check the logic levels on a scope later. Perhaps the PI2 is a slightly higher 3.3V due to less power consumption of onboard components.
The LED tapes I am using are not using logic level converters however they work fine on both devices so ignore those. We know they work.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub
< https://github.com/notifications/unsubscribe-auth/ACB55G6PFB3FYOPKYMJKGYLQ7XGKVANCNFSM4JBAHKZQ
.
-- You are receiving this because you commented. Reply to this email directly or view it on GitHub: https://github.com/jgarff/rpi_ws281x/issues/372#issuecomment-578528951
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/jgarff/rpi_ws281x/issues/372?email_source=notifications&email_token=ACB55GY6UZW56J47T27OOV3Q7XT27A5CNFSM4JBAHKZ2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEJ54PQA#issuecomment-578537408, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACB55GZHEB3TPSV4RK5CRWTQ7XT27ANCNFSM4JBAHKZQ .
2E0PGS
commented
4 years ago Yeah I didn't mention that as it's a given.
Well aware.
-- Peter Stevenson (2E0PGS) http://www.m3pgs.co.uk
On 26 Jan 2020, 20:14, at 20:14, Jeremy Garff notifications@github.com wrote:
Sounds good. I wouldn't rely on getting a Pi2 to solve the problem though. It may work, but may not. Chances are probably it will just based on the number of people who find success. Just remember, it's not the Pi4 at fault. A Pi2 could also not product 3v6 (or 3v3) on the I/O and still be within spec of the Pi. Any use of any Pi with any WS281X without a level shifter is out of spec. Getting another Pi to solve the problem is just rolling the dice. Also keep in mind that even if a Pi works today, it might not tomorrow. Remember temp/voltage/power supply/etc., also are a factor.
On Sun, Jan 26, 2020 at 1:01 PM Peter Stevenson notifications@github.com wrote:
Anyway,
Thank you for the reply and comment's.
Logic level is something I checked loosely with a meter before, however I haven't had time to hook up a scope to it. PWM frequency would effect my testing previously given the hardware detection I mentioned before.
Logic level probably is the cause now we gotten past previous issues. The issue is trying to make it work. Either he buys a PI2 or we try a different PSU and hope the tolerances fit.
Or he buys a new board as you suggested.
I'll get it working one way or another and report back just for closure.
Cheers,
-- Peter Stevenson (2E0PGS) http://www.m3pgs.co.uk
On 26 Jan 2020, 18:25, at 18:25, Jeremy Garff notifications@github.com wrote:
Yeah, that board clearly needs a level shifter and was designed without looking at the specs for either the ws281x or Pi. The highest output of a Pi I/O is ~3.6, and nominal is 3.3, but can be lower. The ws2812 is Vdd(5v) - 0.7, or 4.3V for a logic level high. So this board is clearly out of spec. Most people are lucky in that it works, but chip to chip variation, temperature, etc., is a problem.
See the wiki page on the project for a example Pi hat that I designed which includes a level shifter.
On Sun, Jan 26, 2020 at 11:06 AM Peter Stevenson notifications@github.com wrote:
The board is designed for PI use. It even has examples using this library.
I am not sure if it has logic level converters on board. It looks like does not on my physical inspection.
I just did a bit of googling and found its schematic:
https://www.waveshare.com/w/upload/b/bf/RGB_LED_HAT_Schematic_.pdf
No transister logic level converters. I can check the logic levels on a scope later. Perhaps the PI2 is a slightly higher 3.3V due to less power consumption of onboard components.
The LED tapes I am using are not using logic level converters however they work fine on both devices so ignore those. We know they work.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub
<
,
or unsubscribe
<
https://github.com/notifications/unsubscribe-auth/ACB55G6PFB3FYOPKYMJKGYLQ7XGKVANCNFSM4JBAHKZQ
.
-- You are receiving this because you commented. Reply to this email directly or view it on GitHub:
https://github.com/jgarff/rpi_ws281x/issues/372#issuecomment-578528951
— You are receiving this because you commented. Reply to this email directly, view it on GitHub
https://github.com/notifications/unsubscribe-auth/ACB55GZHEB3TPSV4RK5CRWTQ7XT27ANCNFSM4JBAHKZQ .
-- You are receiving this because you commented. Reply to this email directly or view it on GitHub: https://github.com/jgarff/rpi_ws281x/issues/372#issuecomment-578538427
jgarff
commented
4 years ago Sounds good. Good luck.
On Sun, Jan 26, 2020 at 1:22 PM Peter Stevenson notifications@github.com wrote:
Yeah I didn't mention that as it's a given.
Well aware.
-- Peter Stevenson (2E0PGS) http://www.m3pgs.co.uk
On 26 Jan 2020, 20:14, at 20:14, Jeremy Garff notifications@github.com wrote:
Sounds good. I wouldn't rely on getting a Pi2 to solve the problem though. It may work, but may not. Chances are probably it will just based on the number of people who find success. Just remember, it's not the Pi4 at fault. A Pi2 could also not product 3v6 (or 3v3) on the I/O and still be within spec of the Pi. Any use of any Pi with any WS281X without a level shifter is out of spec. Getting another Pi to solve the problem is just rolling the dice. Also keep in mind that even if a Pi works today, it might not tomorrow. Remember temp/voltage/power supply/etc., also are a factor.
On Sun, Jan 26, 2020 at 1:01 PM Peter Stevenson notifications@github.com wrote:
Anyway,
Thank you for the reply and comment's.
Logic level is something I checked loosely with a meter before, however I haven't had time to hook up a scope to it. PWM frequency would effect my testing previously given the hardware detection I mentioned before.
Logic level probably is the cause now we gotten past previous issues. The issue is trying to make it work. Either he buys a PI2 or we try a different PSU and hope the tolerances fit.
Or he buys a new board as you suggested.
I'll get it working one way or another and report back just for closure.
Cheers,
-- Peter Stevenson (2E0PGS) http://www.m3pgs.co.uk
On 26 Jan 2020, 18:25, at 18:25, Jeremy Garff notifications@github.com wrote:
Yeah, that board clearly needs a level shifter and was designed without looking at the specs for either the ws281x or Pi. The highest output of a Pi I/O is ~3.6, and nominal is 3.3, but can be lower. The ws2812 is Vdd(5v) - 0.7, or 4.3V for a logic level high. So this board is clearly out of spec. Most people are lucky in that it works, but chip to chip variation, temperature, etc., is a problem.
See the wiki page on the project for a example Pi hat that I designed which includes a level shifter.
On Sun, Jan 26, 2020 at 11:06 AM Peter Stevenson notifications@github.com wrote:
The board is designed for PI use. It even has examples using this library.
I am not sure if it has logic level converters on board. It looks like does not on my physical inspection.
I just did a bit of googling and found its schematic:
https://www.waveshare.com/w/upload/b/bf/RGB_LED_HAT_Schematic_.pdf
No transister logic level converters. I can check the logic levels on a scope later. Perhaps the PI2 is a slightly higher 3.3V due to less power consumption of onboard components.
The LED tapes I am using are not using logic level converters however they work fine on both devices so ignore those. We know they work.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub
<
,
or unsubscribe
<
https://github.com/notifications/unsubscribe-auth/ACB55G6PFB3FYOPKYMJKGYLQ7XGKVANCNFSM4JBAHKZQ
.
-- You are receiving this because you commented. Reply to this email directly or view it on GitHub:
https://github.com/jgarff/rpi_ws281x/issues/372#issuecomment-578528951
— You are receiving this because you commented. Reply to this email directly, view it on GitHub
< https://github.com/notifications/unsubscribe-auth/ACB55GZHEB3TPSV4RK5CRWTQ7XT27ANCNFSM4JBAHKZQ
.
-- You are receiving this because you commented. Reply to this email directly or view it on GitHub: https://github.com/jgarff/rpi_ws281x/issues/372#issuecomment-578538427
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/jgarff/rpi_ws281x/issues/372?email_source=notifications&email_token=ACB55G5S3H32PVRHVB2K4A3Q7XWHFA5CNFSM4JBAHKZ2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEJ545PA#issuecomment-578539196, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACB55GYOV7KBXHMMZ3MERHDQ7XWHFANCNFSM4JBAHKZQ .
penfold42
commented
4 years ago The ws2812 is Vdd(5v) - 0.7, or 4.3V for a logic level high.
I thought it was 0.7 * Vdd = 3.5v
2E0PGS
commented
4 years ago I personally assume they're expecting a decent signal given its 5v. So I would consider it lucky to have it working on a PI at all given its logic level 3.3v nominal.
However if you happen to know the specific bit bang high threshold is 0.7 then it calcs to
$ bc -l
0.7 * 5
3.5Oh good catch penfold, thanks. I believe you're right. I think the problem is still that we're out of spec, but the margins are much less than I had originally described, which probably describes why it works most of the time.
Thanks,
Jeremy
On Sun, Jan 26, 2020 at 4:10 PM penfold42 notifications@github.com wrote:
The ws2812 is Vdd(5v) - 0.7, or 4.3V for a logic level high.
I thought it was 0.7 * Vdd = 3.5v
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/jgarff/rpi_ws281x/issues/372?email_source=notifications&email_token=ACB55GYNYQM2VKKT2SBOBATQ7YJ7VA5CNFSM4JBAHKZ2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEJ6AJTA#issuecomment-578553036, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACB55G4EWH3AHKS6AQBWYV3Q7YJ7VANCNFSM4JBAHKZQ .
2E0PGS
commented
4 years ago I personally have ran I2C between PI and Ardunio without logic level converters with sucess.
Given I only did it one direction. Bi-directional would destroy the PI GPIO.
Anyway that was with a PI2 B and a PI Zero W.
So I suspect with the added onboard power consumption of a more powerful CPU of the PI4 and different power circuitry (USB C) it probably has a different GPIO logic level in real world measurements.
2E0PGS
commented
4 years ago So please let me scope it out to see what the real numbers are and differences to prove why it works on a PI 2B and not a PI 4. Regardless of it's debatable design.
jgarff
commented
4 years ago I2c is a different animal entirely. Pull-ups are present on the bus to drive the signal high. Each entity then pulls it low. Additionally, and arduino (typically AVR or such) has protection diodes, which drops the io level to vcc + ~0.7. I'd just be careful here making this comparison. It's not apples to apples. FWIW, I still believe you're on the margins of functionality and this is still very likely a voltage level issue.
On Sun, Jan 26, 2020 at 4:20 PM Peter Stevenson notifications@github.com wrote:
So please let me scope it out to see what the real numbers are and differences to prove why it works on a PI 2B and not a PI 4. Regardless of it's debatable design.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/jgarff/rpi_ws281x/issues/372?email_source=notifications&email_token=ACB55GYBOKTT4LNX5NAKYNDQ7YLENA5CNFSM4JBAHKZ2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEJ6APSA#issuecomment-578553800, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACB55GYBEMLS2MJ7HGNUC23Q7YLENANCNFSM4JBAHKZQ .
2E0PGS
commented
4 years ago I wouldn't rely on a built-in input protection for that. Opto-isolated or transistor based logic level converter, external protection diodes or external resistors is really the only way to not blow it up. However I was referring to PI as the master and Arduino as a slave device so it's safe. Otherway around PI needs protection.
The I2C slave requires a logic level high of 0.6 or 0.7 * VCC so 5V for Arduino is around 3.3v. So it's similair situation having a PI 2B supplying a marginal but working logic level high to the slave device. Like my LEDs expecting 3.5v working on PI2 2B.
However let's not get off topic.
2E0PGS
commented
4 years ago 
Both worked when hooked up via breadboard to those specific pins. So I suspect the hat uses a different pinout (uses a different ground) which is faulty pin on @b4nter 's PI 4. They also both worked on 3.3V supply too.
2E0PGS
commented
3 years ago Hi all, I finally bought a Pi 4 B 4 GB just to confirm there was something wrong with the pins on @b4nter Pi. The above scope tests already proved the logic levels and frequency are the same between my working Pi 2 B and his Pi 4 B 4 GB so here I am just further confirming it with a new board. I am using the original SD card from many months ago to rule out software configuration or version differences.
$ cat /proc/cpuinfo
...
Hardware : BCM2835
Revision : c03112
...
Model : Raspberry Pi 4 Model B Rev 1.2Some .bash_history for others sake
sudo apt install git
git clone https://github.com/jgarff/rpi_ws281x.git
cd rpi_ws281x/
sudo apt install scons~/rpi_ws281x $ git log
commit 6851d9fb090f8a4703d2ceac97da2de617b09e8d (HEAD -> master, origin/master, origin/HEAD)
Author: Stephen Warren <swarren@wwwdotorg.org>
Date: Tue Aug 27 09:24:18 2019 -0600~/rpi_ws281x $ git status
On branch master
Your branch is up to date with 'origin/master'.
Changes not staged for commit:
(use "git add <file>..." to update what will be committed)
(use "git checkout -- <file>..." to discard changes in working directory)
modified: main.c
modified: rpihw.c
no changes added to commit (use "git add" and/or "git commit -a")~/rpi_ws281x $ git diff
diff --git a/main.c b/main.c
index f88414a..e0d0894 100644
--- a/main.c
+++ b/main.c
@@ -59,8 +59,8 @@ static char VERSION[] = "XX.YY.ZZ";
#define TARGET_FREQ WS2811_TARGET_FREQ
#define GPIO_PIN 18
#define DMA 10
-//#define STRIP_TYPE WS2811_STRIP_RGB // WS2812/SK6812RGB integrated chip+leds
-#define STRIP_TYPE WS2811_STRIP_GBR // WS2812/SK6812RGB integrated chip+leds
+#define STRIP_TYPE WS2811_STRIP_RGB // WS2812/SK6812RGB integrated chip+leds
+//#define STRIP_TYPE WS2811_STRIP_GBR // WS2812/SK6812RGB integrated chip+leds
//#define STRIP_TYPE SK6812_STRIP_RGBW // SK6812RGBW (NOT SK6812RGB)
#define WIDTH 8
diff --git a/rpihw.c b/rpihw.c
index 941437c..fd77818 100644
--- a/rpihw.c
+++ b/rpihw.c
@@ -70,7 +70,7 @@ static const rpi_hw_t rpi_hw_info[] = {
.desc = "Pi 4 Model B - 2GB"
},
{
- .hwver = 0xC03111,
+ .hwver = 0xC03112,
.type = RPI_HWVER_TYPE_PI4,
.periph_base = PERIPH_BASE_RPI4,
.videocore_base = VIDEOCORE_BASE_RPI2,~/rpi_ws281x $ scons
~/rpi_ws281x $ sudo ./test
All working.
Got this model https://www.waveshare.com/wiki/RGB_LED_HAT
This RGB_LED_HAT does not respond with Raspberry Pi 4 Model B Rev 1.1 after running test program.
I can confirm LED HAT itself is fully working as I tested it on my colleagues Raspberry Pi 2 with the same software and it was working all fine.
Have tried to run on different GPIO's no luck with that.
Can't tell for sure, but it might be missing some of hardware detection in file file rpihw.c
SDRAM for this model is 0x1800000 got it by running bcm_host_get_sdram_address() reference: https://www.raspberrypi.org/documentation/hardware/raspberrypi/peripheral_addresses.md
Tried to change some variables around, but still can't get it working on my RPI4. Was wondering if anyone else tried to get this RGB_LED_HAT working on RPI4. Or someone more experienced can have a look at this issue.