DotStar not working from 3.4.1

dvmourik

When running 3.3 on MB6HC the DotStar leds (we use only white DotStar) are working perect with M150 X0 S10 P255
Updated to 3.4.1 they stopped working with any command, back to 3.3 makes it working again.

After some step by step updates we found out our system stops working from 3.4.beta6

Only changelog for DotStar is this in the beta6:
[Duet 3 MB6HC] With DotStar LED strip. If the brightness M150 P parameter (brightness) was not a multiple of 8 then the blue LED level was incorrect

dc42

@dvmourik said in DotStar not working from 3.4.1:

we use only white DotStar

Cam you explain what you mean by this? Is there a white-only version of DotStar that I don't know about; or are you using a regular DotStar strip but only setting the colours to White?

achrn

@dc42 There are white-only DotStar: https://www.adafruit.com/product/2434 https://www.adafruit.com/product/2437

I've contemplated using it for something, but not actually laid my hands on any. Adafruit say it's just like ordinary DotStar but with three white LEDs, not a R, G & B.

dvmourik

@dc42 There is a white only dotStar which we use in our machines. We have large machines and can't use others as they are limited to number of leds. And we need at least white (color would be extra).

So there is APA102 (DotStar White) available which we use.

We only use P value to set the brightness level (works up to 3.4 beta 5)

dc42

@dvmourik thanks for clarifying this.

If the red, green and blue LEDs are replaced by white LEDs, surely it follows that with zero specified for the red, green and blue intensities, all the LEDs will be off? Please try providing nonzero R, U and B parameters. I think your LEDs were only lighting up in older versions of RRF because of the bug that we fixed in 3.4.

dvmourik

@dc42 said in DotStar not working from 3.4.1:

rameters. I think your LEDs were only lighting up in older versions of RRF because of the bug that we fixed in 3.4.

That worked out! A little more code to just dim the strip but we can dim it better than with the P value (that stops at 10 and R U and B goes down to 1)