DuetWifi 2 and LIS3DH wiring

jay_s_uk

@nick9one1 SCL is SCLK.
Don't forget power (either 3.3v or 5v)

dc42

@jay_s_uk said in DuetWifi 2 and LIS3DH wiring:

@nick9one1 SCL is SCLK.
Don't forget power (either 3.3v or 5v)

You must use 3.3V power.

SDA is indeed the same signal as SDO.

jay_s_uk

@dc42 why is 3.3v needed when the adafruit LIS3DH has an onboard voltage regulator?
The one I have here works fine with 5v

nick9one1

@dc42

did you mean to say SDA is the same as SDI?

There is already a pin marked SDO

dc42

@jay_s_uk said in DuetWifi 2 and LIS3DH wiring:

@dc42 why is 3.3v needed when the adafruit LIS3DH has an onboard voltage regulator?
The one I have here works fine with 5v

So that the i2C signals have 3.3V signal levels.

nick9one1

I've wired it as per this diagram;

added this to config.g

M955 P0 C"spi.cs3+spi.cs4"

enabled the accelerometer plugin and have run the gcode;

G1 X-50 G4 S2 M956 P124.0 S1000 A0 G4 P10 G1 X50 F20000

but get this error;

If I try

G1 X-50 G4 S2 M956 P0 S1000 A0 G4 P10 G1 X50 F20000

(P0 instead of P124.0) it reports "Error: M956: Accelerometer not found"

DIY-O-Sphere

@nick9one1
In your wiring diagram +3.3V is connected to the wrong Vin pin of LIS3DH.
And spi.cs3 + spi.cs4 are swapped.

sebkritikel

@nick9one1 @diy-o-sphere said in DuetWifi 2 and LIS3DH wiring:

@nick9one1
In your wiring diagram +3.3V is connected to the wrong Vin pin of LIS3DH.
And spi.cs3 + spi.cs4 are swapped.

Yup, that looks to be the case (checked my crappy wiring diagram)
https://forum.duet3d.com/topic/22878/software-package-3-3beta3-released/33?_=1621606715168

SIam

@nick9one1

@sebkritikel
Your test code is wrong

G1 X-50 G4 S2 M956 P124.0 S1000 A0 G4 P10 G1 X50 F20000

must be

G1 X-50 G4 S2 M956 P0 S1000 A0 G4 P10 G1 X50 F20000

nick9one1

@sebkritikel

thanks, swapped those pins and got further
this is the wiring diagram now

G1 X-50 G4 S2 M956 P0 S1000 A0 G4 P10 G1 X50 F20000

now generates a csv, but also gives an error..

csv here:
https://1drv.ms/u/s!AgN3RlWvEHCokZAkg_o_XZeWuFRT7Q?e=i2IFTZ

sebkritikel

@nick9one1 said in DuetWifi 2 and LIS3DH wiring:

@sebkritikel

thanks, swapped those pins and got further
this is the wiring diagram now
G1 X-50 G4 S2 M956 P0 S1000 A0 G4 P10 G1 X50 F20000
now generates a csv, but also gives an error..

csv here:
https://1drv.ms/u/s!AgN3RlWvEHCokZAkg_o_XZeWuFRT7Q?e=i2IFTZ

May be some useful info in these threads regarding your latest error (and the associated ‘garbage’ in the csv)

https://forum.duet3d.com/topic/23021/duet3-6hc-gcode-inside-accelerometer-results/7?_=1621647165826

https://forum.duet3d.com/topic/23013/is-there-a-way-to-debug-accelerometer-connection-issues/22?_=1621647372962

dc42

@nick9one1 see the five bullet points at the end of https://duet3d.dozuki.com/Wiki/Accelerometers#Section_Retrieving_the_data.

nick9one1

@dc42
thanks. I've tried a coupe of different cards with the same result.

The card I'm using is 100MB/s UHS class 1.

I've dropped the sample rate down to 100, and only for the X axis so the gcode looks like this;

G1 X-50 G4 S2 M956 X P0 S100 A0 G4 P10 G1 X50 F20000

the csv contains;
could it still be a wiring issue?

this is how its currently wired

dc42

@nick9one1 your post at 02:20 UK time in which you posted a link the the .csv file shows that it has successfully collected the first 39 samples. That suggests that the wiring was at least working some of the time.

I suggest you install RRF 3.3RC2 and reduce the SPI frequency to 2MHz or 1MHz. There is a M955 parameter to do that in RC2.

nick9one1

@dc42 thanks I'll try that next.

Earlier I did manage to run it again and get what looks like a complete csv of 100 samples with no errors.

https://1drv.ms/x/s!AgN3RlWvEHCokZAlu3J5zuYliA661w?e=bRuxIp

but it seems very temperamental. Sometimes giving me some results with errors, sometimes none.

nick9one1

@dc42

I've installed RC2 but cant see any documentation on how to set the new Q parameter?

dc42

@nick9one1 see https://duet3d.dozuki.com/Wiki/Gcode#Section_M955_Configure_Accelerometer. The default SPI frequency used to be 4MHz in 3.3beta3 but is 2MHz in 3.3RC2.

nuroo

@nick9one1
Thank you for posting your questions and asking for help interpreting the instruction for using this sensor with duet2 board. Some pinout diagrams differ in pin names then instructions. This made the instructions confusing to me.

"On Duet WiFi, Duet Ethernet and Duet Maestro all the SPI CS pins have interrupt capability, so you can use any two spare CS pins on the daughter board connector. For example, on a Duet WiFi or Duet Ethernet you could connect the accelerometer CS pin to SPI.CS3, and the INT1 pin to SPI.CS4. [Note: if you already have one daughter board plugged into the main board and you wish to connect to SPI.CS3 and SPI.CS4 on the top of that daughter board, the daughter board passes SPI.CS3 and SPI.CS4 to the pin positions directly above the SPI.CS1 and SPI.CS2 pins on the main board. This is so that a second daughter board stacked on top automatically uses SPI.CS3 and SPI.CS4 instead of SPI.CS1 and SPI.CS2.] Use this command to tell RRF about the accelerometer:"

Anyway your updated diagram is gold. Something like it should really be included in the instructions.
Until reprap has something as informative - these two videos using it with klipper helped me understand the concepts and practical uses of using an accelerometer on your 3d printer:
https://youtu.be/OoWQUcFimX8
https://youtu.be/IezqWVZZ_iI

Hopefully I can get this specific sensor by the time 3.4 is released.

nuroo

This post got me up and running on a railcore. Had to change some pins in config setup to use daughter board spi. Im getting data thru the plugin as well. Looking forward to rrf 3.4

mule

Just a quick info:
The picture of the adafruit LIS3DH on https://duet3d.dozuki.com/Wiki/Accelerometers is misleading because the V3.3 from the duet is connected to 3Vo (Output) instead of the VIN (Input) pin. Maybe it is better to remove it.