tekkydave

I would think for a bed that size 6mm thick would be best. Its going to be mighty heavy though. Have you considered using leadscrews to drive the z axis? I'm in the process of upgrading my DBot to triple screws driven by a single motor. The v-slot with wheels at 4 corners keeps the heavy bed (300x300x6mm) constrained well in the x-y plane but 3 screws are needed to keep it going up/down without tilting or twisting. At the moment I have 2 screws on L & R of the bed. As it changes direction either the front of the bed or back starts moving first. Probably due to the difficulty of having 2 screws at exactly the c.o.g. of the z carriage/bed combination. Three screws in a triangle arrangement will be a better system.

Might be a good idea to do initial build & test using a cheap RAMPS before hooking up the Duet

Screenshot from my Samsung tablet of the Duet Web Control with the image from Octoprint (running on a RPi3) embedded.

Yes, but via my RPi3 running octoprint as a temporary measure. The Duet doesn't have built-in webcam connectivity but will connect via an ip-webcam or similar. There are instructions on the Duet Wiki on setting up a normal (usb) webcam as an ip-cam using an Rpi zero (or an Rpi2,3 etc).

Resistance is futile...

Yep. My D-Bot is CoreXY. Have a look at The Duet Wiki for all the options. Everything is configured via g-code so no more Marlin Configuration.h and uploading the firmware for minor changes. Most things can be configured on-the-fly and you can define your own g-code macro routines in macro files.

The only place you can but it from in the UK is direct from Duet3D, unless anyone else knows otherwise so no options on price. It's light years ahead of RAMPS so well worth the price. The steppers are virtually silent now due to the 256x microsteps. The recommended mode is 16x interpolated to 256x. Mind you it makes you very very careful when connecting things up - I'm a bit OCD about checking connectors due to my BBC background but even I'm checking more than usual.

Today I have managed to set up the Piezo probe correctly and can home all the axes (G28). I've also set up the mesh levelling (G29) and then tuned the bed & extruder heater parameters. I tried a quick 20mm test cube and it printed perfectly. I love the Duet Web Control interface - it's more slick than Octoprint and can also display your mesh compensation as a graphic. You can also edit the Reprap Firmware (RRF) config directly without having to re-upload your Arduino code. I'm still learning about RRF but it's all good so far. The weird slope on this one is due to the fact that my bed is on 2 leadscrews so rocks slightly when moved up/down. Once I get the extra screw installed it should improve greatly. I have ordered some better motors - 55Nm as opposed to the 36Nm ones I harvested from the Prusa. Next project is the 3 leadscrew mods.

Got a stinking cold at the moment so a good excuse to lock myself away with the Duet. Yesterday I set up the WiFi connection and basic configuration. Today has been mostly crimping. The basic axis motion is working well and I'm amazed how quiet it is. I thought others were exaggerating about this but its true. I need to get the piezo z probe configured before I can try a print. A job for tomorrow...

I'm also planning to upgrade to 3 leadscrews linked by belts with a single motor. I have all the parts. I will have to rearrange/add to the cross 2020s in the base which will affect how I eventually lay out the electronics.

That looks good. Why not post it on Thingiverse - lots of people would find it useful. I was planning to make a case with a fan beneath the pcb where duet recommend the cooling is applied.

Santa delivered

I'm not sure if this is relavant to your dodgy nanos but I read the following in the official D-Bot build guide: "If you cannot connect or if you receive a timeout error, you may need to download and install the CH341SER driver. Some Chinese knockoff Arduino boards use a serial chip which is not the same as normal Arduino boards"

What are the highest and lowest raw output values as you rotate the vane.

What size is ii (short int, long int etc)

I found that measuring the wind speed and direction more frequently than every 15s just produces a jittery output that moves too fast to follow. I measure every 15s and store in a database with an accurate timestamp. Then periodically retrieve all valid values for last 15 mins to do the concensus calculations for the web site. You will need to experiment with what averaging works best for how you want to display it.

I can reproduce this problem but only by selecting the wrong board type (Uno instead of Nano). I'm using IDE version 1.83 on Linux.

I did say I was clutching at Straws I doesn't make sense as you have swapped around all the main components; arduino, cable and PC/laptop (including IDE). Must be something else but IBIIK what.

Reinstall Arduino IDE? Clutching at straws now I'm afraid.

Do you have the corrrect board and port selected? Also - try a different USB lead.

I've adjusted the speed, acceleration and jerk settings in Marlin and it's a lot quieter

Results of M48 repeatability test: SENDING:M48 P5 X150 Y140 V4 M48 Z-Probe Repeatability Test Positioning the probe... Bed X: 150.000 Y: 140.000 Z: 0.031 1 of 5: z: 0.002 mean: 0.0019 sigma: 0.000000 min: 0.002 max: 0.002 range: 0.000 2 of 5: z: 0.007 mean: 0.0044 sigma: 0.002500 min: 0.002 max: 0.007 range: 0.005 3 of 5: z: 0.004 mean: 0.0044 sigma: 0.002041 min: 0.002 max: 0.007 range: 0.005 4 of 5: z: -0.005 mean: 0.0020 sigma: 0.004428 min: -0.005 max: 0.007 range: 0.012 5 of 5: z: 0.008 mean: 0.0031 sigma: 0.004524 min: -0.005 max: 0.008 range: 0.013 Finished! Mean: 0.003125 Min: -0.005 Max: 0.008 Range: 0.013 Standard Deviation: 0.004524

Removed the link - shouldn't have been there. It doesn't really bash into the bed, it just touches lightly. It sounds like it is but the noise is from the motors stopping. This printer is quite noisy I suppose. I think the motors I'm using are a bit under-powered. they originally came from my old Prusa.

Replaced the left hand pot on the pcb with a multiturn to make calibration easier. Also the piezo seems to have reverse polarisation so I flipped the connector on the rear and it works great now. I needed to switch to dual z stepper drivers to get the homing speed up to around 400mm/min to trigger it reliably. Homing on Youtube

I have started using a brim on every print. I started doing it when I was printing the DBot parts on my old Prusa in PETG. I carried on even with PLA and now just do it regardless.