Sell the D3d?

[pro3d]

This is for a CoreXY

I am looking for motivation to try another round of running the Duet Wifi currently on my desk wondering if I want another go or to sell it

How can I get this running with the features I want without using days reading and figuring out macros to write?

My last take on this was a struggle for me and I want to keep it simple if doing it over

Features I struggled with:

• Sensorless homing using stall detect causing false positives I did not like - should I just use sensors?
• Very bad prints probably due to lack of tables for the NTC 3950 thermistor - Is using a temp sensor only good solution?
• 3.3V logic all made a big headache as neither my Z probe or the otical endstops was sure to work with this - I did at last upgrade the probe but not sure what to do for endstops - What options non mechanical?
• A lot of mess with the G29 S2/M561 not working or having to be called in different macros to clear mesh for probing Z0
• Writing a bunch of macros ending up having no idea what they did or how to get them right without using all day researching

My main issue was all the time searching for info on how to write - then write the macros to get the config I am used to from Marlin - Like a simple thing as filament change

I never got to the point where I was satisfied with a simple RRF config that I could push out to other machines so I went back to a $20 controller and are now happy with everything except some noise and some lack of torque

I feel when investing in a $200 controller it should give me pleasure not pain and would like to get to this point or sell it

Thanks for any corexy user input
Change this again?

For this?

[Phaedrux]

Sensorless homing is difficult to get tuned right on a CoreXY. I would just go with switches. It's doable, but it depends on a lot of factors and would maybe require more mucking about than you're after. That's just due to the nature of matching the settings to the specific motors.

A thermistor with known values (like the Semitek E3D cartridge) would work or a PT1000 sensor.

Choice of z probe is up to you, each will have some trade offs and will need to be configured and calibrated. If your bed is flat and solid you can avoid the probe entirely and just use an endstop here as well.

There are optical endstops that are 3.3v tolerant and do work fine if you don't want to add a resistor.

As for a working config, would it help to see an example config set for a corexy to modify from?

Alternatively, if you provide the mechanical and electrical details and a starting config set from the configurator I would happily modify it for your setup.

[pro3d]

@phaedrux

Thanks for wanting to help - The config file it self is not my main pain but all the other macro stuff modifying it to a simple one heater corexy setup with all basic functions

I might have tried to hard to modify it to the high standards I thought I would get out of the box and ended up messing it up - Probably should lower the amount of features I want

Is the G29 S2/M561 issue taken care of now in the latest release? So bed mesh gets disabled properly?

I might take you up on this offer but that is after some assembly of the next build

[Phaedrux]

@pro3d said in Sell the D3d?:

Is the G29 S2/M561 issue taken care of now in the latest release? So bed mesh gets disabled properly?

I believe it is resolved, but I haven't modified my homing files to test it. I still have the work around in place to clear the map twice before homing and then reload it after. I should experiment.

[dragonn]

@pro3d said in Sell the D3d?:

Sensorless homing using stall detect causing false positives I did not like - should I just use sensors?

Yes

@pro3d said in Sell the D3d?:

3.3V logic all made a big headache as neither my Z probe or the otical endstops was sure to work with this - I did at last upgrade the probe but not sure what to do for endstops - What options non mechanical?

Did you ever hear about an voltage divider? End BTW, my optical endstops works fine with 3.3V without any modification.

[dragonn]

@pro3d said in Sell the D3d?:

Very bad prints probably due to lack of tables for the NTC 3950 thermistor - Is using a temp sensor only good solution?

Try that parameters R4700 T100000 B3950, no thermistor table is needed.

[pro3d]

@dragonn

Tried this first - got massive misstemp - got told to buy P1000 witch I now have

Voltage divider? No not really and I dont want to hack it into something useful I just want something that works as good as my Marlin 8 bit setup or better - this did not

[fcwilt]

@pro3d said in Sell the D3d?:

@dragonn

Tried this first - got massive misstemp - got told to buy P1000 witch I now have

Voltage divider? No not really and I dont want to hack it into something useful I just want something that works as good as my Marlin 8 bit setup or better - this did not

There is nothing better than a Duet setup.

There are differences between them and the inexpensive Arduino based boards but you knew that before you purchased the Duet - correct?

My "active" endstops work fine on 3.3. My Z-probes work fine on 3.3.

What devices are you using that will not work on 3.3?

Frederick

[bartolomeus]

I have sensorless homing working very well on my corexy (the one Roy designed). You just need to spend some time in tuning it to get the right parameters. The wiki explains everything quite well. The only reason I wouldn't use it, is if I needed resume after power loss. But that is also described in the wiki.

[pro3d]

@fcwilt

Hi Fredric

I had a 3d touch not ready for 3.3 - I did buy a new one witch supports this though. I also have optical endstops and it was said not all worked with 3.3v

Well the inexpensive arduino boards carrying Marlin gives me all I want of features without writing a bunch of macros I dont have time for just to configure simple features

The only thing the duet gives me more from my point of view is stronger drivers and quieter drivers witch is fine if I can get it running with basic features without to much work

I need some motivation to dig into this a second time and not sure it is for me at this point

[pro3d]

@bartolomeus

I dont have time to read or fine tune the stalldetect I am sorry

I did get it to work with your settings but experiencing some random false positives and more than one is to much as I dont know when it will happen again so if this is not more stable I want endstops

[bartolomeus]

I was actually referring to the others advising again at the use of sensorless homing on a corexy. I really don't see the problem.

@pro3d I think that is part of the problem, not willing to spend some time into the board. A basic configuration isn't really that difficult. You probably spent time the first time using Marlin and learning how to compile and flash with arduino, and now you are used to it. It takes a little/the same effort to understand the Duet. If you can't figure things out, there is always a great community behind the duet, willing to help.

[pro3d]

@bartolomeus

I think you are right about me not using the time necessary

To get to that point I would like to see clearly what the d3d/rrf solution can give me more? Then use the time to get there - So I need motivation to dig into the RRF hackerspace ones more... (lack of atm)

Probably just listing it if I dont se a good reason to put in the effort

[whosrdaddy]

Sell the board, you are clearly not able to change your "Marlin" mindset.
Nothing to see here, carry on...

[dragonn]

I am pretty sure if you first board would be Duet you would have that problems.
I agree, sell it. If you don't have time don't change something witch is already fine working.

[fcwilt]

@pro3d said in Sell the D3d?:

I need some motivation to dig into this a second time and not sure it is for me at this point

You mentioned "macros". There are really not that many and they tend to be small and simple.

I started with 8 bit Arduinos, I then went with Smoothies and they were a step in the right direction but once I discovered Duets I converted all of my printers and never looked back.

Perhaps you attempted to do a complete configuration as opposed to a step-by-step configuration, testing each "step" as you go.

My config.g file consists of 36 lines. Then there are the 4 "homing" files that total 23 lines but they are mostly the same with small variations.

Those are the essential ones to get your printer up and running.

Once you have the "gestalt aha moment" all will be clear and smooth sailing.

Hang in there.

Frederick

[pro3d]

@fcwilt

I guess if i skip the stall detect it would be a much smoother ride. I did have a lot of problem with the BL touch and corner bed leveling + bed mesh leveling and simple filament change with pause and park function

I did want to have at least as many features as from my Marlin config and is just trying to get motivated to give it another go

I know the board is good and it is RRF I struggle with. I wonder what RRF can give me more in an easy way - To me it looks a little unfinished

I will try to hang in there - I have a printer on the way so it might go there in a week or so

[fcwilt]

@pro3d said in Sell the D3d?:

@fcwilt

I guess if i skip the stall detect it would be a much smoother ride. I did have a lot of problem with the BL touch and corner bed leveling + bed mesh leveling and simple filament change with pause and park function

Does you printer have bed leveling screws that you adjust manually?

On my FolgerTech FT5 I have them and have not found the need for a Z-probe or the "bed leveling" features of the Duet firmware.

I tested with a simple Z-probe but it told be what I already knew - that the bed was flat and level.

On a delta, yes you may want those features but the hardest part there is fitting the Z-probe. Once that is done in a suitable fashion the macros needed should be quite simple.

On my deltas I have yet to fit Z-probes and have just used the manual approach using a dial indicator but it can be done with a thickness gauge or a "piece of paper".

Frederick

[dc42]

@pro3d said in Sell the D3d?:

I guess if i skip the stall detect it would be a much smoother ride.

Yes, to use stall detection homing you need to find a good combination of stall detection sensitivity and motor current during homing that works with your motors and mechanics - unless you buy a genuine Prusa i3 Mk3, in which case that work has been done for you. It's not hard to tune the stall detection parameters, but it's probably easier for a beginner to use endstop switches because they don't need tuning.