Smoothieware to Marlin 2.0 - My story on the Azteeg X5 GT

[@3dprinting221]

Morning, thanks for this video, searched high and low for something like this. Good luck with your channel.

[@LukesLaboratory]

Thanks for letting me know! I did the same when I tried to convert over, and just wound up stumbling through. If you have any questions, feel free to ask!

[@B0M0A0K]

Very interesting story, it's made me rethink my choices for a board upgrade. Thanks very much for posting.

[@LukesLaboratory]

No problem. Any other questions?

[@kho75]

I was in a process to replace my main board for my Rigidbot Big. I saw your video and notice you have Rigidbot Big and regular. Do you have any main broad beside Azteeg X5 GT because this board is out of stock in many place. I was having hard time to understand where to start and how it wire everything from main board back to the printer working again. I saw your video on Marlin 2.0. It was helping on firmware edit and update. If you have link where should I get all my parts, it will help a lot.

[@LukesLaboratory]

Hi! I would not recommend getting the Azteeg X5 GT over more popular and cheaper boards that are out today. A very common board is the SKR 1.4 (amzn.to/2S6OYe5) (affiliate link) which has very wide community acceptance and great capabilities for the price of $62 with drivers included. If you want something that works out of the box and is rather easy to use, a duet 2 wifi will probably serve you pretty well.

[@kho75]

@@LukesLaboratory Thank you so much When come to wire back together, how do I mod the heat bed cable? does Duet 2 using Marlin open source? Sorry to ask like a newbie on this. I just want to make sure I understand all the steps before order either board. When I use those board how much improve that I will get in term of print speed? I still using the stock broad that came with.

[@LukesLaboratory]

@@kho75 There is a guide on the rigidbot wiki on how to join together the heating wires to make one "big" wire (since its really just positive and negative leads with 2 wires seperated for the thermistor) the Duet 2 uses RRF, which i believe at this time is a superior firmware - it uses gcode to configure it, no compiling or any other nonsense, you just edit settings in a single text file (and can edit them on the fly) and away you go! When you get your new board you may be able to go faster, especially if you follow it up with hardware mods.

[@kho75]

@@LukesLaboratory Thank you so much to explain to me. Since yesterday I been watching a few video on those two boards setup. You are so helpful.

[@zdenekvalek1538]

I have some serious issues with Azteeg X5 GT as well. 1. The hotend path gets totally out of control when it gets to higher Y values mid print, which results in priter crashing to hardstops. At first I thought it was caused by using grid leveling and WCS, but it still happens without WCS. 2. The printer freezes when starting print at beginning of first line (when it touches build plate). I found that if I change feedrate to faster, the issue is gone and print completes successfully. Now I am completely lost, because there is no clear information on the firmware running the Azteeg. Comments here are the best information source I found so far and it does not look good. Probably I will throw the printer to trash after I have upgraded it, because I am not going though all the configuration (and/or changing board) again.

[@pgbilbo]

Thank you for making this video. It convinced me to try Marlin on my X5 GT as well. Did you also have the problem that as soon as the sdcard option is enabled the communication with the TMC drivers is broken? I am also using the viki2. If so, did you find a solution or workaround?

[@viktarsamuel8822]

Hi, I have Smoothie running, no problem, but and also would like to run Marlin 2.0 on it to get more features. Can I ask you some questions? Thanks

[@LukesLaboratory]

Sure, no promises tho

[@corynardin]

Did you find that Marlin on the azteeg is giving you the crispness that you were getting from the 8 bit setup?

[@LukesLaboratory]

Yessir! That's exactly it. Once I have Mongo back together I'll do a comparison video between the quality of the two.

[@ArthurWolf]

Hey. Freezes like that *absolutely never* happen with a Genuine Smoothieboard. Azteeg uses a less powerful microcontroller than Smoothieware is designed for, and doesn't keep it's config and firmware up to date with the project. Both of these issues are likely to be the source of the print quality issues you seem to have. If it failed with heaters on, it must be using an *extremely* old firmware, this has been impossible in Smoothie for like half a decade. The stuff about the motion planner being better in Marlin than Smoothie is absolutely completely definitely incorrect, somebody was just talking to you out of ignorance, the Marlin devs themselves agree with this. Also, the difference in contributor base for Marlin vs Smoothie isn't as large as you make it out to be, and that's by design, due to the fact Smoothie also supports CNC and lasers, making it useful to more people, resulting in a wider range of contributors... If you take another crack at Smoothieware at some point, make sure you reach out to the Smoothie community if you run into issues, what happened to you are definitely things we would have been able to help you with...

[@LukesLaboratory]

Hi - Its been some time since i've posted this, thanks for the response. A few things - 1) As I said - I actually tried quite a few versions of the firmware, updating/downgrading the config as I went along from legacy releases to the edge builds. The lockup issue persisted through all. 2) Here's the forum link where I reached out as I reported in the video - forum.smoothieware.org/forum/t-4483417 I understand that the Ethernet board is an add-on, but for all I can tell it's just a simple hat that uses smoothies internal web server. Lockups happened regardless of whether or not I was connected to the web server. 3) I just confirmed that the Azteeg X5 GT uses the LPC1769FBD100 Chip, just the same as Smoothie V1. No difference. 4) After I troubleshot&resolved the lockup issue, I did some additional printing comparisons since it was easy to swap back/forth between firmwares, and to try to substantiate what I read around the web. Using identical Gcode I got better (subjective, but repeatable) and faster results with Marlin's planning than smoothies. Same thing now that I've migrated to klipper - Same Machine, same board, improvement when switching to different firmware. (klipper is adding a rpi for calculations to the mix, but still) I'd love to do a trio of comparisons between Klipper, Smoothie, and Marlin (maybe including stepd) and the LPC1769 can even run RRF with some tweaks for a quartet! I can also be reached on twitter @LukesLaboratory or my email [e-mail]

[@ArthurWolf]

@@LukesLaboratory There's no version of the firmware and config that's supposed to work well with that board, because Panucatt hasn't made it so they would by keeping things up to date and testing it out It sounds like your issue was related to the Ethernet extension though as you said in the video. My comment was more about other issues we've seen users have with Azteeg products, that were due to Panucatt not keeping things up to date. About the forum post, I was actually trying to help you and get to the bottom of this, but at some point you just stopped participating. You ultimately gave an update saying you figured out it's the module, but if you had kept interracting with me / participating in issue resolution, we would have figured it out together, the way we do with hundreds of users every year We've seen lots of panucatt products ship with the 68. If yours has the 69 that's great, he has finally decided to fix that at some point. Doesn't change the fact he's not doing what needs to be done for his board and the firmware to work along nicely. The firmware and config he provides are extremely outdated ( last time I checked ). About print quality, I talk weekly to users who say Smoothie provides them with the best quality of all the systems they've tried ( which is usually mostly just Marlin, but not always only ). And there are good reasons why that would be true, we put a lot of work into that, more than any other system. However, as I said Azteeg doesn't make it so their hardware works well with the firmware, so you can expect all sorts of issues, some of which will cause print quality issues, which I would guess is what you are seeing.

[@LukesLaboratory]

@@ArthurWolf as for the comments on the board what kind of things could cause the issues with both hangups/print issues? All of the features worked (when not frozen) to the best of my knowledge regardless of firmware or functional config. Marlin and klipper work flawlessly, as long as pins were assigned appropriately which seems to really be the only kind of customization needed to run smoothie as well on this board. Sofar the only main difference seems to be tmc2660 drivers. I will do some digging to find my old config, but what I can tell you is that regardless of whether I started with panucatts config or started from scratch and configured following documentation, it did not make a difference either way. When I identified the issue as the ethernet module, I was expecting a line of questioning related to the module, if any follow up was needed/desired, especially since I was running a non-official board w/ add-on hat As for print quality - I have nothing concrete here. All of the settings I can touch were more conservative than their marlin counterparts as I tried to diagnose their root, but even slowing down acceleration considerably did not get me the results I desired. Similarly, marlin cannot deliver as fast and as consistent prints as klipper does, even with considerably more tuning. I chalked it up to internal motion planning differences and called it a day. I would love to find out why the differences are the way they are (why smoothieware is the only firmware to not play nicely on this board) and if possible, to maybe give it another fair shake now that marlin 2.0 has some serious hours under it's belt.

[@ArthurWolf]

@@LukesLaboratory The thing is, Smoothie v1 is extremely optimized for the current chip, it's using the chip much more intensively than firmwares that got to run there more recently. But this isn't true for Panucatt stuff: they have not done the intense testing / optimization we have done with v1+smoothieboard. Issues like you describe, nowadays we pretty much never see with Smoothieboard v1 under normal circumpstances, while they are fairly common with panucatt or clone hardware. You'd need to use the very latest firmware and very latest config, adapted to work for azteeg, but if you did that, you'd be in an extremely untested setup, so there's really no way you can win / be in a normal situation with this hardware, right now...

[@ChrisHarmon1]

@@LukesLaboratory I have experienced the exact opposite and I'm talking smoothie 4+ years ago vs todays Marlin 2.0. From what I have read on the google Delta forums over the years Arthur is very proud of his planner. I was actually just researching to see if I could flash Smoothie on this printers STM32 board. Wish I could afford to support you Arthur but this is hopefully next best thing :)

[@silicon.alchemist]

background music is really annoying...