Upping my umbilical game

[m00dawg]

Printer Model

Trident 300

Extruder Type

Orbiter

Cooling Type

Other

Recently I decided to try XOL/OB2. One thing I didn't realize is that it was meant for an umbilical setup so after a few other gotchas, I've been quite happy with it! I know I can improve things more though and was curious what folks are using with XOL/OB2/NH36 (with umbilical guide).



I've been happy enough with the setup but it is a little janky. I did buy some piano wire to help add rigidity, but it would seriously dig into the plastic on my previous setup. Right now I'm using a length of spare nylon filament from a PrusaMK3 setup, but it's not long enough. I know some folks run things through a cable gland and I am pondering this while also just putting the spool underneath the printer so it can go up and over (that's how I have my V0 setup). Any other thoughts perhaps? This thing only prints PLA and maybe on rare occasion PETG so I don't plan on ever having covers, which gives me plenty of room along the top.

Anyways just curious to see how folks setup their umbilical for XOLs, TLDR

Also as an aside any tips to help make sensorless homing of the Y less bangy/clungy that'd be appreciated - I tried several combinations of speed and stalls but while the X seems to have the perfect, quiet, touch, the Y doesn't exactly have finesse.

 

[NoGuru]

Also as an aside any tips to help make sensorless homing of the Y less bangy/clungy that'd be appreciated - I tried several combinations of speed and stalls but while the X seems to have the perfect, quiet, touch, the Y doesn't exactly have finesse.

You can try and drop voltage when homing. Here is an example of my config. If you look, I drop my X by .59% and Y by .60%

VoronV2.4/printer_data/config/homing.cfg at main · NoGuru2/VoronV2.4

Backup of Config. Contribute to NoGuru2/VoronV2.4 development by creating an account on GitHub.

github.com

 

[m00dawg]

You can try and drop voltage when homing. Here is an example of my config. If you look, I drop my X by .59% and Y by .60%

VoronV2.4/printer_data/config/homing.cfg at main · NoGuru2/VoronV2.4

Backup of Config. Contribute to NoGuru2/VoronV2.4 development by creating an account on GitHub.

github.com

Oh thank you!! I had that opposite I guess. I thought I had to raise voltage, whoops! Yeah I'll give this a read, thanks for sharing!

 

[m00dawg]

You can try and drop voltage when homing. Here is an example of my config. If you look, I drop my X by .59% and Y by .60%

VoronV2.4/printer_data/config/homing.cfg at main · NoGuru2/VoronV2.4

Backup of Config. Contribute to NoGuru2/VoronV2.4 development by creating an account on GitHub.

github.com

By the way, your printer config proved incredibly helpful for getting my sensorless homing working on my 2.4. The key turns out was my homing speed was way too low. I bumped it to 40 and all my problems seemingly went away. This plus having a working example of homing_override while using a Beacon was also helpful! I'll be applying these changes to my Trident as well since the X homes well but the Y is still a bit rough.

On the umbi front, both my Trident and 2.4 are now both setup with XOL + Orbiter 2 + Beacon + Revo + NH36. The Trident is still using 2 cables, but on the 2.4 I was able to wire up the Beacon to the NH36 connector. For the Trident, I printed a cable clamp sort of things for the rear right motor to run the cables through which was super helpful in fixing the clutter. So I followed suit on the 2.4 though I had to zip-tie the cable to the PTFE tube to raise it up and out of the way. I'll need to get a cable sheath I think to avoid the zipties from rubbing but otherwise it seems to work well.

XOL ended up being a pretty huge change given all the periphery changes (umbilical and sensorless homing included) to the point it feels like I almost built 2 entire printers after everything was done, but oh man is the 2.4 soooo much better now that I have a solid probe solution. They're both working pretty great!

TLDR thanks so much for all your help, NoGuru!

 

[NoGuru]

Glad to assist! I have an SLS XOL toolhead I have never put into use. I would like to hear how you like it? Maybe someday I will get around to using it.

 

[m00dawg]

Glad to assist! I have an SLS XOL toolhead I have never put into use. I would like to hear how you like it? Maybe someday I will get around to using it.

For me it was quite an upgrade given all the things that had to change at once so hard to judge it fully in isolation. But I'll try! Notably the 2.4 definitely feels like a new printer now having done all that. Feels more like a "2.5" or some such. This is mostly because of Beacon, but the QGL works soooooo much better now. The XOL helped me realize this by way of being able to use NH36 and make use of its on board USB hub. I only have one umbilical cable now (I still have two on the Trident but I'll be updating it once I get more connectors and some wire to make my own cable).

For the XOL itself, to start with the negatives, the big thing is it requires moving to less standard things and mandates a lot of changes compared to a stock Voron (Moving to BFI, Umbilical, NH36 or similar) and some of those changes aren't "on rails" nearly as much as a standard Voron. Meaning there's lots of implementations and no standard. For well seasonsoned folks, this isn't as big a deal, but does mean XOL isn't currently something I would suggest for someone new to Voron.

Other minor gripes are the LEDs seem to be harder to source and aren't the same size. I just opted not to use them in my builds as a result. And you have to screw some things direct into plastics such that maintainability might require having to print replacement parts more often. I guess last thing is, I do wish the bed probe was part of the toolhead rather than the carriage, but that's rather minor.

On the plus, I found I got much better hotend cooling, even with the small front fan. I still need to convert to 5V Delta fans but even with the ones that came with the kits I got, I've been pretty happy. I did finally get a jam on my Trident but this seems to be due to my nozzle being gunked up. Swapping it with a new one solved that (reminder I need to do cold pulls more often). Hard to tell how much of this is XOL vs using an Orbiter 2 as I swapped to both at the same time.

The toolhead is lighter and I got notable improvements on my Trident with ressonance testing. So while I don't actually print fast, those claims seem to be real. Part cooling is also improved but I also don't go ham on that either.

It does seem easier to disassaemble and work on. And while the plastic into screws thing is a negative, being able to remove the toolhead with 3 screws, much like the V0, is rather nice. In fact it does feel a lot like the Mini-SB on my V0, which I quite like, except it supports bed probes.

Oh and the belt mounts are really nice once I got used to them. To this point I've always been wary of having to remove the carriage because of having to re-secure the belts, but XOL makes that basically a non-issue. That's very nice!

Overall yep I'm very happy! Though I will admit it was quite a slog to get there, but a lot of that were periphery issues not directly related to XOL itself, just requirements to use XOL.

All that said, I didn't know there were SLS options for XOL. Curious where you got yours?

 

[NoGuru]

Back when XOL was in its first stages, the guys did a group order from PCBway or some other site that can do SLS...or was is SLA...I forget since I am old but anyway yeah, just sent the designs out and I want to say it was about $70 to have it made. But the downside is we had to do the tapping for the screws. Lucky I work in an office that has a Pilot plant for manufacturing and the tooling guys in back did the tapping for me.

I know what you mean by getting away from the standard Voron build. Yes you can see improvements in some areas but you loose the support and simplicity to an extent.
I think that is why I have switched but I think I will for the fun of it.

 

[3dCase]

First of all sorry for tagging along on this thread. It deals very much with a subject that is made of nightmares for me so I am not wanting to hijack this thread, simply want to learn. Still lurking here and trying to get as comfortable and proficient as you two guys ( @m00dawg @NoGuru )with changing things on my V2.4 that require cfg changes as well.
Ran into a problem recently after an update and now I think I will force myself to do a complete rebuild to incorporate all the bits I bought already, but never have fitted yet because it was working ok. This is actually an embellishment of sorts, because the truth is I am shitting bricks because of having to change my whole cfg, macros and all that, which is simply too far out of my comfort zone to volunteer into. But, I got a cartographer, EBB36 canbus, new mount for the toolhead plus all the cnc milled parts from chaotic lab to do away with the plastic parts. I already fitted my WS7030 centrifugal fan on the feame and I am using it as a parts cooler. I also converted to sensorless x and y and removed all the cable chains so basically my wiring is kind of umbillical at the moment but without using the canbus EBB36 for it.
I will write myself a gameplan for the conversion and it will be a step by step guide. First question now is will I still need my nozzle z limit switch? The one sitting behind the bed?
Everytime I read some of these stories about these changes I am impressed at how it comes out in the end, so hat off for the achievements. I hope one day I will be like that, but I am but a simple machine engineer, and a dinosaur at that. These coding things scare me a bit (much).

 

[m00dawg]

Yeah I feel your pain 3dCase, I wasn't comfortable changing so much at once and actually is why I have a BTT CANBus solution sitting here that I've never used. Similarly I wasn't super comfortable getting too far away from the Voron standard build, really trying to stick to cable chains as much as I could (noting the USB cable that comes with the LDO Nitehawk's is rated for chains).

My 2.4 was out of commission for about 3 weeks to do the changes, but 1 week of that was a vacation and I also was taking my time. The actual hardware conversions I did up above took maybe less than 4 hours total if you ignore waiting on parts in the mail and printing the ABS parts.

The software config took me a few hours, in part because I didn't peruse NoGuru's printer config closely enough. I got tripped up with Beacon's reference config vs using homing_overrides. Also got tripped up thinking I needed to slow down the homing speeds for sensorless homing when in reality, for me (and seemingly NoGuru), it works much better with faster homing.

All that said though, I opted to upgrade because I wanted better first layer adhesion. So what prompted this whole thing for the 2.4 was getting Beacon going. For my Trident it was more that I wanted to try and improve my jamming issues (by testing out the Orbiter 2). I said that to say, if you're getting good results with your printer as-is, there's a case to be made to leave it be unless you have a reason to do so. That is if you're making prints and want to keep doing that. If you want to tinker as part of the journey, well that's part of the Voron ethos so you should! But in my case since I print customer printers with all my printers in addition to my own stuff, I had to be somewhat pragmatic. In the end, the improvements are GLORIOUS and that also translates to my customer prints. But I did have a specific target of improvement in mind and generally revolved around that for my upgrades. Just so happens that to get it setup the way I wanted, I ended up changing a lot more than intended.

 

[3dCase]

Yeah I feel your pain 3dCase, I wasn't comfortable changing so much at once and actually is why I have a BTT CANBus solution sitting here that I've never used. Similarly I wasn't super comfortable getting too far away from the Voron standard build, really trying to stick to cable chains as much as I could (noting the USB cable that comes with the LDO Nitehawk's is rated for chains).

My 2.4 was out of commission for about 3 weeks to do the changes, but 1 week of that was a vacation and I also was taking my time. The actual hardware conversions I did up above took maybe less than 4 hours total if you ignore waiting on parts in the mail and printing the ABS parts.

The software config took me a few hours, in part because I didn't peruse NoGuru's printer config closely enough. I got tripped up with Beacon's reference config vs using homing_overrides. Also got tripped up thinking I needed to slow down the homing speeds for sensorless homing when in reality, for me (and seemingly NoGuru), it works much better with faster homing.

All that said though, I opted to upgrade because I wanted better first layer adhesion. So what prompted this whole thing for the 2.4 was getting Beacon going. For my Trident it was more that I wanted to try and improve my jamming issues (by testing out the Orbiter 2). I said that to say, if you're getting good results with your printer as-is, there's a case to be made to leave it be unless you have a reason to do so. That is if you're making prints and want to keep doing that. If you want to tinker as part of the journey, well that's part of the Voron ethos so you should! But in my case since I print customer printers with all my printers in addition to my own stuff, I had to be somewhat pragmatic. In the end, the improvements are GLORIOUS and that also translates to my customer prints. But I did have a specific target of improvement in mind and generally revolved around that for my upgrades. Just so happens that to get it setup the way I wanted, I ended up changing a lot more than intended.

So are you still having a fixed dedicated z homing limit or is that gone now and the beacon, in my case the cartographer, is doing that as well now?

 

[m00dawg]

So are you still having a fixed dedicated z homing limit or is that gone now and the beacon, in my case the cartographer, is doing that as well now?

The switch is still there but it's unhooked and unused. Beacon (or in your case Cartographer though I haven't tested it to say) handles it all.

The one cautionary tale there is the Beacon has to be over the build plate to do that. So you need to home X/Y first, then center the toolhead over your build plate (or at least assuredly on part of it) and then do the Z level.

Beacon apparently supports Z level by contact but I haven't tried that (I prefer touchless leveling).

 

[3dCase]

The switch is still there but it's unhooked and unused. Beacon (or in your case Cartographer though I haven't tested it to say) handles it all.

The one cautionary tale there is the Beacon has to be over the build plate to do that. So you need to home X/Y first, then center the toolhead over your build plate (or at least assuredly on part of it) and then do the Z level.

Beacon apparently supports Z level by contact but I haven't tried that (I prefer touchless leveling).

That sounds exactly what I wanted, one does it all kind of thing. Thanks for that, one more tick on my list.

 

[m00dawg]

It's great! Easily the best leveling solution I've used across all printers. That's not saying there aren't many good ones out there, but Beacon hits all my requirements.

 

[NoGuru]

I can tell you have tried almost all Z probes except Cartograher and the Piezo one...one on a Voron. Well Beacon is by far my favorite and Contact for Z is amazing since I change to different build plates time to time.

I would almost always suggest doing one change at a time but its hard to do that when its a case of "while I am in there" to do lots more.

 

[m00dawg]

I would almost always suggest doing one change at a time but its hard to do that when its a case of "while I am in there" to do lots more.

Yep! I strive for that too and definitely recommend it, although there are rare exceptions such as is when things are tightly correlated such that one can't get sufficient information due to the relationship between multiple parameters. In the case of my XOL upgrade I would have much preferred doing one thing at a time but since it required so many other changes, it was a hard proposition without adding a lot of steps and cost in between. I'm still not quite done, at least on the Trident, as I need to better wire up the Revo to the NH36. It's a tad sketch right now. having done all that though, I am definitely happy with the results.

 

[3dCase]

I can tell you have tried almost all Z probes except Cartograher and the Piezo one...one on a Voron. Well Beacon is by far my favorite and Contact for Z is amazing since I change to different build plates time to time.

I would almost always suggest doing one change at a time but its hard to do that when its a case of "while I am in there" to do lots more.

Yes doing upgrades should be done gradually since it can end up like building a whole new printer in one go. But for me this time it actually was a gradual process, maybe you remember my first post about a year ago now. Built it out of a suitcase having transported it from western europe to eastern europe on a plane. Then the slow process of finding all its weak spots. Finally got rid of the drag chains and went with an ugly umbilical of a bundle of wires. Have to say that I did not have one wiring issue after that so for me that proves that umbilical is far superior as a build method. Then replaced that clockwork monster and its trianglelab hotend. Then removed limit switches for sensorless homing and finally went with a parts cooler which was frame mounted via CPAP hose. Now I feel I am going to complete the process with the addition of the cnc frame parts from chaotic lab. So this last step is to go canbus with a proper canbus cable from Igus, cartographer with its cnc mounting plate for the whole hotend. And see what else I will find that I don’t like during the strip down.
Very gradual with a sprint for the finish line