I don't know if this happens to any of you, but anytime I see a "new Alex Kenis video" notification, a smile shows up on my face, no matter if I don't care at all about the subject of the video (though in this case I care)
I am afraid most of us are just a bunch of fanboyish followers that will enjoy whatever subject you dive into Alex. In my case it is not about lack of care or interest regarding the particular field of the video, but that I've found myself interested in subjects that I'd never vote for if you had asked me in advance. Several times. Whatever topics your patreons prefer I'm 100% into. Thank you guys, too.
Would love to hear a long rant on Backlash and the hillbilly solutions for it. Lots of people use those acme spring loaded backlash nuts for their Z-axis with no regard for X-Gantry weight vis-a-vis spring tension. I have been battling some serious backlash slop in the Z for both directions on an Ender 3. I've resorted to melting/tapping (meltapping?) a Delrin block over the T8 screw to get an exact mold of the threads and then affixing that to the X-Gantry as my nut. Yeah... things got that bad over here.
I may end up doing that since a lot of people seem to be asking lately, and I was emailed some pics of backlash nuts installed completely wrong on some stock printers. But I actually used to do the same thing you did with 3d printed nuts, threaded into a scrap of leadscrew which I heated up with a torch. Then id use another leadscrew that was scuffed up to make the surface abrasive and ran that through it for a while and then slathered it with lube. Still using those nuts too.. kinda. They are on two homebrew printers that are packed away until I get more space.
@@AlexKenis I'm sure you know of these alrdy, but here's a 3D printed threadless ballscrew m.ruclips.net/video/Rkhm-hp9788/видео.html. I'll be printing and trying this instead since it's....well it's just sexier than an ugly block of Delrin. Many versions of it on thingiverse ofcourse. Best Regards
I have to admit, I really hate youtube videos/tutorials/information and watching them, I much prefer a big wall of text with a couple of images... I think I'm alone in that.... but, Alex, your videos are one of the very few that I actually enjoy listening to/watching and playing at normal speed (I usually x2 videos if I have to watch them and can't find the info elsewhere). you've done some fantastic work, even an idiot like me can learn from you, I hope you continue creating content for years to come... even if it's one video a year. thank you for testing these things so we don't have to!
Belt tension and rollers sounds interesting (although according to gates you are supposed to use their app and tune the belt to a specific frequency?). Its incredible the amount of work you do.
29:35 This is the same approach I've been taking. No matter what, it seems like single sided assemblies don't grip the rail very hard, or for very long. And the tri-wheel is much better for tension than the quad wheel designs most printers ship with. These choices in original design have always confused me.
So much cool content! Since you're asking us on how to prioritize: Please let it be the topics you have closest to 'complete' that allow you to post many in-depth videos and fast :). As for personal preference of exciting topics: 1) what is the best belt tension? talk about your belt tensioner? (want to print and/or design my own) 2) lapped bed (don't you need a surface plate to test?), carbon fiber Y-sled and how it compares to stock bed (weight? stability?) 3) ALL the diy test jigs. I'd also like to see how you'd take advantage of decent cheap tools (0.01mm calipers that I honestly think every hobbyist must have, 0.001mm dial indicator) for more precise measurements. Happy to contribute some of my best buys (reply and I'll email). 4) Everything on parts cooling.
Commenting here but am a patreon .. pullies teethed vs not is interesting (my guess: esoterical differences) as well as sensorless/nozzle leveling and off-printer cooling vs cooling near the nozzle .. Practical aspect interests me the most .. as for printing surfaces, I've heard these (could look for an translation if you're interested) pressed/glued sawdust plates were good for lots of plastics Btw I was against live streams when you asked your patreons but your's are so different/interesting, I ended up watching all your (few) streams in post. .. I d now vote for more less quality in favor of more vids. Your 3d channel is so different from all the other's and it's would be cool to hear more from you. I'd compare you to cnc kitchen, though styled differently equally interesting and in-depth
Hey Alex, have you posted your X-Axis belt tensioner files anywhere? I've looked around your accounts on various sites but unless I missed it, it doesn't seem like you post many of your designs. I'm trying to do testing of my own and I really like your solution. I realize this video is a bit old but I'd appreciate it if you see this. I'll happily throw in a donation for the time. Even just the mesh would be helpful. Thanks and great work!
Great to see you well, hope you stay that way. :) I always learn a lot from your videos, so really looking forward to the upcoming ones. As for priority, and what I'd like to see: parts cooling jigs/performance/categories ( it would also be nice to compare cooling performance vs. noise) What you did not mention, but would be nice is some tool recommendations, what's worth paying extra money for, and what's just throwing money out the window ( multimeter, oscilloscope, ferrules, crimping tools, calipers, etc. )
What about using a piezo sensor and amplifier on the hotend mount to perform the trigger event when the nozzle taps on the print bed. The sensitivity can be tuned to prevent the nozzle digging into the printing bed. I have this setup using the precision piezo
for sensor less bed leveling, would it be possible to use a tmc2209 with the sensor less homing? just bang that nozzle into the bed until the driver realizes you have hit the limit of travel? :) Then you can use a glass print surface.
Hey man, just found your videos and started going through them when I saw this with your belt tensioners! Do you have those available anywhere? I have tried a few but I just have not fallen in love with anything out there. I like the long adjustability of your design too. If you don't have them online maybe I can just modify something i have to include the longer adjustment range but figured it never hurts to ask first :) Keep up the videos man!
have you ever thought of plopping a stepper on a cold bed and printing against the side of the shaft to make a 4th axis print? for that matter, is there ever a time where you would want a heated part fan?
One idea I might try (once it arrives) is to use one of those $1 external nozzle-mounted pressure sensors that ship with many delta (kossel) printers for auto mesh bed leveling. Search for "kossel pressure sensor". Perhaps not for everyday homing, but for the occasional EEPROM-stored mesh, it would be way faster than manual probing.
Information overload! Sooo much stuff going on you are a true hero. By the time we improve the ender 3 there will be a v5 version out 😂. Looking at the steel plate and v-roller that you made reminds me of the exoslider carriages. It uses bearings to roll on the flats of the extrusion rather than the V. Some DIY versions around on thingiverse try profile slider some designs better than others enter the thingiverse engineer disclaimer.
i am converting one of my enders to smooth rods, sort of readapting a crearibo design to an ender 3 pro, and since the v slot wheels are not going to be there anymore, your floppy sturdier setup that you use on your test machines (I am curious about that thing and would really like to print/ try it out) might fit on the e3 pro without interfering with build volume.
I've been thinking about doing an automatic z-offset finder thingy and I'm wondering if a piezo sensor off the side of the bed might be a good way to do it.
I realize this is an old video but I'm wondering about a mod idea. With all the speed boat benchys, I know there are endless settings related to acceleration jerk and other mechanical problems with moving that fast. What about replacing the stepper motors and belts with hydraulic pumps & pistons? I've worked with large hydraulic systems for a long time & I know one major advantage compared to electrical systems is that hydraulic systems can change direction almost instantly.
Might be a dumb question but you've incorporated some pretty unique design elements into a lot of your printers. I consider myself a bit of a tinkerer and would love to compare your designs to improve my machines. I mostly have deltas but I also maintain the prusa style machines at our local maker space. Do you have somewhere that you decently documented the machines and their print quality? I have been going through a ton of your videos and only see little snippets.
I was watching a guy print on a corexy at 800mms, and I had a thought. sometimes people put 2 steppers on the z axis, in a corexy setup, what if you had 6 steppers? 2 for z, 2 for diagonal x, and 2 for diagonal y. if you put a spring tensioned idler pulley along the belt path, you can monitor pulley position and change belt tension be giving an extra step to one stepper or the other. if you put your steppers at the 4 corners of the frame, closest to the end of the belt, you can ensure optimal tension at the belt ends. has this been done before, and I just haven't found it yet, or could I be on to something?
It'd take a bit more math, and a special start print sequence, but couldn't you get a pretty great solution using both an inductive probe and the hotend as a switch? The inductive probe can do a mesh bed leveling sequence, then the nozzle itself can be used (say in an exposed corner or other exposed metal that's the same height as the bed surface) to figure out the offset between the inductive probe and the tip.
I never saw so many heads explode when I called a raspberry pi a toy computer... I did not think the raspberry pi would end up with that fanboyism like it has in the 3d printing community.
Another idea for sensorless bed leveling: Use Trinamics Stallguard Feature (TMC drivers with SPI/UART required) to detect the Nozzle running into the bed. It basically measures how much resistance the motor faces when moving and you set a threshold for when it should trigger. Determining the threshold value might be finicky but once it runs it's good. Prusa uses this for X and Y afaik.
stall guard will also fight with the spring tension, it will not be triggered exactly upon touching the the surface. well unless you didn't use any sprint to mount your bed.... perhaps it will work.
For X and Y it's fine, but you shouldn't be using it for Z since stallguard won't produce consistent enough results in that order of accuracy that is required for bed leveling.
@@martijnpeters yup, because it really doesn't matter much if X/Y homed 1mm or 2mm off unless you are trying to maximize your print space. but for Z it is crucial since your print will not stick if it is too far off, or your nozzle will simply dig into the bed.
@@evertchin The accuracy is enough to re-home X/Y and then resume the print seamlessly after power loss or step loss. You can read out the load intensity with 1024 step resolution via SPI/UART. For the z-axis you could detect the gradual rise in resistance when the print-head starts depressing the springs. This requires a fast readout of the stepper drivers registers and some calibration though. It's not easy but it should be possible. Check this out for further information: www.trinamic.com/fileadmin/assets/Support/Appnotes/AN002-stallGuard2.pdf
I don't know if this happens to any of you, but anytime I see a "new Alex Kenis video" notification, a smile shows up on my face, no matter if I don't care at all about the subject of the video (though in this case I care)
You like learning new things and he tends to deliver on that
Subjects doesn't matter he Is so good trying stuff
I am afraid most of us are just a bunch of fanboyish followers that will enjoy whatever subject you dive into Alex. In my case it is not about lack of care or interest regarding the particular field of the video, but that I've found myself interested in subjects that I'd never vote for if you had asked me in advance. Several times.
Whatever topics your patreons prefer I'm 100% into. Thank you guys, too.
Everything!
With subtitles please!
You do a really good stuff, thank you.
Would love to hear a long rant on Backlash and the hillbilly solutions for it. Lots of people use those acme spring loaded backlash nuts for their Z-axis with no regard for X-Gantry weight vis-a-vis spring tension. I have been battling some serious backlash slop in the Z for both directions on an Ender 3. I've resorted to melting/tapping (meltapping?) a Delrin block over the T8 screw to get an exact mold of the threads and then affixing that to the X-Gantry as my nut.
Yeah... things got that bad over here.
I may end up doing that since a lot of people seem to be asking lately, and I was emailed some pics of backlash nuts installed completely wrong on some stock printers. But I actually used to do the same thing you did with 3d printed nuts, threaded into a scrap of leadscrew which I heated up with a torch. Then id use another leadscrew that was scuffed up to make the surface abrasive and ran that through it for a while and then slathered it with lube. Still using those nuts too.. kinda. They are on two homebrew printers that are packed away until I get more space.
@@AlexKenis I'm sure you know of these alrdy, but here's a 3D printed threadless ballscrew m.ruclips.net/video/Rkhm-hp9788/видео.html.
I'll be printing and trying this instead since it's....well it's just sexier than an ugly block of Delrin.
Many versions of it on thingiverse ofcourse.
Best Regards
Prioritize human cloning so that you can prioritize all of it! Seriously cool stuff man, loving the channel!
I have to admit, I really hate youtube videos/tutorials/information and watching them, I much prefer a big wall of text with a couple of images... I think I'm alone in that.... but, Alex, your videos are one of the very few that I actually enjoy listening to/watching and playing at normal speed (I usually x2 videos if I have to watch them and can't find the info elsewhere). you've done some fantastic work, even an idiot like me can learn from you, I hope you continue creating content for years to come... even if it's one video a year. thank you for testing these things so we don't have to!
Belt tension and rollers sounds interesting (although according to gates you are supposed to use their app and tune the belt to a specific frequency?). Its incredible the amount of work you do.
29:35 This is the same approach I've been taking. No matter what, it seems like single sided assemblies don't grip the rail very hard, or for very long. And the tri-wheel is much better for tension than the quad wheel designs most printers ship with. These choices in original design have always confused me.
So much cool content! Since you're asking us on how to prioritize: Please let it be the topics you have closest to 'complete' that allow you to post many in-depth videos and fast :).
As for personal preference of exciting topics:
1) what is the best belt tension? talk about your belt tensioner? (want to print and/or design my own)
2) lapped bed (don't you need a surface plate to test?), carbon fiber Y-sled and how it compares to stock bed (weight? stability?)
3) ALL the diy test jigs. I'd also like to see how you'd take advantage of decent cheap tools (0.01mm calipers that I honestly think every hobbyist must have, 0.001mm dial indicator) for more precise measurements. Happy to contribute some of my best buys (reply and I'll email).
4) Everything on parts cooling.
I am excited for all the parts cooling fun!
Great to see a new video - some of the best info available! Really looking forward to getting hold of your belt tensioner STLs.
me too
So, you have a busy decade ahead of you!
Commenting here but am a patreon .. pullies teethed vs not is interesting (my guess: esoterical differences) as well as sensorless/nozzle leveling and off-printer cooling vs cooling near the nozzle .. Practical aspect interests me the most .. as for printing surfaces, I've heard these (could look for an translation if you're interested) pressed/glued sawdust plates were good for lots of plastics
Btw I was against live streams when you asked your patreons but your's are so different/interesting, I ended up watching all your (few) streams in post. .. I d now vote for more less quality in favor of more vids. Your 3d channel is so different from all the other's and it's would be cool to hear more from you. I'd compare you to cnc kitchen, though styled differently equally interesting and in-depth
Great video Alex, Like the others I’m keen for those belt tensioner STLs ....
The "induktive sensor"/radar works by sensing water (you are 80% water)
It dont work with only metal. Sorry to say
Great update by the way :-)
Hey Alex, have you posted your X-Axis belt tensioner files anywhere? I've looked around your accounts on various sites but unless I missed it, it doesn't seem like you post many of your designs. I'm trying to do testing of my own and I really like your solution. I realize this video is a bit old but I'd appreciate it if you see this. I'll happily throw in a donation for the time. Even just the mesh would be helpful. Thanks and great work!
I still plan to do the follow up vid, I just get derailed too easily and got tied up in the Kingroon.
@@AlexKenis No worries man, appreciate everything you're doing.
thanks as always
Great to see you well, hope you stay that way. :)
I always learn a lot from your videos, so really looking forward to the upcoming ones.
As for priority, and what I'd like to see:
parts cooling jigs/performance/categories ( it would also be nice to compare cooling performance vs. noise)
What you did not mention, but would be nice is some tool recommendations, what's worth paying extra money for, and what's just throwing money out the window ( multimeter, oscilloscope, ferrules, crimping tools, calipers, etc. )
What about using a piezo sensor and amplifier on the hotend mount to perform the trigger event when the nozzle taps on the print bed. The sensitivity can be tuned to prevent the nozzle digging into the printing bed. I have this setup using the precision piezo
for sensor less bed leveling, would it be possible to use a tmc2209 with the sensor less homing? just bang that nozzle into the bed until the driver realizes you have hit the limit of travel? :) Then you can use a glass print surface.
Thinking of playing about with this soon myself
Will you share the STL of that belt tensioner
Hey man, just found your videos and started going through them when I saw this with your belt tensioners! Do you have those available anywhere? I have tried a few but I just have not fallen in love with anything out there. I like the long adjustability of your design too. If you don't have them online maybe I can just modify something i have to include the longer adjustment range but figured it never hurts to ask first :) Keep up the videos man!
have you ever thought of plopping a stepper on a cold bed and printing against the side of the shaft to make a 4th axis print? for that matter, is there ever a time where you would want a heated part fan?
One idea I might try (once it arrives) is to use one of those $1 external nozzle-mounted pressure sensors that ship with many delta (kossel) printers for auto mesh bed leveling. Search for "kossel pressure sensor". Perhaps not for everyday homing, but for the occasional EEPROM-stored mesh, it would be way faster than manual probing.
Thanks for the add and great information.
Information overload! Sooo much stuff going on you are a true hero. By the time we improve the ender 3 there will be a v5 version out 😂.
Looking at the steel plate and v-roller that you made reminds me of the exoslider carriages. It uses bearings to roll on the flats of the extrusion rather than the V. Some DIY versions around on thingiverse try profile slider some designs better than others enter the thingiverse engineer disclaimer.
i am converting one of my enders to smooth rods, sort of readapting a crearibo design to an ender 3 pro, and since the v slot wheels are not going to be there anymore, your floppy sturdier setup that you use on your test machines (I am curious about that thing and would really like to print/ try it out) might fit on the e3 pro without interfering with build volume.
I've been thinking about doing an automatic z-offset finder thingy and I'm wondering if a piezo sensor off the side of the bed might be a good way to do it.
I realize this is an old video but I'm wondering about a mod idea. With all the speed boat benchys, I know there are endless settings related to acceleration jerk and other mechanical problems with moving that fast. What about replacing the stepper motors and belts with hydraulic pumps & pistons? I've worked with large hydraulic systems for a long time & I know one major advantage compared to electrical systems is that hydraulic systems can change direction almost instantly.
Might be a dumb question but you've incorporated some pretty unique design elements into a lot of your printers. I consider myself a bit of a tinkerer and would love to compare your designs to improve my machines. I mostly have deltas but I also maintain the prusa style machines at our local maker space. Do you have somewhere that you decently documented the machines and their print quality? I have been going through a ton of your videos and only see little snippets.
what cad package are you using there?
I'm running Klipper with moonraker and Mainsail on a Voron. It runs fast on any pi.
I only watch this channel to see what the explanation for the background noise is.:)
I was watching a guy print on a corexy at 800mms, and I had a thought. sometimes people put 2 steppers on the z axis, in a corexy setup, what if you had 6 steppers? 2 for z, 2 for diagonal x, and 2 for diagonal y. if you put a spring tensioned idler pulley along the belt path, you can monitor pulley position and change belt tension be giving an extra step to one stepper or the other. if you put your steppers at the 4 corners of the frame, closest to the end of the belt, you can ensure optimal tension at the belt ends. has this been done before, and I just haven't found it yet, or could I be on to something?
Looking forward to this "season". And was that part of a Tele pickup in with your water cooling stuff?
You should build a Voron and make a review of it.
It'd take a bit more math, and a special start print sequence, but couldn't you get a pretty great solution using both an inductive probe and the hotend as a switch? The inductive probe can do a mesh bed leveling sequence, then the nozzle itself can be used (say in an exposed corner or other exposed metal that's the same height as the bed surface) to figure out the offset between the inductive probe and the tip.
thank you
Z improvement and remote drive extruder....
I never saw so many heads explode when I called a raspberry pi a toy computer... I did not think the raspberry pi would end up with that fanboyism like it has in the 3d printing community.
Did he belt out Sledgehammer by any chance? Judging by how drunk he sounded he might feel like he got hit with one later.
Pls prioritiose what eva u has da parts for, that you enjoy doin, I love to watch when you're havin fun.
you dont need a raspberry pi for octoprint. i used my old laptop to run octoprint until last week.
True but Raspberry PI is very low power compared to an old laptop so that's why people choose it.
@@the_DOS thats why i swiched to a pi. but when you dont run the printer 24/7 a laptop can be an alternative when you have one.
36v stepper mod?
SKR should be easily wireable.
There are pretty cheap Mean Well 36v-100w passive cooled PSUs.
Actually, how much power do steppers draw?
CFD!! =D
Another idea for sensorless bed leveling: Use Trinamics Stallguard Feature (TMC drivers with SPI/UART required) to detect the Nozzle running into the bed. It basically measures how much resistance the motor faces when moving and you set a threshold for when it should trigger. Determining the threshold value might be finicky but once it runs it's good. Prusa uses this for X and Y afaik.
This would create imprints in soft print surfaces.
stall guard will also fight with the spring tension, it will not be triggered exactly upon touching the the surface. well unless you didn't use any sprint to mount your bed.... perhaps it will work.
For X and Y it's fine, but you shouldn't be using it for Z since stallguard won't produce consistent enough results in that order of accuracy that is required for bed leveling.
@@martijnpeters yup, because it really doesn't matter much if X/Y homed 1mm or 2mm off unless you are trying to maximize your print space. but for Z it is crucial since your print will not stick if it is too far off, or your nozzle will simply dig into the bed.
@@evertchin The accuracy is enough to re-home X/Y and then resume the print seamlessly after power loss or step loss. You can read out the load intensity with 1024 step resolution via SPI/UART. For the z-axis you could detect the gradual rise in resistance when the print-head starts depressing the springs. This requires a fast readout of the stepper drivers registers and some calibration though. It's not easy but it should be possible.
Check this out for further information: www.trinamic.com/fileadmin/assets/Support/Appnotes/AN002-stallGuard2.pdf