I ran those for over a year. I saw no appreciable acceleration increase either. It did let the gantry sag more easily when the motors were turned off, leaving me to set the after-print park position at 350, 175 so the next QGL didn't take forever. What I did find was that, with the GE5C mod, I could support the gantry evenly high, loosen all the bolts to the gantry extrusions and let everything find it's relaxed position. Then tighten all the GE5C through bolts, and the rest of the gantry bolts. That made the gantry more rigid between the Z rails, and led to more accurate prints with better finish in the X and Y surfaces.
I didn't really do it for the acceleration tbh. The stock Z joints tend to come loose after a while and require re-tightening. They are, as you mentioned, never fully tightened in the first place. As such, they will loosen even more just by running your printer for a little while. With this mod, you can fully torque down and even thread lock the bolts, and they virtually never come loose on their own. There's just less guesswork overall as well, since under/overtightening the stock Z joints can impact performance. Overall it's quite a nice mod, even if it's not totally necessary.
I agree, it's a very nice mod. It simplifies the z-stack and as you state, eliminates the guessing. But I can't help wondering if there is something else that might need to fix for better results. Thanks for the comments and watching!
When I did my GE5C mod. I used the bigger Block to keep my magnetic end stop sensors and double stacked the magnets.. I also double stacked x2 m5 washers to add some height tolerances and fixed the bolts firm but not loose and not extreme tight. I did it to help with a more level gantry. Never tested speed or resonance..
The end stop need makes alot of sense. But since relocating the endstops, I did not need it. If you get the chance, try resonance testing. I'm curious if you see an improvement. I'm still wondering if I should have seen an improvement (like some claim) or not.
Great video as usual. I watch your videos before everybody else's. Even months after you post they are spot on. I watch many of them many times. As noted below, I don't think it is making an important difference. However the simplicity and the fact that we can tighten everything now is better. My screws come loose after 20-30 hours of printing. So threadlocker is a good addition as well with this mod.
The next printer I build will have these as well. From an engineering perspective, they seem like the right way to do it. Thanks so much for the nice words!
I'm noticing that some builders have tightened down the z-stack once they racked and tuned the printer (I did). And some others have not. That last group might be seeing bigger improvements. Thanks for the comments and watching!
Thanks for making this video! I am also curious to hear if other people have further insights about this. I did the same upgrade about a month ago. I had very similar results and came to more or less the same conclusions. As you point out in the video it seems that proper tension on the the stock z stack should work just as good. Theoretically this mod alters the geometry of the quad gantry leveling calculations, but I haven't had any issues. It seems to do roughly the same amount of pecking as before...
Yes, looking forward to what others think. It is an elegant design and it simplifies the z-stack. It feels like in the long run, it's a better way to go. That said, my printer might have other issues. Thanks for the comments and watching!
Hello Daniel. It appears you left a message, but for some reason I cannot read the full message as I cannot get to it. Its about the EBB36 and resonance testing. Please post again or contact me on Discord. Thanks!
@@KapmansBasementWorkshopI left a message because I suspected that resonance testing with ebb36 was unreliable due to the mount vibrating more than the toolhead. But it seems that I was mistaken and it works well enough. So I deleted the comment to avoid any confusion. Thanks for noticing! 😊
I had purchased some steel GE5C bearings and I would not recommend them. I thought I'd check for play in the bearing/block out of curiosity. The bearing fitted nicely into the block as expected however once I clamped the bearing with a bolt, washers and a nut whilst in the block, I found that the inner sphere could move over 0.2mm inside the race. So I did not install them.
Hmmmm. I was planning on purchasing metal GE5C bearings to see if I could find any performance differences. Thanks for this information. I now need to consider this more deeply. Thanks for watching!
@@KapmansBasementWorkshop How are the ceramic bearings holding up? I have some on the way but maybe again will check before installing them and the blocks.
@@KapmansBasementWorkshop Update... my ceramic bearings have arrived and I have done a basic check for play compared to the steel bearings and there is no play. The steel bearings I had are obviously now poorly manufacturered and over priced. I'd recommend paying more for quality and directly from a repitable brand/source.
I fixed the gantry without bearing and it works perfectly. The only extra thing which I;ve done, I have used corners to make the geometry of the frame and bed squared. so bed levelling is natural by design and little to do with gantry adjustment. What I experience is gantry dropping after motors current off. so I need probably higher torque motors for Z
Kinda don't expect change here more as a reliable z tension when performing QGL with less friction at corner points. I believe this is the primary goal
I believe the gantry adjusts a little better wit these installed. That's about it. Also, compared to the standard solution, this feels better engineered.
I have used the metal version of this type of bearing. It increased x acc to 18k and y acc to 7k. But I just thought about how the bearings are inserted in the holder. That’s in y-direction. So it leaves perhaps for more wiggle room? Even though it’s a snug fit. Also Y direction will always be lower because there is more mass that is moving compared to x direction. (Voron 2.4r2 300mm, LDO 0.9degree steppers, run current 0.8A)
Hmmm. Wiggle room. You may have a point there. I also wonder if there is a real difference between metal and what I used. Now you've got me wondering...Thanks for the comments and watching!
Im considering this on my next teardown and realized there are metal and these version bearings. Interesting to see this small increase in numbers. Is this a stock build otherwise ? Your input shaper numbers seem on the lower end of most ive seen
I don't attribute any improvements in the resonance numbers to this change. But, my next build will use these as well. I think this is the correct way to allow the gantry to move as it eliminates all guesswork around correct tightness vs the stock design. I went for these vs metal as I have experience with these materials in the past and and like the fact that they are maintenance free. It's the largest version of this printer which means it will always be on the lower side due to all the mass, but this printer is so fast, I'm not sure it matters that much anymore. That said, there are quite a few things that can be done to lighten the x-axis and print head. I will get to those soon enough. Thanks for watching and the comments!
I have done this upgrade quite some time ago, indeed the improvements are not that huge, but for me it eliminated a lot of additional vibrations. In your graph it seems you have a lot of sub harmonics. Especial at the 5Hz region? For me I basically have 1% vibrations with the VZ shaper and everywhere else its zero. With the stock gantry blocks it was recommending me the MZV shaper with around 4500 accel, now I am on the VZ shaper with 6500. (and probably could go higher when I am willing to take more vibrations) Btw what accelerometer do you you run? I run it with a generic ADXL345 on a pico pi and there you have to pull the pmw pin high, the get more accurate results, because without the high power mode the noise in the signal of the ADXL is a lot higher than with it. If you really want to hunt for higher accelerations on the y axis you probably would have to switch to a mini stealth burner and a mini sherpa extruder and go for a skeletonized x gantry and that way saving a lot of weight. A stock stealth burner with cw2 is roughly 430g, the combo mentioned is 260g and the skeletonized profile would save another 160g saving a total of roughly 300g, also could switch back to a MNG9 rail for a few addtional grams saved.
While I got no performance increases, I really like this mod and I believe a 2.4 should be built with these standard. From a mechanical perspective, these are superior in every way. I think some of those harmonics are probably due to belt tension and a few items in the base such as the power supply where I believe I need a stronger mount. Slowly, I'm going to be taking those on one by one. I used to use a generic ADXL, but now I have one built in with the EBB36. I'm not sure what model that is. I need to see if there is a way to pull that pin high...thank you for this tip! I have some plans in the future as there are different print heads I would like to try. But it's a tradeoff between getting prints done and endless improvements/modifications to this printer. But the changes will be coming. Thanks so much for the tips, comments and watching! I will be keeping these in mind as I move forward.
@@KapmansBasementWorkshop yeah belts can be an issue that can introduce quite a few vibrations, maybe the results for my test are quite ok, because with the GE5C replacement I also changed most of the gantry parts to the pin mod (5mm pins instead of the bolts), including the alternative front idlers. Which also means I basically had to re-tension all the belts in the machine itself. Hope you can figure out and track down some of those vibrations, than again as long as they don't cause actual artifacts while printing they are mainly an annoyance, you know they are there but don't actually affect something. Edit: Also very important, the right table, my printer stands on quite solid table and on that there is a 60x60x3 Granit tile which is roughly 16kg. I added that after watching CnC kitchens video about noise reduction, which also basically reduces some of the vibrations of the printer. (czcams.com/video/y08v6PY_7ak/video.html)
@@hornet331 I have been thinking about a big stone tile as well. So it's good to hear that it helped with resonance too. I just went with pin mod last week. I have yet to retest, but it's printing well. It's on a very heavy bench. I don't this it's that, but I'll know soon. Thanks very much for the great tips!
Could it be that you missed the figures on the y-axis of you PSD plots? For your X-printer axis, it's down from 7.2*10e4 to 6.4*10e4, and for the Y-axis it's halved, from 7*10e4 to 3.5*10e4. According a thread in the Klipper discord, a higher PSD could translate to a stiffer system, which is not necessary bad. I think the rubber bearings are the cause of loosing stiffness.
It's been a while since I made these change. I think you have a strong point. It's actually gotten worse since then. Hmmmmm. Thank you very much for the comment! It might be time to take a fresh look at this.
You're welcome!😃 Maybe you'll find the following video of @3dprintersawhiteboard401 helpful as well. It's called: "How to read and analyze input shaper graphs generated by Klipper". I think this guy knows a lot, and some of his other videos are also definitely worth watching.
Interesting seeing the IS results of a floating gantry v2, on my v0 it reccomended me 7500 accel with stock gantry. So I guess the floating gantry looses some amount of rigidity, makes me wonder if adding some sort of reinforcement like 2040 extrusions for the gantry would help it
It might. But then I also wonder if the way the gantry is built with ABS corners may be an issue with the 2.4 at 350mm. I wonder if replacing those with aluminum would add that stiffness. That said, for a printer this size, it's still very fast....Thanks for watching!
I think you have other issues with your build, somewhere, as the percentages for vibrations are pretty high. on my 2.4, i have 0% for mzv on y, and 1.1 for zv, at 6800 accel on x, 0 at mzv and 2.4% at 14600 on x. for me, it tighted up the graphs alot, but i didnt expect to see 40% accel improvement. i dont have CAN
I tightened up every screw I could previous to this. I don't think that is the issue. But I don't know what it could be. I've been wondering if this could be coming out from under the printer from the electronics compartment? Thanks for the comments and for watching!
it recommends 3 washers on top of the ge5c bearing, not one just fyi. otherwise you won’t have full movement & it’s restricting itself with a flat surface hence the 3 washers. it only touches the inner bearing & giving you full rotation. you can see a big different in the allowed space the gantry is able move. i also had one washer & it was actually worse for me until i read the readme clearly & i’ve missed the 3 washers lol trust me that you’ll thank me ✌️
I only used a single 1mm thick spacer, which has an OD of 8mm (left over from the Formbot kit). There's enough play in it that the back of the gantry will sag a couple inches when the printer power is off. I'd say one each is sufficient. But more wouldn't hurt. 👍
I ran those for over a year. I saw no appreciable acceleration increase either. It did let the gantry sag more easily when the motors were turned off, leaving me to set the after-print park position at 350, 175 so the next QGL didn't take forever.
What I did find was that, with the GE5C mod, I could support the gantry evenly high, loosen all the bolts to the gantry extrusions and let everything find it's relaxed position. Then tighten all the GE5C through bolts, and the rest of the gantry bolts. That made the gantry more rigid between the Z rails, and led to more accurate prints with better finish in the X and Y surfaces.
This is a good idea. I will need to give this a try. Thanks you!
I didn't really do it for the acceleration tbh. The stock Z joints tend to come loose after a while and require re-tightening. They are, as you mentioned, never fully tightened in the first place. As such, they will loosen even more just by running your printer for a little while. With this mod, you can fully torque down and even thread lock the bolts, and they virtually never come loose on their own. There's just less guesswork overall as well, since under/overtightening the stock Z joints can impact performance. Overall it's quite a nice mod, even if it's not totally necessary.
I agree, it's a very nice mod. It simplifies the z-stack and as you state, eliminates the guessing. But I can't help wondering if there is something else that might need to fix for better results. Thanks for the comments and watching!
Appreciate the videos about Voron mods.
Thank you for watching and the nice comment!
When I did my GE5C mod. I used the bigger Block to keep my magnetic end stop sensors and double stacked the magnets.. I also double stacked x2 m5 washers to add some height tolerances and fixed the bolts firm but not loose and not extreme tight.
I did it to help with a more level gantry. Never tested speed or resonance..
The end stop need makes alot of sense. But since relocating the endstops, I did not need it. If you get the chance, try resonance testing. I'm curious if you see an improvement. I'm still wondering if I should have seen an improvement (like some claim) or not.
Great video as usual. I watch your videos before everybody else's. Even months after you post they are spot on. I watch many of them many times.
As noted below, I don't think it is making an important difference.
However the simplicity and the fact that we can tighten everything now is better. My screws come loose after 20-30 hours of printing.
So threadlocker is a good addition as well with this mod.
The next printer I build will have these as well. From an engineering perspective, they seem like the right way to do it. Thanks so much for the nice words!
Had the almost the same result as you, in the end I think this is sexier way for the coupling of the ganty
I'm noticing that some builders have tightened down the z-stack once they racked and tuned the printer (I did). And some others have not. That last group might be seeing bigger improvements. Thanks for the comments and watching!
Thanks for making this video! I am also curious to hear if other people have further insights about this. I did the same upgrade about a month ago. I had very similar results and came to more or less the same conclusions. As you point out in the video it seems that proper tension on the the stock z stack should work just as good. Theoretically this mod alters the geometry of the quad gantry leveling calculations, but I haven't had any issues. It seems to do roughly the same amount of pecking as before...
Yes, looking forward to what others think. It is an elegant design and it simplifies the z-stack. It feels like in the long run, it's a better way to go. That said, my printer might have other issues. Thanks for the comments and watching!
Hello Daniel. It appears you left a message, but for some reason I cannot read the full message as I cannot get to it. Its about the EBB36 and resonance testing. Please post again or contact me on Discord. Thanks!
@@KapmansBasementWorkshopI left a message because I suspected that resonance testing with ebb36 was unreliable due to the mount vibrating more than the toolhead. But it seems that I was mistaken and it works well enough. So I deleted the comment to avoid any confusion. Thanks for noticing! 😊
@@danielpeterson2017 No problem. I was worried it would appear I failed to respond!
I had purchased some steel GE5C bearings and I would not recommend them. I thought I'd check for play in the bearing/block out of curiosity. The bearing fitted nicely into the block as expected however once I clamped the bearing with a bolt, washers and a nut whilst in the block, I found that the inner sphere could move over 0.2mm inside the race. So I did not install them.
Hmmmm. I was planning on purchasing metal GE5C bearings to see if I could find any performance differences. Thanks for this information. I now need to consider this more deeply. Thanks for watching!
@@KapmansBasementWorkshop How are the ceramic bearings holding up? I have some on the way but maybe again will check before installing them and the blocks.
@@jchristensen2022 I've had no problems with them. Or at least not that I'm aware of. They have been working well.
@@KapmansBasementWorkshop Update... my ceramic bearings have arrived and I have done a basic check for play compared to the steel bearings and there is no play. The steel bearings I had are obviously now poorly manufacturered and over priced. I'd recommend paying more for quality and directly from a repitable brand/source.
@@jchristensen2022 Lesson learned. Thank you for posting this!
I built my 2.4 with the ge5c bearings from day one. I can't really imagine how these would help with acceleration on either the x or y axis.
Thanks for that feedback!
I fixed the gantry without bearing and it works perfectly. The only extra thing which I;ve done, I have used corners to make the geometry of the frame and bed squared. so bed levelling is natural by design and little to do with gantry adjustment. What I experience is gantry dropping after motors current off. so I need probably higher torque motors for Z
I did the same thing. That might be why I'm not seeing such an improvement. Thank you for the comments and watching!
Kinda don't expect change here more as a reliable z tension when performing QGL with less friction at corner points. I believe this is the primary goal
I believe the gantry adjusts a little better wit these installed. That's about it. Also, compared to the standard solution, this feels better engineered.
I have used the metal version of this type of bearing. It increased x acc to 18k and y acc to 7k. But I just thought about how the bearings are inserted in the holder. That’s in y-direction. So it leaves perhaps for more wiggle room? Even though it’s a snug fit.
Also Y direction will always be lower because there is more mass that is moving compared to x direction.
(Voron 2.4r2 300mm, LDO 0.9degree steppers, run current 0.8A)
Hmmm. Wiggle room. You may have a point there. I also wonder if there is a real difference between metal and what I used. Now you've got me wondering...Thanks for the comments and watching!
Im considering this on my next teardown and realized there are metal and these version bearings. Interesting to see this small increase in numbers. Is this a stock build otherwise ? Your input shaper numbers seem on the lower end of most ive seen
I don't attribute any improvements in the resonance numbers to this change. But, my next build will use these as well. I think this is the correct way to allow the gantry to move as it eliminates all guesswork around correct tightness vs the stock design. I went for these vs metal as I have experience with these materials in the past and and like the fact that they are maintenance free. It's the largest version of this printer which means it will always be on the lower side due to all the mass, but this printer is so fast, I'm not sure it matters that much anymore. That said, there are quite a few things that can be done to lighten the x-axis and print head. I will get to those soon enough. Thanks for watching and the comments!
I have done this upgrade quite some time ago, indeed the improvements are not that huge, but for me it eliminated a lot of additional vibrations. In your graph it seems you have a lot of sub harmonics. Especial at the 5Hz region? For me I basically have 1% vibrations with the VZ shaper and everywhere else its zero. With the stock gantry blocks it was recommending me the MZV shaper with around 4500 accel, now I am on the VZ shaper with 6500. (and probably could go higher when I am willing to take more vibrations)
Btw what accelerometer do you you run? I run it with a generic ADXL345 on a pico pi and there you have to pull the pmw pin high, the get more accurate results, because without the high power mode the noise in the signal of the ADXL is a lot higher than with it.
If you really want to hunt for higher accelerations on the y axis you probably would have to switch to a mini stealth burner and a mini sherpa extruder and go for a skeletonized x gantry and that way saving a lot of weight. A stock stealth burner with cw2 is roughly 430g, the combo mentioned is 260g and the skeletonized profile would save another 160g saving a total of roughly 300g, also could switch back to a MNG9 rail for a few addtional grams saved.
While I got no performance increases, I really like this mod and I believe a 2.4 should be built with these standard. From a mechanical perspective, these are superior in every way.
I think some of those harmonics are probably due to belt tension and a few items in the base such as the power supply where I believe I need a stronger mount. Slowly, I'm going to be taking those on one by one.
I used to use a generic ADXL, but now I have one built in with the EBB36. I'm not sure what model that is. I need to see if there is a way to pull that pin high...thank you for this tip!
I have some plans in the future as there are different print heads I would like to try. But it's a tradeoff between getting prints done and endless improvements/modifications to this printer. But the changes will be coming.
Thanks so much for the tips, comments and watching! I will be keeping these in mind as I move forward.
@@KapmansBasementWorkshop yeah belts can be an issue that can introduce quite a few vibrations, maybe the results for my test are quite ok, because with the GE5C replacement I also changed most of the gantry parts to the pin mod (5mm pins instead of the bolts), including the alternative front idlers. Which also means I basically had to re-tension all the belts in the machine itself.
Hope you can figure out and track down some of those vibrations, than again as long as they don't cause actual artifacts while printing they are mainly an annoyance, you know they are there but don't actually affect something.
Edit: Also very important, the right table, my printer stands on quite solid table and on that there is a 60x60x3 Granit tile which is roughly 16kg. I added that after watching CnC kitchens video about noise reduction, which also basically reduces some of the vibrations of the printer. (czcams.com/video/y08v6PY_7ak/video.html)
@@hornet331 I have been thinking about a big stone tile as well. So it's good to hear that it helped with resonance too. I just went with pin mod last week. I have yet to retest, but it's printing well. It's on a very heavy bench. I don't this it's that, but I'll know soon. Thanks very much for the great tips!
Could it be that you missed the figures on the y-axis of you PSD plots? For your X-printer axis, it's down from 7.2*10e4 to 6.4*10e4, and for the Y-axis it's halved, from 7*10e4 to 3.5*10e4. According a thread in the Klipper discord, a higher PSD could translate to a stiffer system, which is not necessary bad. I think the rubber bearings are the cause of loosing stiffness.
It's been a while since I made these change. I think you have a strong point. It's actually gotten worse since then. Hmmmmm. Thank you very much for the comment! It might be time to take a fresh look at this.
You're welcome!😃
Maybe you'll find the following video of @3dprintersawhiteboard401 helpful as well. It's called: "How to read and analyze input shaper graphs generated by Klipper". I think this guy knows a lot, and some of his other videos are also definitely worth watching.
@@wingunder, I will check that out. Thank you!
Interesting seeing the IS results of a floating gantry v2, on my v0 it reccomended me 7500 accel with stock gantry.
So I guess the floating gantry looses some amount of rigidity, makes me wonder if adding some sort of reinforcement like 2040 extrusions for the gantry would help it
It might. But then I also wonder if the way the gantry is built with ABS corners may be an issue with the 2.4 at 350mm. I wonder if replacing those with aluminum would add that stiffness. That said, for a printer this size, it's still very fast....Thanks for watching!
@@KapmansBasementWorkshop Maybe printing the parts in PC-CF would help add that stiffness without having to buy the custom parts
Audio messed up around 2:15
Yes, I just noticed. I need to go back and fix it....thanks for watching!
anychance of a voron tap tutorial?
Yes, there is a chance. Thanks!
Looking forward to that as well, my tap kit is on order! 💪🏼
@@lukasmatzinger I'm still evaluating the modification but will probably pull the trigger soon....
I think you have other issues with your build, somewhere, as the percentages for vibrations are pretty high. on my 2.4, i have 0% for mzv on y, and 1.1 for zv, at 6800 accel on x, 0 at mzv and 2.4% at 14600 on x. for me, it tighted up the graphs alot, but i didnt expect to see 40% accel improvement. i dont have CAN
I tightened up every screw I could previous to this. I don't think that is the issue. But I don't know what it could be. I've been wondering if this could be coming out from under the printer from the electronics compartment? Thanks for the comments and for watching!
What size 2.4 are you running?
I run a 350
@@MikeStammer I'm running a 350 as well. Thank you.
Mike - that’s a fast 350 v2.4 !
it recommends 3 washers on top of the ge5c bearing, not one just fyi. otherwise you won’t have full movement & it’s restricting itself with a flat surface hence the 3 washers. it only touches the inner bearing & giving you full rotation. you can see a big different in the allowed space the gantry is able move. i also had one washer & it was actually worse for me until i read the readme clearly & i’ve missed the 3 washers lol trust me that you’ll thank me ✌️
Can you send me a link to that? I have looked, and I only see 4 spacers in total. Thanks!
I only used a single 1mm thick spacer, which has an OD of 8mm (left over from the Formbot kit).
There's enough play in it that the back of the gantry will sag a couple inches when the printer power is off.
I'd say one each is sufficient.
But more wouldn't hurt. 👍
@@81XS11 I have the one spacer and I see some good spacing that might also give me a inch or so of change across the gantry.
I agree wit this. I found 2 washer was enough for me to have some axial alignment play
@@jc84com I may need to add a second and give it a try....