3D Printer Belt Tension Effects on Creality Ender 3
Вставка
- Опубліковано 12 гру 2019
- Chuck tests the effects of Belt Tension Effects on Creality Ender 3. It's often stated to keep the belts tight on your 3D printer but how tight is tight? Chuck installs belt tensioners on his Creality Ender 3 and tests different tightness levels to find out the effect. Check on the results in this week's Filament Friday.
X-Axis Tensioner
USA: amzn.to/38BKbYP
UK: amzn.to/38zuVf3
Banggood: bit.ly/2rJ0QZV
Y-Axis Tensioner
USA: amzn.to/2PhOdhi
or amzn.to/366jbiD
UK: amzn.to/35l6pMP
Creality Ender 3:
From Banggood : bit.ly/2wfD5Ia
From : amzn.to/3fSIISy
From MatterHackers: bit.ly/2VVveZW
Creality Official Reseller Comgrow: bit.ly/3hDTvSf
Filament Friday Tool Kit:
USA - amzn.to/2N8NuOR
UK - amzn.to/2OP6pNY
Filament Friday Filament:
USA - amzn.to/2rgz3Ma
Creality Ender 3 Parts/Upgrades
-------------------------------------------
Creality Ender 3 Upgrade Package: bit.ly/2MNvmcs
PTFE, Couplings and Cutting Tool:
amzn.to/2TgzRyP
CHPOWER Metal Extruder:
USA: amzn.to/2rpMJsn
UK: amzn.to/2YpeMUN
Banggood: bit.ly/2YnuwHO
WINSINN Dual Gear:
USA: amzn.to/37fB4eF
UK: amzn.to/36bUKju
Banggood: bit.ly/34YWDzN
EZR Extruder: bit.ly/2rlKCpp
Nozzle Removal Palm Ratchet:
USA: amzn.to/3u6ncm7
Premium Version with Rubber Grip: amzn.to/3qPs3pB
UK: amzn.to/2YnPqXo
Creality Silent Stepper Board: amzn.to/2YPM0Nh
Cap Tubes & Couplings: amzn.to/2Vf8BCf
CR10/Creality Ender 3 Nozzles: amzn.to/2m39K1y
CR10/Ender3 Silicon Heater Block Cover: amzn.to/2PaxJ89
Solder Sleeve Kit: amzn.to/2Pqqngm
Creality Ender 3 Fan Set (24v): amzn.to/38u9aOl
CR-10 Fan Set (12v): amzn.to/2PrcCxK
T-Nuts Kit: amzn.to/2LsQZg8
Screw Kit: amzn.to/2Va8uGK
Temperature Meter Used in Videos: amzn.to/2JpUJOC
Screwdriver set used in Videos: amzn.to/2PbWocM
Creality Ender 3 Upgrade Package (Extruder, Silent Board & Capricorn Tubing): bit.ly/2MNvmcs
CHEP Cura Profiles:
www.chepclub.c...
CHEP Prusa Profiles:
www.chepclub.c...
CR-10 V2
From Creality Official Reseller Comgrow: bit.ly/2WwT8gz
From Banggood: www.banggood.c...
From
From : amzn.to/2C2naiy
CR-10:
From Creality Official Reseller Comgrow: bit.ly/2PF01uW
From Banggood : bit.ly/2UdMcpc
From : amzn.to/2JDO9oP
CR-10S Pro
From Creality Official Reseller Comgrow: bit.ly/2JGLLh4
From Banggood : bit.ly/2WXWasO
From : amzn.to/2C4MGnw
From Matterhackers: bit.ly/2Kwk3G6
CR10 Mini
From Banggood : bit.ly/2u5XNJk
From .com : amzn.to/2l8B7mX
Creality Ender 3X - Creality Ender 3 plus Glass Bed and 5 Extra Nozzles:
Creality Official Reseller Comgrow: bit.ly/2NguWOD
Creality Ender 5:
.com: amzn.to/2IZaEFy
Banggood.com: bit.ly/2PwwHV7
Creality Official Reseller Comgrow: bit.ly/2L2SweG
Creality Ender 3 Pro:
From Banggood.com: bit.ly/2DBG0AI
From .com: amzn.to/2QUjRUL
From MatterHackers: bit.ly/2MfPCFT
Creality Official Reseller Comgrow: bit.ly/3hET7CL
Filament Friday Tool Kit: amzn.to/2YhMtea
Filament Friday 3D Printer Starter Kits: bit.ly/2EuojUq
Filament Friday Filament: amzn.to/2rgz3Ma
********* $3 - Ender Style Cura Profiles Download Links: *******
Latest V5.x Profiles: rb.gy/5k981c
TPU Profiles: shorturl.at/V08bk
PETG Profiles: shorturl.at/RLEfI
(Free to Patreon Members)
**** Support the Channel *****************
EBAY USA: ebay.to/1wtxfiU
EBAY UK: ebay.eu/2qdC5iy
EBAY Canada: bit.ly/2qdGRwh
Support the Channel here and Shop at MatterHackers.com and you automatically donate to the channel at no cost to you:
bit.ly/1Ugcijp
Buy thru this link to Slice Engineering for the Best Hot Ends:
bit.ly/3yuyuze
Note:
As an Amazon Associate, I earn from qualifying purchases.
Financial Support through Patreon:
/ elproducts
****************************************************
Filament Friday Sticker:
Send Self Addressed Stamped Envelope to:
Electronic Products
PO Box 251
Milford, MI 48381
****************************************************
Filament Friday Playlist:
goo.gl/yJLNhr
Chuck's Arduino Book:
amzn.to/3tA31JP
Intro animation by Pablo Licari based on Filament Friday Logo by Brian Carter. - Наука та технологія
On my Ender 3 my straight walls are pretty perfect but I recently started getting issues with curved walls. The surface quality on any curves was looking really crappy with very uneven leayer lines, zits/blobs and little pockets/voids. I honestly never though that my belts were the issue. They always seemed tight enough to me. Anyway, after watching this video I decided to try giving them a bit more tension. Bingo! The difference was like night & day. Thanks Chuck!
Great feedback. Glad I could help.
I guess I'm quite randomly asking but do anyone know a good place to watch new tv shows online ?
@Legend Westin thanks, I signed up and it seems like a nice service =) I appreciate it!
@Peyton Mohammed no problem =)
Just be aware that the bearings in the motor and wheels will wear sooner, but they are fairly cheap, so probably not a big issue.
Just got a printer recently and your channel has been a godsend! I go bonkers when reading instructions and hearing that a thing should be tight...but not too tight...Argh! Having you demonstrate the effect is amazing for a newcomer like myself.
Thank you for the effort put into making these videos and doing the experiments.
I was a professional tech for years and overtightened belts are horrible for your shafts and bearings. You'll be replacing them (or the whole assembly) in a few months.
I was really surprised that he said to keep them tight like that. There is mathematics involved in how tight they should be longevity sake and resonance frequencies!
By far the best belt tension video thus far.
seriously though, I don't think anyone has done a A/B comparison on belt tension.
One of the trickiest part of 3D printing is areas like belt tightening, where it needs to within a range. Not too tight. Not too loose. If you are too tight, problems can show up over time, especially if any of the idlers or pulleys are supported by 3D printed material. Good video, Chuck. Thanks.
In my experience "it' can't be tight enough" is actually incorrect.
I was getting 1st layer inconsistencies for a while (using a bltouch and a glass bed): things sticking differently on different parts of the bed, even when disabling auto bed leveling. I stumbled upon a reddit post from someone with those same issues, and one of the answers listed a number of items to check. I ran through all of them, and I ended on "check your belt tensioning". The author explained that while undertensioning is obviously bad as we all know, overtensioning can result in less predictable bed movement. I tried to slowly remove tension on X and Y belts and I found a point where my first layer issues completely disappeared.
Long story short: make sure your belts are properly tensioned, but don't over do it!
Why would the extra tensioning only affect the first layer? That's just bs.
@@alejandroperez5368 it can cause the axis to warp. since the following layers will take the countour of the first one, adhesion is not a problem, but since the bed is flat...
Wouldn't that be detrimental to the motorshaft if the belt is too tight?
Remember, sideward force on a motor will cause bearing damage or premature failures on motor bearings. Best for backlash is going with leadscrews, this way there is no sideway tension and the motor could last a lifetime.
Yes you are right that tightening too much may damage the bearings,
but the accuracy and the print quality is important and I believe
that changing the ball bearings of the nema17 is an acceptable
maintenance operation.Regards
@@pascalgarcia5161 sure thing but if you need that sort of quality, I'd rather invest in converting to leadscrews as adjusting backlash on them VS. changing bearings inside a stepper. Just a better option in the long run
Sure lead screws are more precise, but they are also far more expensive. The kind of printers we are talking about are full of mechanical aberrations that should be solved before investing in such a solution.
Regards
@@pascalgarcia5161 are lead screws really far more expensive? You can get them built right into the steppers if you want and hardware to interface with an axis is pretty cheap or easily printable.
They are not so far expensive, but transforming the printer designed with belts to use lead screws has a cost that is not justified too my mind as long as you do not solve the other mechanical problems and bad design. With this test we now that the belts are rigid enough to be tighten hard enough whthout affecting significantly the size of the printed object. Let us go for that on these printers without investing more than needed.
I am thinking building my own designed printer, and yes it will incorporate lead screws.
Two examples of default in design:
The gantry is guided at each end, and there is nothing done to ensure the 2 guide are parallel, one can not even access the screws when the gantry is in place, what would allow to tight everything in a better position.
The Z axis is often supported directly supported by the coupler that is not design for that and behaves as a spring. Some printers have a rigid coupler what is not good also as the coupler cannot cope with misalignment and involves extra constraints on the motor and the lead screw. I am in favor of normal coupler but with a 8 mm ball inside what solves this problem as the axial effort is not any longer supported by the coupler and the ball does not affect the capability of the coupler to bent. This really improves the accuracy of Z axis, and I advise everybody to do that. But then the gantry is still supported by the motor that includes also a spring stronger than the coupler of course, but still a spring that may have consequences if the gantry is too heavy or there is a small blockage on the axis.
We
I have experimented with that before on my am8 build and it destroyed my stepper motors. I took the motors apart and found the armature was indeed rubbing the field. Highly not recommended. Run another 50 hours with it that tight and give another update. Little steppers are not designed to handle that kind of side load. There is a formula for proper belt tension for a reason. The printer can only print so fast / quality no need to try and cheat and pay for it later. Be patient.... or spend $$$ on a printer made to do the job with speed.
This is exactly what I came here to ask. Every time I tighten my belts I wonder if the tension could bend the shaft, or see-saw torque the bearings inside. Now I need to go find this formula you speak of.
@@debren27
Most nema17 stepper motors have a radial load rating between 3-5 Kg. I could be wrong but I beleive I found something on it from Prusa's site. I ordered 2lb tensioner springs off Ebay and E3d stepper motors with there 9mm belts. Been printing smooth as glass ever since.
Always great info Chep Thanks a mill Dude!
Finally some real life tests and some decent information. Straight to the point. Thanks.
I'm using the same x-axis adjuster and had the gap when I installed it. I shortened the belt 15mm to fix. This gives a great improvement over the stock - the idler shaft is supported on both sides. Some people claim a toothed idler is better for the belt.
Nice video. Could you please make a followup to show what exactly you do to remove that ekstra filament on the bulging corners? Thanks.
I plan to
There are belts with steel core. You can tight them as hard as you can. However, the downside is the bearings won't last long.
I have worked in robotics design for ~10 years and tight enough it dependent on speed and backlash, if its tight and moving slow its fine, if its moving fast you need more... you are slapping a rubber belt, the faster you go the less slack you can have
Interesting, thanks for taking the time to try this out and share it!
With them super duper tight you can damage the bearings in the idlers or even the motors themselves (learned that the hard way). My method now is to get them tight enough to be taut, but not super stretched. Your belt size and quality can effect this too, so you just have to do a little trial and error. The boogers on the corner of the cube are probably from using cura 4.3 or 4.4 as they have disabled retractions on the outer wall (even if it is enabled) due to a bug they are working on (don't understand why as with the supposed bug it seemed to work fine so no idea why they disabled it internally).
Thanks.
Where do you buy good quality replacement belts? Tons of videos about belt tension but NONE about how to replace, what size how long and some decent links on where to get. Please consider this topic.
I'm in the process of upgrading a CR-10 and have used polyeurathane belts with steel belts. I've fitted a metal adjustable tensioner on the X axis, and a bearing support on the notor end to prevent the tension bending the motor shaft. On the Y axis, the standard metal 'tensioner' (moving belt idler) seems tough enough, but on the motor end it has a motor support (it's a bit tight to fit a bearing support for the tip of the shaft). All these add up to a lot more tension in the system.
Chep: you really need to adjust the length of that X axis belt; if not replace both X and Y with higher quality belts. You can see your new aluminium tensioner deforming as you tighten it. A little shorter belt and that far side would butt up against the extrusion and prevent this.
Could we see how you set up the additional motor support? Thanks!
Steel core belts bent around small pulleys will fail quickly. The steel wires will break inside the belt and then the polyurethane will stretch. Print quality will quickly go bad. You'll know this has happened when you find yourself retightening the belts every couple days. Steel core belts need very large pulleys if you want them to last.
tightening it further than necessary will ruin your motors because of the torque that the bearings/bushing aren't supposed to feel. it might even really cause the internals of the motor to rub on itself.
Do you know the rated torque that can be applied sideways to the motor shaft before it causes any main issues? In curious as to how tight I should get my belt but the internals of the motor should be fine. Bearings are not meant to turn inside motors. Before the internals ever rub up against each other you'd need enough force to actually rip the motor off its mount on the printer
How is every problem I try to look up, accompanied by your videos. You are inevitable...
Thanks for that info. You solved my mystery questions. So it was the belt that wasn't tight enough. My prints had wavines,or I had suddenly printer nozzle offset and start printing different location. So it's better to have it tight. In worst case I can replace the steper motor if it breaks down, but at least I'll have excellent prints.😄
I'm kinda late in the game but I'd like to add that for those using motor dampers, too much tension will pull the motor at an angle (via the rubber between the dampers).
Can't agree more. I printed a part using PETG on my Y-Axis so I can use the damper without cutting the bed level knobs and that too started to warp. I caught it just before the belt was about to slip off the pulley. Im removing all dampers and my new SKR 1.4 Turbo board and new drivers come in 2 days. Should have skipped the dampers all together.
I overtightened my e3v2 belts. The problem I ended up having was vibration. It seemed that the tighter the belt, the more artefacts I got on smaller details due to vibration. I loosened the belts by 1/4 of a turn and I got perfect prints again. On large objects with lots of straight faces, you don't notice it. On small parts it would cause them to break off. Especially thin walled supports.
Honestly i think these adjusters just add more points that might become loose later. i agree no tight is tight enough but its better to keep the set points to a minumum bc otherwise you loose stability of the axis which is the greatest game of ender printers over others. Again great video, thanks.
I have an s5. I have recently had issues because I tightened the belts too much. One thing is the motor mounts aren’t the best. Too tight also puts extra wear on your motor bearings if you did have good mounts. I think I caused layer shift on my last print due to x-axis belt being too tight. I haven’t nailed down the root cause yet, but that is where I am planning to start my investigation. It is tough to figure out the correct tension.
I agree. They should be tight, but not overtight.
I have also had a bearing fail in a motor and had to replace it. So take caution as you can overtighten and create more problems later.
another great video Chuck. belt tension is something i have been considering too recently!!
About belt tension and motors/bearings.
Gates, who make Gt2 belts, say 30 Newtons for best precision.
NEMA says 27 Newtons on the pulley or you fry the bearings.
27 Newtons on the belt is 54 Newtons on the pulley!
Pulley tension is belt tension 2x, mostly.
If we honor NEMA, 13.5 Newtons is the best belt tension we can expect!
Or is my assumption wrong?
Excellent - now how do I measure this cheaply?
CHEP, After watching this video because I was having problems. I had an ah ha moment followed your advise tightened my belts and had a great print Thank you.
Me too! Just upgraded to a all metal hotend and remembered that I also messed with the belt tension when I saw this video
I've had similar results with tight belts but also had some bearings die after running them super tight for a few weeks.
Thank you. I have always made sure that they were not as tight as in your video. I'll tighten them up a bit more to see the results.
this is great... ringing and loss of quality imo is caused by loose parts and belts in most cases... when you print same file for years with same printer and you get ringing etc to me it makes little sense its a software issue and can't be a g code issue...thanks for sharing... i came by to check this out before i go through my machines to try to improve ringiing issues ..
my prints have been suffering from an undiagnosed, slightly detrimental to quality and getting progressively worse problem. I never looked for this video it came up in my suggested as I have watched others you have uploaded. I'm sure it is exactly what i needed to see. im getting these tensioners and i have liked and sub'd thankyou
Edit: I will be sure to use your affiliate links when i come to buy them
Yes, sometimes a loose belt will cause an occasional shift also.. then keep printing normal again for a few layers then slip again.which can mess up print.. this depends on your print... tight belts are a must... my ender 5 only needed this done once after several hundred hours of use... i like that fancy hardware,but it really is not worth doing for me...
it can be too tight. belt wear and bearing/bushing wear in the moving parts and motors are the result. tight enough is good, no need for a death grip.
Keep in mind, there is 'too tight'.
Draw two circles, one to represent the inner bearing race, one to represent the outer. Draw a circle in between them, touching both, and realize at that point of the bearing, they can only get within exactly 1 bearing diameter, the bearing keeps them apart.
Now instead of 1 bearing at that point, put a bearing before and after that point, off to each side. Realize the two circles can NOW be pulled in closer than the 1 bearing diameter.
Ball bearings do not form a perfect circle to follow the perfect circle of the races. They form a sort of point polygon, if you set high tension and pull the outer race in close it will destroy the bearings or races as the races have to ride over the high spot of the bearing.
This is why bearings have loading limits and similar, and get destroyed really fast with high loads..
Long term if you want higher precision you need higher precision linear items.
Since these bearings are fairly cheap, you can set the tension high, get great results, and just consider them a wear item. That's perfectly fine too, just understand you're shortening their life and they'll need wear replacement..
My 4 year old CR-10 is still very smooth and works perfectly, probably way better than most. I can easily turn all of my POM wheels with two fingers, barely any force needed. They are linear guide items that need no excess play, but don't need tension. Same for the belts, easily deflected and low tension, but very low play in the positioning direction. Properly used bearings have a near indefinite lifespan for something as low load as a 3d printer.
If I needed it I'd wring that last bit of precision as well, and just consider the wheels, bearings, etc wear items. But if you do it, do understand you're going to be intentionally point overloading the bearings, and they'll changing now and then.
So don't hesitate to do it for the extra results, it's a lot cheaper than the next step up all around for high precision in a whole printer. Just keep in mind there's a trade off, and more maintenance required..
Also, notice there's still a small, regular pattern on the prints. I wouldn't be too surprised if that's the bearing peak pattern showing up in the output.
Keep in mind, these belts stretch.
@@FilamentFriday 'these belts stretch.' Of course they do a tiny bit. Most of them are fiber reinforced or better, they're still plenty capable of overloading small bearings. Small steel wire is easy to bend, it's similarly easy to put pressure on a small bearing and have the 1 or 2 balls that support load from any one direction be digging in to the race slightly. Bearings have a wear curve, and wear shoots up very quickly once you exceed a certain loading, and it will be amazingly low for small, cheap type bearings.
At the 4:00 in this video, you can still see a vertical line pattern in the surface. Bearing rolls along, then forced over a bearing. Rolls along, whoops over a bearing. You can see the bearing high spots in the output, almost guaranteed that's what that vertical pattern is coming from, one of the bearings as it's rolling along.
It would be interesting to get a scale in there, see exactly what the bearings are loaded at. Then put that weight on a bearing with a test rig set to move it back and forth slowly. Small bearings don't like loading at all, they probably wouldn't last a day or two of accelerated testing before they're obviously worn..
1 pound of force is relatively low on a square foot. Much higher on 1 square cm. It can easily become steel surface destroying when you're talking about the near point contact patch of a tiny steel ball in a small bearing. Small bearings do not like loading, it is part of scale effect.
@@ModelLights There has to be specs for the steppers on how much sideways force they are rated for. it's not just the adjuster bearings taking the force. There has to be a point of negative, or at least non positive results from over tightening.
Thank you CHEP (and youtube algorithm), I upgraded to an all metal hotend and got weird unexplainable quality issues. Suddenly remembered I messed with the belts a bit too when I saw this video, looked at those again, made a few tests prints and sure enough :)
Thanks a lot for testing it, man! It really helps! 😃
Great advice as always Chuck
This upgrade is available on thingiveres and 100% printable
lil hotdog yeah was thinking the same thing, why not print it?
@@esqpain123 metal vs plastic
One of the first things i did while assembling was to sand the rough edges from the inside of the profiles....
I so appreciate your channel. You have truly been amazing for sharing your knowledge of 3d printers and the Ender 3. Thank you so much!
Do all the stepper motors have ball bearings or some cheaper ones have sleeve bearings? Tight is right but I wonder if that much load will effect the life of the front bearing in the stepper.
The good thing is nema 17 stepper motors are only $10. I'd thing most people would sacrifice a shorter lifespan rather than print wavy parts (I would, at least). 🙂
@@3dPrintingMillennial Oh for sure...I agree. Just dumb stuff that runs through my head.
The belts will stretch quicker but over tightening. It should be tight but you should be able to push the belt down a little bit. It shouldn't be tight like a guitar string.
Note the Ender 3 Pro has the wider channel and the adjusters won't fit the Y axis.
They make these adjusters for the 40x40 size extrusion beam that the Ender 3 Pro uses.
Yes, it’s good result when the belts have a maximum tension, BUT, it have a very bad influence on the idle pulley’s bearing and it could broke in the near future.
There's got to be a rated optimal tension on these. Cranking as hard as you can sounds like a great way to &^%$ up your gear, IMHO
loose belt are increasing skid at sharp corners.
tension also affects the axial scale of the model.
too much tension are increasing wear out and leads to sudden belt or pulley/motor bearings failure.
nuclearzerg Why would tension change scale? There's no reason it would do that, steps are steps, all the motion is relative and still controlled by the same system independent of tension.
It may cause accelerated wear, but certainly not sudden failure. Bearings wear faster with more loading but don't suddenly fail unless you somehow manage to crush them, which you won't do with a rubber belt on this scale.
@@boarder2k7 try to tighten the belt when printing, you will notice how the layer becomes wider.
@@nuclearzerg Is this based on the logic that a stretched and tighter belt then covers more space per ridge?
Another great video, I like the adjusters, since the manual way is tricky. I have also fount that tight is better. I converted one of my Ender 5's to linear rails and found that they need to be tight to perform correctly. Keep up the good work!!
Might be true for an ender 3. But I think this is bad advice. On all of my CoreXY and Cartesian printers. I run tension just high enough to not skip but as low as possible otherwise to save my motors and bearings and stop terrible vibrations.
Gates, the creator of the GT2 belt, recommends a tension of 6 to 8 pounds of tension when using a 6mm GT2 belt for registration like in a printer.
Your post is more important than it may seem as Gates are pretty much king of industrial belt drives. Thanks for looking up the data so I don't have to!
i think the gates belt doesnt need as much tension since it uses aramid fiberglass reinforcement.
Pretty daring to buy the Y tensioner that had a one-star rating and was only brought to market less than four months ago.
Thanks for the video! I am going to order those tensioner brackets myself. Very nice. I've seen some I could print myself but I am concerned they wont hold as well over time as then metal ones you featured.
How did you end up fixing the over extrusion on the edges/corners?
Gosh this guy's videos are so useful
Mechanical setup is just as important as software setup, its not just belt tension, its making sure everything is square and perpendicular, the roller wheels arnt worn, the rolling tracks are clean, the rollers arnt tight etc etc.
CHEP I feel some tensions here. LOL. Interesting video. Thanks.
Just belting things out, are we?
@@theginganinjaofficial yes right, 11 months later, we need to sit tight and keep the BELTs great to avoid strange situations.
Looking forward to test it out.Great videos as always. Thanks for all the research @CHEP.
Wait, why didn't you print one?
Excellent episode Chuck!
This video made me rip my x axis belt in half. Thanks a lot :)
@Joseph Joestar Thanks, and now I've been running my ender 3 for over a month on a not so tight zip tie. And it is printing better than before...
So, yes the belt will break. My Y axis shredded on the Ender 3 a couple months ago. It was only a few months old. I didn't have the belts overly tight. Not sure why but it broke at the motor end.
Motor end gets all the wear is why.
Educational as always! Keep it up! 👊🏻
The belts come about 2% undersized, so they need a fair bit of stretch to be set at the belt's intended tension, so you get them taught, and then stretch them a further 4-5mm, then you're set. These poly fibre reinforced rubber belts (as opposed to steel reinforced PU belts) won't actually break. But they can damage the motor bearings! I don't think those motors are intended for this kind of side load that the belt expects of you, you expect some kind of coupler to a shaft that's supported on both sides which then carries a pulley, which would be a little too complex for a low-cost desktop 3D printer.
But i don't want to make it into a big issue. You can just open the steppers and replace the bearings when time comes.
I'm not a fan of the x-axis belt tensioner only being mounted to one side of the gantry. It just seems to cause the slightest bias when tight. Anyway.
I'm having an issue with layer height after about 8cm up. The extruder seems to hit the print, and the entire thing skips back on the y axis. I get this layered stepped effect for a couple of layers. Then it's fine for 1cm, then repeats the issue.
I'll try the belts (again), and I really like the idea of having a tensioner. I'm ordering these.
And I'll make sure everything is tight, and no bolts are loose holding the frame together.
It has a bracket on the other side, that is supposed to butt up to the gantry (no fasteners). The problem here is that the travel distance of the tensioner pulley, vs the length of the belt, means you have to pull the whole bracket further out than intended, to enable it to tension the belt. Then the rear bracket is no longer contacting the gantry. As the side, brackets are screwed on, you could opt to print a longer bracket. Could do the other side bracket too.
In regards, to your Z axis issues, have you checked the condition of those lead screws? That they're not gunked up from filament rubbing on it, other gunk, deformed, or lack of lubrication? Is there something weighing/pulling on the print head, once it starts getting to those Z-heights?
I did wind up going over the entire printer, checking all the bolts. Four bottom bolts were loose. Tightening those, along with getting those pullies extra tight, and enabling z-hop, eliminated my delamination issue.
But lubrication is something that I should perform. It's been a year of lots of printing.
I got a ender 3 pro, but heavily modded
, and will certainly test this when i get my E3D V6 heated block replacement and new LGX extruder, once with the old recommended low G twang people say to use, then another cube with the tighter and see if cube changes.
Can't believe i didn't find this video until now lol
Good work Chep. I think you should also consider the fact that you changed from smooth pulleys to toothed pulleys. Running the toothed side of the belt on a smooth pulley can cause surface ripples, so I think that might also be a factor.
I’ve tested that theory by flipping the belt and saw no improvement on a smooth pulley.
Thanks for the video, I wonder if any1 has tried a 3D printed belt tensioner? How does it compare? Probably going to buy these
There are many on Thingiverse. I wanted a strong metal one to experiment with.
@@FilamentFriday I agree with wanting a metal 1. Though 3D printing is the most popular way for us guys so I think I'll try both.
Cars use plastic belt tensioners now so I am sure a 3d printed 1 will work
@@wires4auto just make it beefier than this one. The original prusa printers actually use 3d printed y axis holders
Thanks for this informative and extremely useful channel!
I bought the tensioners linked, and the X fell apart withing about 6 hours into a print, causing the belt to slip and creating a significant layer shift. This was with the linked "Pulley Version". The bearing used in the Pulley Version is very poor quality, and it only has one. The other side seems to be just a small presses in grommet, which is was fell apart. I removed the bearing and printed a 4.9mmOD x 3mmID x 8.5mm L bushing replace the bearing, so hopefully that helps.
hmm... smaller pulleys than the original larger ones.... that would bring the belts slightly closer to the center of the inside of the profile and possibly rub... I'll stick with the standard manual belt tensioners on the Ender 3 Pro... wish the Ender 3 V2 had those same mounting holes and option to purchase the traditional belt tensioners...but we'll see how long these plastic V2 tensioners last. : / Yeah, too tight, not good... too loose, obviously wrong period. Great videos as usual!
I was actually working on sorting out how tight the belt needs to be. I was seeing some articating and binding when the belts got too tight -- and that was not making sense. The thing that happens when you tighten the belts is if your idler is not smooth as silk, the gantry will show binding as the idler is struggling against the tighter belt -- super tight belt will expose other issues. I just re-assembled all my idlers with smaller shims so they can be snugged up and still spin freely -- that's really important for a tight belt.
DON't OVER TIGHTEN OR TIGHTEN THEM "AS TIGHT AT YOU CAN GET THEM"!!! This caused my mainboard to catch fire due to the extra stress on the drivers.
Too tight is a major problem and nobody seems to understand that. My Anet A8 prints perfect with the belts fairly loose. My Wanhao Duplicator i3 came with the belts fairly loose and it prints great, so I used it as an example on how to set my A8 belts.
No, that's not how it works. The stepper drivers on the board are constant-current devices. You can damage them if you wire the motors incorrectly or if the wiring breaks, not due to high tension. Still, if you run high set current, you also need a fair bit of cooling.
@@SianaGearz they were wired correctly. The over tightening caused failures in the motor itself and cause a short that cause the main board to fail.
@@BreakingYTown That's not plausible. Each output on the driver chip has its own current sense shunt, using which it limits the current and thus power to a safe set value. It fails however if the shunt is bypassed, i.e. the winding outputs got connected together. But to cause the sort of insulation failure in the motor that would potentially connect unrelated windings, you'd need to get the inside 170°C hot or so. It would probably reach well beyond 90°C on the outside, which i mean, you think you wouldn't notice before?
It was probably an unrelated failure. Sometimes things just fail due to a latent defect, not because the use was inappropriate.
@@SianaGearz I wish I still had the stepper to show you. You can have your assumptions. I go off what actually happened
Great Video! I'm new to 3D printing and have found your Videos very helpful especially with bed leveling But just one question,will this shorten the life of the belt ? Because if you're almost maxed out on the tension screw ,where do you go from there?
I bought the LEOWAY version of the x-axis tensioner because that’s what I could get. It came pre assembled. I installed it and found that the belt road high on the tensioner side and rubbed in the extrusion. I tried adjusting to no go. Then I figured maybe it was assembled upside down because the hole the pulley rides on is high. The hole for the tensioner screw is offset too. Re assembling let me to no fix. Sadly I am back to stock
CHEP: The tension can't be raised too high
All Commenters: NOOOO CHEP CUZ BEARING DAMAGE!!!
that makes perfect sense...theres a sweet spot which is best for machine..no reason to over tighten belts...
Mine have been running for years. No bearing damage. Belts stretch long before that happens.
I also have some wavy prints on my ender 2, I also need to sort out my da-vinci 1.0A your videos are a big help as alway's!!!! Keep well.
4:02 I also have those round bulged corners, as if the extruder, right before the corner, went out a little, made a half circle and then came back in. If you find a fix for perfectly sharp corners, or at least round corners that don't bulge out, please share. Thanks!
PS: I already have the printer frame perfectly square with shims, tight belts, new marlin, e-steps calibrated, extrusion flow compensated (90%) for a perfect 0.21mm wall and still get those bulged out round corners at 80mm/s (40mm/s outer wall, 60mm/s inner).
I've been experimenting with Klipper settings to get those bulges tuned out. Square Corner Velocity is interesting and seems to be getting the corners better. I tried getting pressure advance to do anything useful but it doesn't seem well suited to bowden tubes. I wasn't getting great results with jerk settings on the stock firmware so I went to Klipper, which doesn't actually do jerk in its motion planner.
I would like to see the exo slides with the belt tensioner
Hi Chuck. Followed your advice with purchasing the winsinn tightener. I must say that this tightener is terrible. in order to use it you have to do an extra fitting and eventually you can not get it much tighter than using the original tensioner. the winsinn tensioner is not designed properly to get the needed tension. I can relate to the need of putting tension to the belt for better results though.
Which axis?
@@FilamentFriday the X axis
It also depends on the belt itself, if its a steel cable or fiberglass cable variety, there are other kinds of belts as well. I know the Tronxy printers use a steel cabled belt.
I must be the worst at installing upgrades cause my prints have just gotten worse after doing them on my Ender 3 :D
You don't need these belt tensioners. The firm tension is more than enough. It is much more important to get vibrations from the print. My observation is that the more you tighten the belt, the more vibrations are transferred to the print!
Not all "Upgrades" are a good thing. Some will fuck up your previous tuning. Don't mess with the Zohan.
Very informative, I'm going to thingiverse now to print me some belt tensioners. One of the things that's hard for me is match the visual problem with the right cause. Your video shows the result of loose belts nicely. Next problem will be Z-binding which I can't resolve even after all sugestions tried
Nice, could you make a test with unitta gt2 belt and high quality pulleys, then in my experience, fingertight is enough. When printing fast 100mm/second or above you will get gt2 aretefacts on your print.
Aren't there ones online you can just print that do the same thing?
Yes. I have used a printed version for over a year (made with PLA with about 50% infil).
This is why I'm doing adapters to take the tension out of the stepper motors from my CR10S, so I can apply much more tension.
Right now with the dampeners, etc they twist and the axis sufers a lot of force.
I've been wondering about this. Thanks!
Dam.... I get problems with my ender 3 and here is chep with video how to fix it 😅🤣 ty chep
I'm using a printed belt tensioner for a while but until now never tightened this much.
Will try to tight a little more.
There is only 2 solutions : keep them belt's tight or go to a bearing rail system. Again like with everything maintain your Equipment proper and you get very good results.
That was something I've been wondering about, not touch the tension on my Ender 3 since I built it, So will check that out.
This was an awesome video. I was looking for examples of what a loose belt could cause. I have to check mine. I am experiencing a strange issue where the right walls and bottom right fills of any object, no matter where it is places, happens to each when I print multiple don't squish together. I think it might be a loose belt but so far no luck with x height or clogs or what not. Very odd.
How hard is that on the stepper motors? Thanks Chep, love your vids brother!
I think the biggest damage would be inflicted upon the ball bearings inside the stepper motors. Better prints maybe but more frequent maintenance on replacing stretched belts and worn out bearings.
As much as any force on the motors. It’s a flaw of direct drive of the belt.
There are printable belt tensioner on thingiverse that Work well
thank your videos help so much thank you just got my x axes tightener i wish this came with the ender 3 pro
Great walk-through
Thanks for sharing :-)
Thanks again chep really helpful as always ✌
I heard brand of the belts can make a big difference as well. Like the generic cheapo ones you get from aliexpress or banggood the fiberglass is thin and weak and can stretch much more or even snap. I hear the Gates brand is where it's at quality wise, but I need to order some (apparently available through E3D or Filastruder)
thanks for the info, but I manually tightened the belt instead of buying new parts, it should work
Just thinking out loud, but it seems a spring needs to be incorporated into the belt tightener. The spring could maintain a nearly constant force on the belt even if it stretched. And with graduations on the tightener and using springs with a consistent spring rate, the same force could be dialed in on any printer with that belt tightener.