Not sure if z wobble or z banding or just object too small it sways at high heights (read for more detail)

Hi all, so I finally finished building my BEAST HEVO (hypercube evolution) (with a build plate of 500 x 500 x 780) and am currently doing some tests/torture tests when I see this "problem"

So the thing passed the calibration cubes and 1st layer test flawlessly almost (the x and y are 0.05mm within each other and I get 20.05mm and 20.04 mm on x and y )

So I started testing the height to see if there's possibly any z axis issues (being such a tall print height) when i encountered some problems.

The first print was t somewhat thinner wall cube (NOT the calibration cube) that was extended to 630mm, and the problem was the thing was small and thing that it itself sways and wobble a bunch at that height so that didn't work.

Then I tried the spiral vase to 610 mm and there was no "layer shifting", but I noticed this:

[www.dropbox.com] (this is a video)

so there seems to be some small banding? however it doesn't appear until around after 300mm ish, so I'm not sure if it's actually z banding or is this vase once again too thin that it gets affected.

[www.dropbox.com] (some more pics)

any help would be greatly appreciated, right now my lead screw is like this at the top:
[www.dropbox.com]

do I possibly need to have to design a bearing holding to hold the lead screw in place? Any help is greatly appreciated, thank you very much!

Normally Trapezoidal T8-2 single start lead screws are fairly true.

The Lead Screw couplers cause a little Z-wobble, put a single wrap of electricians tape on the end that slips into the coupler to eliminate the Wobble.
On the other end you could use a bearing support holder to secure the top of the Lead Screws.

"sample Top Lead Screw holder with bearing"

---

Computer Programmer / Electronics Technician

The XY mechanism is at the top of your printer. The bed plate is held in place by the guide rails that run from the top of the printer to the bottom. When the printer is running at the start of the print, the bed plate is near the extruder and the top anchors of the Z axis guide rails. In other words, the bed plate and XY mechanism are very closely coupled. As the XY mechanism moves during printing, the top of the machine shakes more than the bottom (because the bottom is coupled to the floor). At the start of the print, the bed shakes along with the XY stage and the print quality is fine- the print is tightly coupled to the bed plate. As the print gets taller, the bed gets further from the Z axis guide rail mounts at the top and gets closer to those at the bottom. The bed decouples from the XY stage and while the XY stage is shaking away at the top of the printer, the bed is more solidly coupled to the bottom of the printer that isn't shaking. The result is that the print quality on tall parts gets worse as the print grows taller. Tall prints with a small footprint on the bed will suffer more because the print is going to wobble on its own, too.

You may also have a problem with the lead screws top constraint. If the screws aren't straight, as the bed moves down away from the lead screw top bearings, the screws are able to shift the bed plate laterally. As the bed moves to the bottom of the of the frame, the lead screws are less able to shift the bed around again (but then you have the frame rigidity problem described above).

If it's a frame rigidity issue, bolting the frame to a wall may help (but has extremely low WAF!). If it's a leadscrew problem, relaxing constraint at the top of the screws may help, or redesigning the way the nuts attach to the bed plate -allowing the plate to slip laterally on the nuts- may help.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Oh I forgot to mention I'm using T12 -2 single start, so it can support the heavier weight, but I guess that doesn't matter, ah I'm using aluminum flex couplers, should I maybe try spider/jaw couplers? I'll try the electric tape method, do you mean like this or actually take the thing out and put it on that end and then slip it back in? Thanks
[www.dropbox.com]

@the_digital_dentist I don't have a bearing holder at the top, it's rotating freely at the moment, I'm hearing two arguments one saying I should add a bearing support, and others saying no should rotate freely >.<

Edited 1 time(s). Last edit at 01/24/2019 10:19AM by iamthebest22.

Yeah, it's like that. Try it both ways and see if either way works better for you. It depends on a lot of factors (are the screws and motor axles in perfect alignment, etc.)

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

eh sorry when you said trying it both ways are you referring to spider jaw couplers or having a bearing or not at top?

This has already been discussed so many times, yet always the same basic design error like the bed on 4 adjusting screws

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

Quote
iamthebest22
eh sorry when you said trying it both ways are you referring to spider jaw couplers or having a bearing or not at top?

Yes, all that.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Okay so the bearingless holder didn't do much, changing it from aluminum flex to Jaw/spider/plum coupling though, it seems to have made a huge difference, *fingers crossed* but right now the z wobbling is almost gone, it still "seeable" if I shine a light on it, however it pretty much not sensible by touch.

My question then is this, I will be trying it without the bearingless holder as a last check, but about the jaw coupling, I've seen a couple of videos, but they don't really specify, so what I'm wondering is this:

as you can see in these pictures [www.dropbox.com]

and [www.dropbox.com]

are the lead screw and motor suppose to go in all the way and touch the red "spider" part, or is it suppose to be like the aluminum flex coupling just 5mm to 6mm or just barely past the set screw? Right now I got it just barely past the set screw and NOT touching the red part. Thanks.

Edited 1 time(s). Last edit at 01/26/2019 05:31PM by iamthebest22.

so an update now, took out the bearingless holder and it doesn't seems to make much of a difference though possibly slightly better. However I think I found part of the problem (other than the original aluminum flex coupler):

I think I have to slow down my print speed.

So I'm currently using around 80-90mm/s print speed, now not all is using that speed for obvious reasons, but when the perimeters and infill goes at those speeds, I've noticed my bed shakes well quite a little bit as this video will some what show you (not very clear as it's vibrating too fast to capture on camera)

However the vibration does not happen if I go at 60mm/s or slower, so I'm probably gonna have to do that for now.

With that being said, is there any way to eliminate these vibrations, as these are most likely what's causing my print to possibly have more "z wobble " at the top and the things to sway more.

Here's the video (again it doesn't show the vibration too well as vibrating too fast for camera to capture all but it's there you can see if looking closely): [www.dropbox.com]

Can you post a picture of the whole machine? Is the bed cantilevered?

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

oh it's a Hypercube Evolution, a corexy, here's a pic of the whole thing:
[www.dropbox.com]

Having 4 guide rails makes about as much sense as using 4 screws to level the bed. Using 4 rails means that any slight misalignment (and there will beat least slight misalignment) will cause the bed to either bind or walk as it moves. Walking is tilting back and forth as it moves up and down, which could cause print defects you're seeing. As the print gets taller, the lateral displacement at the top of the print increases because the print is tilting with the bed.

One round guide rail provides vertical guidance, but the bed can rotate around it. Adding a second rail, parallel to the first, stops the rotation. That is all you need. Using 3 or 4 rails is asking for alignment problems. I would put two of the rails close to two of the lead screws.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Hmm okay, just thought I'd put in an update though, here's a print WITHOUT the bearingless top holder, and it looks like i should put in the top holder, as interesting if you you, you'll see that the thing "layer shifts" at the exact same distance after a certain height.
[www.dropbox.com]
[www.dropbox.com]
[www.dropbox.com]

this didn't happen with the bearingless holder on top, is this caused by what you are talking about the smooth rails/rods?

@ the digital dentist, so I'll be following your instructions and removing two of the smooth rods/rails. Based on your observations and experience. should I remove the front two smooth rods (the ones closest to the side lead screws as you saw in the pictures) or the back two smooth rods (the ones closest to the back lead screw)?

I'd put one guide rail as close as possible to each lead screw.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

crap I don't think it's quite possible for the back one, as it would hit the hotend part cooling fan when it print towards the edges >.<. I was wanting to do that (putting one beside each lead screw as you said) until I saw that wasn't really possible (it could be but possibly very hard). If that's not possible and I could only use two smooth rods, which ones should I remove?

Okay so I managed to design myself a smooth rod holder at the back, unfortunately due to clearance issues I could only fit in a LM12UU (30mm one) instead of a LM12LUU (57-58mm), However, you are right, after I put in the back smooth rod and removed the two back ones, then put the bed all the way down, , the bed straightened out and holy the nut was off from the jaw couplers, so basically before despite having 4 smooth rods, the bed like you said was tilted like front is almost 8mm higher than the back, and if I try to make them same level the motors would bind. I couldn't figure out why until this, and it makes sense, it was tilted so the lower it went the more vibrations and cause the lead screws to very likely bind and "walk" like you said. Here are the pictures of the new setup and the new print.

[www.dropbox.com]
[www.dropbox.com]
[www.dropbox.com]
[www.dropbox.com]
[www.dropbox.com]
[www.dropbox.com]
[www.dropbox.com]
[www.dropbox.com]
[www.dropbox.com]

Ignore the "wobbling" on the top, that's caused by the thing still swaying because it's just so high, so basically if I was to print something, it can't be smaller than 100mm (maybe 80mm) x 100mm) or else as you can see from the top the item wobbles and cause those "wobbles" that makes it looks like z wobble. HOwever i know it's not because it's printing out 701mm height when the thing was meant to be 700mm, unlike before with the 4 smooth rods, 700mm but it did 630mm only.

This also makes sense cause if you look at the sides, some of the sides at the top doesn't have "z wobble", etc.

The pictures also show the new setup, I've moved the lead screw closer to the front smooth rods as you've mentioned, the_digital_dentist now they are only 80mm away from each other, can't put it any more closer unfortunately. The top bearingless holders also seemed to help.

Are there any more suggestions though? Thank you very much!

The long extensions from the nut and the bearing to the bed support at the back look "iffy". You have the nut coupled to the bed support via a long printed plastic arm, and the bearing coupled to the bed support by a different arm. I'd probably build the bed support larger so that the bearing and nut can be more closely coupled to it and each other.

It looks like you still have a lot of z wobble- the uniform pattern of banding every couple mm as you go up the print. You also have a lot of ringing, which is more a function of acceleration, jerk, and resonance.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Hi Dentist, undfortunately I am really tight on budget right now, or else I would immediately buy a longer 2020 extrusion and make the back closer to the back motor, unfortunately I can't, so is another possible option to add another smooth rod to the right of the back side motor, or would that again cause more problems as you said,

as for the z wobble, I printed a much bigger object, the bigger vase, this is the spiral vase, and here's a very interesting thing, the banding isn't there, because this item barely sways at the top. but here are some pictures.

You'll notice that some the z wobble (it's also exaggerated by the light, I have to be within 10 cm to see it barely) but it's more noticeable for the back side (closest to the back motor) and the front and two sides barely have or have no "wobble" at all. SO it could be as you said due to the long iffy arms, but I can't buy another extrusion until maybe next month so hence the question above if I should add another smooth rod to the right side of the BACK motor (as you saw in the pictures there's one on the left side of the motor).

[www.dropbox.com]
[www.dropbox.com]
[www.dropbox.com]
[www.dropbox.com]

Oh and I forgot to mention, this z wobble only starts to happen around after the 650mm height mark, no z wobble before that, so I'm wondering what causes that, could it be the bearingless lead screw holder? All three have them, but it looks like the back one as the bed gets lower starts to push into the block a little, could that be causing it?

Just a thought, should I maybe instead remove the back motors, and instead go for a 4 motor setup? so put the two smooth rods at the back back again, but put two motors there for a 4 z motor setup? Or is the triangle setup better?

here's what I mean by 4 motor setup: [www.facebook.com]

Edited 1 time(s). Last edit at 02/03/2019 10:30PM by iamthebest22.

You could even use a spoon to cut your meal.
But the results will be not what you expect.

4 Points on a Printbed was and is always a bad idea. - Never more than 3 Points.
More than one motor for Z is the next bad idea. - They'll never stay in sync.
So please do yourself somethong good - start reading the 2 blogs mentioned in the dentists signature, read, understand, and change your System to something approved working.

And please: Change your Nickname - it hurts to read your silly questions with this Nickname - Sorry

Quote
Dancer
You could even use a spoon to cut your meal.
But the results will be not what you expect.

4 Points on a Printbed was and is always a bad idea. - Never more than 3 Points.
More than one motor for Z is the next bad idea. - They'll never stay in sync.
So please do yourself somethong good - start reading the 2 blogs mentioned in the dentists signature, read, understand, and change your System to something approved working.

And please: Change your Nickname - it hurts to read your silly questions with this Nickname - Sorry

Indeed, indeed but they never learn, never read, always come back with these bad ideas !
Oh yeah, sometimes it works, it happens when a bad idea is compensated by an other bad one. For ex. 4 rails with a flimsy frame will not seize, just may be grind.

Edited 1 time(s). Last edit at 02/06/2019 11:32AM by MKSA.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

Sorry but not changing it, it has nothing to do with this subject , just a childhood memory .

Thanks for the constructive criticism, as I am indeed a self learning guy, and I did do research after that post was made and realized that 4 is indeed a bad idea, hence why I posted in the other one topic and left this one alone, because this one needs to be closed, just never bothered asking. However sometimes using a spoon is necessary if you don't have a knife . Also, I am not the designer, I am merely following working with other people who said to do this, I'm just a poor guinea pig

MKSA, thanks for the constructive criticism, I see you're a fan of thunderf00t as well so don't worry, the idea isn't mine, but I do learn from it, and that's just the process when learning to work with other who design it, and I just test it, please tell the designer of the hypercube evolution that, not me

I'll try to change the motors to only one as you said, as I've seen many people do that, the question is this, the two side ones are from the same manufacturer, and are connected in series, is it possible for me to just disconnect the back one and have the two side ones in series pull that back motor, or no it has to be absolutely one only? Thank you very much

@iamthebest22
I and others already mentioned these errors here and there. Usual answer is others do the same and it works !
People fail to understand that often it works because there is enough play and flexibility in their set up that prevent binding and they put powerful enough motors.
Then when problems arise, they do like you, ask why ? Act of God or what else ???

There are at least 5 "designs" that popped up in the last 2 months, same issues etc. Digital Dentist is very patient as he keeps giving explanation, links ... Yet many are stubborn and keep going with their "design". Note most disappear.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

Okay and I can understand that, however, here's my story, and you should read this before you start to get frustrated.
Like many others, I relied ALOT and ALOT on facebook groups (I'm literally in 6 Hypercube groups) which has a huge groups (over 30k people in total) to find answers and ask questions. However, as I've started to learn more and more about 3d printer design only just recently (after being a guinea pig for so long) that I started to question some of the answers. They never thought of this forum actually and just follow along something because lots of people are talking about it. In fact, if you join the Hypercube Evolution main group, you'll see that just recently there's a huge following of people who are using 4 motors ALOT for their 500 x 500 or bigger beds, and alot of them think it's a great idea because well, like you said, they don't know. I strongly suggest you should join that and really comment on that. ANyways, to continue on my short excerpt.

So basically FB is fast and so as I've mentioned, being a beginner like many others, I've just blindly followed answers (and costing me a few hundred bucks in the way), until recently when I've started learning more and especially thanks to Thunderf00t actually, I've begun to question some of the answers I've gotten and especially the person who helped me design this (despite the errors) as we went along. It's not that I didn't question it before, but it was more like hey I'm a beginner and so I need to just listen and try everything, and now that I'm more experienced (even more now thanks to you guys so appreciate that) I've realized that facebook can have quite a big skill gap between people. For example, before I came to ask here, some of the answers I've gotten from this same problem was this:

- Bed heating causing problems, try turning off bed
- First layer adhesion issues
- Your lead screw is maybe bent

Not one person thought it was as you guys mentioned using three motors

I'm not making fun of these people either, because like me, they are new to this and just want to help (despite not being knowledgeable)



I can understand you're frustrated, and I understand, being a band/music AND English/ELL/ESL teacher myself, I have to re teach the same thing over and over again every year, and it can get frustrating after all, but here's the thing, despite being new to thunderf00t (only subscribed to him what about 3-4 months ago, interestingly enough when I've started this build with another person designing it. He says that quote(the quote you have in your signature), but after watching alot of vids, his main problem, at least my take away, isn't that people are making mistakes, it's that they are making mistakes and NOT learning from them.

Before you say I'm just like those 5 designs, well here's the thing, I've never seen those, and as I've mentioned I'm not the designer. The person who help me took his idea from the upcoming THE BEAST V2 3d printer, which apparently has 3 separate motors, and so he (and at the time I thought it was a brilliant idea , not so much now lol) thought by doing that and using a closed loop belt, it would be an ingenious idea. So yeah...

We're in the HEVO pro modified FB group and when we asked for suggestions, people are just like wow so cool, and well obviously that doesn't help with the situation, in fact for the modified, he just made it 4 motors >.<. So what I'll do is, I'll tell him about what I've learned and I strongly suggest experts like you to do this is as soon as possible. Some how get more people who are really popular (like CNC Kitchen, Thomas Sandelerer [I can never spell his name correctly], 3D printing NERD, 3D Maker Noob) to talk about this or something, because like I've said before, most people on facebook never bother or really even seem to know about this forum, they use FB because they can get super fast responses usually in like a few minutes, not realizing sometimes patience (like I'm doing right now) pays off, they just want fast fast fast "results"

I've personally only found out about this forum after being a Duet wifi user and finding out about this (because dc42 is on here too and Duet Wifi guys)

As you'll see with my conversation with the digital dentist, I'm not gonna disappear, I know why those people probably disappear, but I like to learn from mistakes and it's how I did, good thing it's usually small mistakes

I'm friends with the admin of the biggest HEVO group, and if you message him, he will let you or the other experts put a sticky post right on top telling people why not to use those designs, and that will hopefully greatly decrease any future posts like mines popping here in the future. You guys need to get this out there, blog posts etc just aren't as used often in people who like me back then, just almost heavily rely on FB for answers.

Alright so back onto this,
so I guess one motor only, so I guess I'll use the back motor as the single one, and use that to pull the front two side motors, will report back. Btw one reason he and I thought (well he thought, I just agreed at the time ) of this design is because we were worried that one motor cannot handle the weight of the whole thing. Guess we're wrong too xD.

This site may help you determine of one motor produces sufficient torque to lift your printer's bed: [daycounter.com]

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

wow thanks you very much! Wow looks like my one motor is definitely enough! Please tell me if I did it right though:

So for Force I put 13000 g (that's max area my thing can handle/print out, I fully know that it will never get that close, but better to be safe then sorry)
then for pitch diameter I put 2mm because if I interpret it correctly (and if I didn't please tell me) I have a T12 x 2mm lead screw 2 mm pitch, single start, 12mm diameter

Thread density is what I"m confused, but I put in 5 per cm, cause I"m assuming it's a 2mm pitch, so 2 x 5 = 10 mm which is 1 cm
Coefficient friction since I"m using steel lead screw + Machine oil lubricant, I'm assuming 0.17 as the worse case scenario.

Is that correct? If so holy I could go up to 120kg before my one motor would fail. BUt I am running at 1/4 microstepping, which means that's around 38% max of what it could handle, so that's still more than enough, if I did the inputs correctly.