3 points levelling system

If the "engineer" has a degree it certainly isn't in mechanical engineering Using four points will over constrain the bed which is never a good thing. If you cannot get the plane of the bed parallel to the X and Y planes with a three point system then you either have a warped bed or don't understand how to adjust each point.

Have you determined if your X and Y planes are parallel?

Quote
Tinchus
Ok... I didnt abandoned this issue. I converted my prusa i3 to a 3 point level system as I already stated. I experimented with it for a while. And then I found this:

[youtu.be]


In this video it is perfectly explained why a 4 point levelling system should be used. I got an almost same explanation from prusa itself. I aske them why they use a 4 point levelling method instead of a 3 points and they game me the same explanation. Also, I did some research, looking for people with 3 points systems... they suffeer the same issue, there will be always a corner wich seems to be always not leveled.

Thing is: in real world, it is alm ost imposible to get a perfectly flat surface. Even if we get one, that surface wont remain flat for much long, since for example, a printed piece, really hard atached to it, when we try to remove it there is a chance is slightly warping it while removeing the pice for example. Also thermal expansion has ore room in a 3 point system than in a 4 point system.

Looking at the video, it is explained that a somehow flat surface can be "leveled" in the corners, but it will be a little warped in the center, down or up. And that is perfectly normal. That has always been my problem and I though it was a bad thing. So also, modifying first layer settings is the way to go.
I have gotten a really flat aluminium bed now and returned to a 4 points system. and with ew first layer settings, I can now say that after 2 years printing... Im seeng the light lol
Thanks all for the help, look the video, really usefull information there from an engineer.

With an additional adj screws in the middle, you could adjust to get rid of this "warped center". So 5 points is even better than 4. In fact you could have many more . You got to be patient to adjust all this though.
This because your bed is flexible. In fact everything is in these cheap machines is flexible and luckily because considering how they are designed and built they would bind !

Even engineers may rely on duct tape and steel wires to fix a flaky machine but that doesn't mean it is the correct way to do !

---

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

Quote
etfrench
If the "engineer" has a degree it certainly isn't in mechanical engineering Using four points will over constrain the bed which is never a good thing. If you cannot get the plane of the bed parallel to the X and Y planes with a three point system then you either have a warped bed or don't understand how to adjust each point.

Have you determined if your X and Y planes are parallel?

An engineer works far more by finding the best compromise than by strict adherance to rules. For example, an aircraft needs very rigid wings for best control but flexible wings are better able to take sudden onset forces such as are found in turbulance or violent manuoveres. Prusa has found a compromise that works and their printers are among the best of the FFF type - finding that good compromise is good engineering.

For myself, I do prefer the three point system but I put the adjusters somewhat in from the very edges to give the best support and allow a little bit of mass on the table to be saved.

Mike

The i3 starts with a compromised bed and compromises the support and leveling structures under it, then uses autoleveling (autotramming) to compensate, and they seem to have it working, but when the autoleveling fails you won't be able to print. You have traded time spent leveling and releveling the bed for time spent debugging the autoleveling system. Some may see that as an improvement, and it may indeed be an improvement over using such a bed and leveling system without autoleveling. For users who don't know or understand how a printer works or how to fix things when they don't, the machine becomes a doorstop when the autoleveling fails. But that's not anyone posting in this thread.

If you're going to the trouble of designing and building your own printer, what is the point in copying all the compromises that are made to get a commercial product out the door for a relatively low (?) price? For a lot of people, building something like a 3D printer is done so they can build something better than they can afford to buy off-the-shelf. It is neither difficult nor very expensive to build it right, and there are no secrets. You can easily get a flat, rigid, thermally conductive bed, three point leveling works, and no autoleveling is needed.

It comes down to philosophy, budget, and how you prefer to spend your time. If you're a zip tie and duct tape sort of "engineer", copy the commercial product and tinker with it to get it to perform its basic functions. Some people buy junky cars and work on them because they like doing that sort of thing. That's OK if that's what you like. Some might prefer to see how cheaply they can build something that still functions as a printer. But if you prefer something that doesn't require endless tinkering to find the magic combo that keeps it running, spend a few more $ and build it better, and you can spend more time designing and printing and less time tinkering.

Of course, designing and building your own machine takes a lot of time, too. I tend to keep going back to my designs and tweak them or sometimes make major changes to the printer's construction. For example, SoM was built to solve problems I discovered in MegaMax's design after using it for a while. SoM lives at the Milwaukee Makerspace and is the main workhorse printer there because it is the most reliable one. People keep discovering new ways to break it that I didn't anticipate in the design, so I have to fix it every once in a while, and do my best to fix it so it can't be broken the same way again. I'll soon be going back to SoM and replacing the bed with the kinematic mount that I used in UMMD. UMMD, while printing reliably, is still a work in progress. Right now I'm testing a low cost replacement for the expensive worm gear drive in the Z axis. If you like to tinker, designing and building your own machine without all the commercial compromises still leaves plenty of room to tinker and tweak. You just won't have to tinker and tweak to get the machine to perform its basic function, so you can stop when you get tired of tinkering.

---

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

@digital_dentist, I can see where you are coming from, but I don't entirely agree with some of what you say in the above post. I suspect you have never had a 3D printer with a good Z probe, so you don't appreciate the advantages of having one even though your printers don't need auto bed compensation.

Running a 3D printer without a Z probe as you do is possible only if all of the following are true (unless you are prepared to spend a lot of time doing manual levelling frequently):

1. The bed is sufficiently flat (e.g. its height varies by no more than +/-50um). This is possible if you use cast aluminium tool plate.

2. The X gantry does not sag significantly. This is possible if you use sufficiently thick rods or linear rails.

3. The Y motion system does not sag significantly and has no twist. Again, this requires stable mechanics.

4. The bed support is sufficiently stable that the levelling does not drift significantly.

All of the above are possible to achieve, but are not cheap and they make the printer heavy. For low-cost or lightweight printers, it is more economical to use a lighter weight construction, and to use a Z probe to provide auto bed compensation. That's not an excuse for using a bad design of course, but IMO it is an acceptable trade-off in a low-cost printer to use cheaper/lighter materials, as long as the amount of bed compensation required is small.

Even if you have a printer that is stable enough not to need bed compensation, a Z probe has a number of other uses:

1. Getting a perfect Z=0 height just before starting the print. Depending on the design of your printer, the Z=0 height after homing will depend to a greater or lesser extent on temperature. So if you print with different materials, bed temperatures or chamber temperatures (or even different ambient temperatures), the Z height after homing may vary. Probably not an issue if you use a low-end homing switch and always print at the same temperatures.

2. Adjusting the levelling screws. I use the Z probe on my Cartesian and SCARA printers to tell me how much I need to adjust each screw by to get the bed level.

3. Checking that the bed is flat and level, for example after moving the printer. I use the Z probe occasionally to generate a height map, even though I rarely enable mesh bed compensation. You can do the same thing manually using a dial gauge, but it takes a lot longer. Here are two examples of height maps generated by probing, one from a Cartesian printer built from a budget kit (which nevertheless I can get good prints out of) and one from a delta printer that I designed myself.



As for the Z probe failing, good Z probes are easy to get working and very reliable. Some types such as capacitive sensors are best avoided.

Edited 5 time(s). Last edit at 04/01/2018 01:15PM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].