Threaded Rod vs Pitch

Quote
Dark Alchemist
One thing the Dentist mentioned above is dead straight they will lose sync but according to Prusa just slam Z all the way up until you hear the motor(s) doing the ratatatatat sound for a few seconds as it will not hurt the motors and both motors will be back in sync. This is his way he tells everyone to do it on his MK2 series of printers.

Alternatively, if the Z leadscrews are driven by independent stepper drivers and you have a Z probe, you can probe the bed and use firmware to adjust the leadscrews to level the bed. RepRapFirmware had this functionality, and I've heard that Repetier has it too.

---

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].

Quote
dc42

Quote
Dark Alchemist
One thing the Dentist mentioned above is dead straight they will lose sync but according to Prusa just slam Z all the way up until you hear the motor(s) doing the ratatatatat sound for a few seconds as it will not hurt the motors and both motors will be back in sync. This is his way he tells everyone to do it on his MK2 series of printers.

Alternatively, if the Z leadscrews are driven by independent stepper drivers and you have a Z probe, you can probe the bed and use firmware to adjust the leadscrews to level the bed. RepRapFirmware had this functionality, and I've heard that Repetier has it too.

That is the whole point is that on a Prusa 2 motor design (I am sure all of them are like this) that they share the same driver. In a real world situation it would use separate drivers and have some precise way to drive both to precision but the costs would rise substantially. Having just an ordinary endstop on each Z side (in a dual controller setup) would not suffice as I found they bounce and a mechanical switch is only good to the hundredths digit (some to the tenths). Now I loved when I changed over from a mechanical endstop to my home made magnet as it is so precise and that might work in a two motor setup but with auto bed sensing (ABL) that wouldn't work.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

There is a simpler, cheaper way, just ONE stepper, one belt, one pulley per leadscrew. No coupling that introduces wobbling. This two steppers config is a DESIGN FLAW. Yet, people keep using it, poorly justifying it, keep reinventing unsatisfactory if not blatanly wrong ways to circumvent it etc.... Worse, others than the Prusa are afflicted by this, some even worsening it by using THREE motors plus leadscrew !

Quote
MKSA
There is a simpler, cheaper way, just ONE stepper, one belt, one pulley per leadscrew. No coupling that introduces wobbling. This two steppers config is a DESIGN FLAW. Yet, people keep using it, poorly justifying it, keep reinventing unsatisfactory if not blatanly wrong ways to circumvent it etc.... Worse, others than the Prusa are afflicted by this, some even worsening it by using THREE motors plus leadscrew !

Three motors for Z? OUCH. Which one does that?

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

Three leadscrews driven independently plus a Z probe allows you to perform automatic bed levelling (and I really do mean bed levelling, not bed compensation). Three leadscrews driven neither by independently-driven motors nor by a single motor and belt drive gives twice as much to get out of sync as two leadscrews does.

Edited 2 time(s). Last edit at 09/07/2017 11:02AM 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].

Indeed, like many, I too had thought about this kind of leveling. (Done in fact in optic and antenna ). Fact is, considering the sizes we are dealing with, on a cartesian machine, I prefer to make the machine bed flat and rigid enough. Even with my Prusa, I managed to get good results and no autobed level at all. I don't have to bother about the type of material used, sensor quality, firmware etc...

Quote
Dark Alchemist
Three motors for Z? OUCH. Which one does that?

On Thingiverse, reinvented regularly for CoreXY, H bot ... machines. You can't tell it is wrong, they take it personally

Quote
dc42
Three leadscrews driven independently plus a Z probe allows you to perform automatic bed levelling (and I really do mean bed levelling, not bed compensation). Three leadscrews driven neither by independently-driven motors nor by a single motor and belt drive gives twice as much to get out of sync as two leadscrews does.

The thing is we are talking about the setup all of these systems use which is two motors one cup, errr driver. In a perfect world 3 motors and 3 separate controllers would rock and precisely for the reason you mention.

Quote
MKSA

Quote
Dark Alchemist
Three motors for Z? OUCH. Which one does that?

On Thingiverse, reinvented regularly for CoreXY, H bot ... machines. You can't tell it is wrong, they take it personally

If it was done with separate motor controls then it would be wonderful as it could do an actual bed level instead of compensation, as DC42 mentions, but that is not what we are talking about here as they all share the same controller which is beyond retarded.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

I have noticed that the 3D printing community as a whole exhibits a propensity for overly complicated solutions to 3D printing problems. Why use solid construction and a flat bed plate when you can throw a bunch of motors/drivers, screws, sensors, power supply, and software at the problem? There's no cost benefit analysis done to figure out whether it might actually be cheaper to build the thing solidly instead of using a bunch of active measures to fix the problems that cheesy construction creates. People would rather spend money on fixes for problems than on preventive measures. No one seems to care about the reliability of the proposed solutions, either. If it looks good in a youtube video, it must be good!

---

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

Quote
the_digital_dentist
I have noticed that the 3D printing community as a whole exhibits a propensity for overly complicated solutions to 3D printing problems. Why use solid construction and a flat bed plate when you can throw a bunch of motors/drivers, screws, sensors, power supply, and software at the problem? There's no cost benefit analysis done to figure out whether it might actually be cheaper to build the thing solidly instead of using a bunch of active measures to fix the problems that cheesy construction creates. People would rather spend money on fixes for problems than on preventive measures. No one seems to care about the reliability of the proposed solutions, either. If it looks good in a youtube video, it must be good!

We all know where you are coming from after all of these years. Hell, make the damn thing 1,000 kilogram that requires a special concrete slab that was engineered dead flat and none of this would be needed.

Edited 1 time(s). Last edit at 09/07/2017 02:56PM by Dark Alchemist.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

Digital dentist, you appear to me to be confusing using mesh bed compensation (often incorrectly called mesh bed leveling) to compensate for beds that are not flat and gantries that sag, with using multiple independently-driven leadscrews to get a flat bed level - as an alternative to using manual levelling screws and a belt drive to keep the leadscrews synchronised.

If I were building a printer with a bed moving in the Z axis, I would probably view independently-driven leadscrews as a simpler solution than a belt drive, because I am more comfortable with electronic engineering than mechanical engineering. I guess the reverse is true for you. What I think we are both agreed on is that having multiple motors without independent control and a Z probe is the worst solution.

---

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].

Quote
dc42
Digital dentist, you appear to me to be confusing using mesh bed compensation (often incorrectly called mesh bed leveling) to compensate for beds that are not flat and gantries that sag, with using multiple independently-driven leadscrews to get a flat bed level - as an alternative to using manual levelling screws and a belt drive to keep the leadscrews synchronised.

If I were building a printer with a bed moving in the Z axis, I would probably view independently-driven leadscrews as a simpler solution than a belt drive, because I am more comfortable with electronic engineering than mechanical engineering. I guess the reverse is true for you. What I think we are both agreed on is that having multiple motors without independent control and a Z probe is the worst solution.

See, what he says is right but not practical as well due to costs, weight, etc... Sure, grab a Bridgeport that can weigh 500-1000 kilogram and it would make one hell of a 3d printer but would be slow as hell too but it would be dead flat. Having a tooling plate is all nice but I have seen tooling plate buckle, and warp, over time when applying 450-700 watts from heating elements too and that was with Mic 6 so far from cheap and even a Bridgeport has to be made true every six months to a year and, as I mentioned earlier, weight close to 500-1000 kilogram.

I personally would love to walk next to the printer and tell it to level the bed and three screws move then my bed is actually leveled (not close or approximate but really level) with a flip of a switch. It means that when maintenance time comes I never need worry about diddling with the bed. I don't care to hear anyone say they never had to recalibrate their bed because that person would be delusional at best or at least fooling themselves since there is not a piece of equipment in the world that doesn't go out of check every so often. I don't think you would find a single machinist that would say otherwise.

What is so wrong with a bed that actually levels itself? Dentist has always sounded like this old guy in the shop saying all of this CNC stuff is not needed and is just an overly complicated solution when I can get it done "on the mill" before you have even puched out the code. Trust me I have heard that a few times and when a machine can save me labor, and time, from having to do it myself you can bet your bippy I will choose that over anything else.

Edited 1 time(s). Last edit at 09/07/2017 04:27PM by Dark Alchemist.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

I can definitely see the value in mesh compensation in large machines in which keeping everything flat and level enough to print is impractical. In hobby-size machines I just don't feel like it makes sense when it isn't particularly difficult or expensive to build the machine so that the bed and X axis stay flat and level enough to print on without frequent adjustment. The sort of things you do in the construction to make the machine stable also contribute to print quality, and consistency. Solid construction is generally more reliable than active solutions to the bed leveling/flattening problem, and when the inevitable failure occurs, much faster and easier to understand and repair.

In a 3D printer, without the sort of mechanical loads a milling machine is subjected to, the need for adjustment can be minimized. The only time I have to adjust the leveling in Son of MegaMax is if I take apart the Y axis for modifications or maintenance, nether of which have been done in the last 6 months. The 1/4" tooling plate bed stays flat enough to print on when heated, and if it needs retramming because I have disassembled the Y axis, the three point leveling makes the job quick and easy. I don't have as much history on it yet, but Ultra MegaMax Dominator is on track to match SoM's reliability and minimized need for repairs and maintenance.

System reliability is a function of the reliability of each part of that system. Two parts that make up a system, each with 90% individual reliability,result in 81% system reliability. Now imagine a cascade of 3 motors and their cables and connectors, 3 drivers, a power supply, a sensor and its cable, and software running on the controller, each with less than 100% reliability assembled as a system (actually, that just about doubles the complexity of the whole printer, which requires 3 motors, drivers, cables, etc. to move the mechanism). How reliable can/will it be? How does that compare to the reliability of a rigid structure held together with bolts and driven by a single motor? 3D printing is an inherently unreliable process due to the complexity of the mechanism and the the vagaries of molten plastic of unregulated composition. I prefer to keep the mechanism simple because I believe doing so makes it more reliable. My experience with my own and commercially produced printers at the makerspace tells me that my approach has value. Maybe you have experienced something else.

---

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

Tooling plate that is 6mm thick and has 500 watts shoved at it (300 sq mm) will distort over time and all I ever hear is to just use the largest layer height you can for the first layer which, to my ears, is just a bandage to an issue. I really wish we didn't need a heated bed because then you could set it and forget about it if the bed is stationary. This is the one thing I love about a Delta or a Polar setup is the bed never needs to be trammed because once trammed, and if it doesn't use a heated bed, it stays trammed as long as you don't pick it up and move it around. Now it might need a tweak every year or two or not it just depends on weather conditions, etc... but heated beds make the tooling plate, or any thing, flex due to some parts being cool and other parts are hotter then cool it down and some parts cool down quicker. This is why a thinner metal plate will warp on you fast.

You are against glass on the bed but people use it because it is dead flat and stays dead flat unless someone bull dog clips it. Look at a router where you face the sacrificial board which is tramming the board but we can't do that on a 3d printer so we can only ever get close. Lets say we could face our beds and it is 0.00mm of deflection across the entire bed. Now add 500 watts to it and see if it stays 0.00mm deflection. It shouldn't and if it does then for how long? The answer I get back from the 3d community is "but it doesn't need that sort of accuracy just make the first layer .2 or .3mm layer height to compensate". UGH!!!

Anyway the three motor setup I am talking about is on a stationary bed that doesn't have a heating source because if either of those two change you can, and probably will, have issues.

Edited 1 time(s). Last edit at 09/08/2017 01:42AM by Dark Alchemist.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

Quote
Dark Alchemist

Quote
dc42
Three leadscrews driven independently plus a Z probe allows you to perform automatic bed levelling (and I really do mean bed levelling, not bed compensation). Three leadscrews driven neither by independently-driven motors nor by a single motor and belt drive gives twice as much to get out of sync as two leadscrews does.

The thing is we are talking about the setup all of these systems use which is two motors one cup, errr driver. In a perfect world 3 motors and 3 separate controllers would rock and precisely for the reason you mention.

Quote
MKSA

Quote
Dark Alchemist
Three motors for Z? OUCH. Which one does that?

On Thingiverse, reinvented regularly for CoreXY, H bot ... machines. You can't tell it is wrong, they take it personally

If it was done with separate motor controls then it would be wonderful as it could do an actual bed level instead of compensation, as DC42 mentions, but that is not what we are talking about here as they all share the same controller which is beyond retarded.

The people coming with this don't do it with that in mind !

How about making it and showing us ? Then, based on it, one can see the pro and cons. So, let's rock !

Quote
Dark Alchemist

Quote
the_digital_dentist
I have noticed that the 3D printing community as a whole exhibits a propensity for overly complicated solutions to 3D printing problems. Why use solid construction and a flat bed plate when you can throw a bunch of motors/drivers, screws, sensors, power supply, and software at the problem? There's no cost benefit analysis done to figure out whether it might actually be cheaper to build the thing solidly instead of using a bunch of active measures to fix the problems that cheesy construction creates. People would rather spend money on fixes for problems than on preventive measures. No one seems to care about the reliability of the proposed solutions, either. If it looks good in a youtube video, it must be good!

We all know where you are coming from after all of these years. Hell, make the damn thing 1,000 kilogram that requires a special concrete slab that was engineered dead flat and none of this would be needed.

Speaking about concrete, it is what I intended to do, use a slab of concrete as a base. So far, I just use a 40mm thick piece of kitchen table top with the printer bolted to it. Granite would be great.
In fact, the bed could be made out of granite too. The kind used in some cooking plate [www.steakstones.com]; If it doesn't work, you can still use it . BTW, I am of course serious about this, not for a Prusa of course !
Note: I have a similar plate, 25X25X1.6 cm, mass 2.7kg, flatness better than 0.05mm !

Edited 3 time(s). Last edit at 09/09/2017 02:00AM by MKSA.

Quote
the_digital_dentist
I can definitely see the value in mesh compensation in large machines in which keeping everything flat and level enough to print is impractical. In hobby-size machines I just don't feel like it makes sense when it isn't particularly difficult or expensive to build the machine so that the bed and X axis stay flat and level enough to print on without frequent adjustment. The sort of things you do in the construction to make the machine stable also contribute to print quality, and consistency. Solid construction is generally more reliable than active solutions to the bed leveling/flattening problem, and when the inevitable failure occurs, much faster and easier to understand and repair.

...

System reliability is a function of the reliability of each part of that system. Two parts that make up a system, each with 90% individual reliability,result in 81% system reliability. Now imagine a cascade of 3 motors and their cables and connectors, 3 drivers, a power supply, a sensor and its cable, and software running on the controller, each with less than 100% reliability assembled as a system (actually, that just about doubles the complexity of the whole printer, which requires 3 motors, drivers, cables, etc. to move the mechanism). How reliable can/will it be? How does that compare to the reliability of a rigid structure held together with bolts and driven by a single motor? 3D printing is an inherently unreliable process due to the complexity of the mechanism and the the vagaries of molten plastic of unregulated composition. I prefer to keep the mechanism simple because I believe doing so makes it more reliable. My experience with my own and commercially produced printers at the makerspace tells me that my approach has value. Maybe you have experienced something else.

The only times I have experienced issues with electrical/electronic parts when building 3D printers have been either because I was using an electronic part that I didn't design and wasn't up to the job, or because I was crimping wires to connector pins and the wires (which I didn't choose) were too fine for the crimp pins.

I have also experienced loose belts, loose grub screws on pulleys, and backlash in belt/pulley systems. So IMO mechanical components - for example, the belt and pulleys driving 3 leadscrews from one motor - can be at least as troublesome as electrical/electronic components. I keep my printers as simple mechanically as I can. That's why I like delta printers. I also like them to be accurate and stable so that they need little or no mechanical adjustment, and little or no electronic compensation for mechanical deficiencies.

For me, a Z probe isn't extra complexity, it's an essential component of any printer I build, to get an accurate Z=0 height just before starting a print, even if I don't want to use it for mesh bed compensation (which I hardly ever use) or anything else.

In a well-designed and built 3D printer using good components, I think the system reliability won't be significantly affected by either the electrical and electronic components or by the mechanics of the motion system, rather it will be dominated by extrusion issues.

---

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].

Quote
Dark Alchemist
Tooling plate that is 6mm thick and has 500 watts shoved at it (300 sq mm) will distort over time and all I ever hear is to just use the largest layer height you can for the first layer which, to my ears, is just a bandage to an issue.

What units of time are you thinking in terms of? I've had a 450W heater on SoM's 12.5" x 12" x 1/4" tooling plate bed for the last 3 years of daily printing, and as far as I can tell, it hasn't warped or distorted. Prints still stick reliably over the entire surface. I think if it did warp or distort, I'd just replace it another piece of tooling plate. If I get >3 years use out of a bed plate that costs <$15 (random rack at Howard Precision Metals), I can live with that. How many dropped, cracked, or chipped glass plates, and at what cost, will the average printer be replacing in that time?

The only place I've seen anything about thicker first layers is a tool-tip in slic3r. I believe that is aimed at people with the typical, sheet metal and PCB beds that bow when heated and are "leveled" with 4 screws. I always print with the same first layer thickness as the rest, typically 200 um, which works reliably from edge to edge. But you're right, I'd probably have some trouble printing a 100 um first layer over the entire bed surface. Thank goodness I've never needed to do that!

---

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

Quote
the_digital_dentist

Quote
Dark Alchemist
Tooling plate that is 6mm thick and has 500 watts shoved at it (300 sq mm) will distort over time and all I ever hear is to just use the largest layer height you can for the first layer which, to my ears, is just a bandage to an issue.

What units of time are you thinking in terms of? I've had a 450W heater on SoM's 12.5" x 12" x 1/4" tooling plate bed for the last 3 years of daily printing, and as far as I can tell, it hasn't warped or distorted. Prints still stick reliably over the entire surface. I think if it did warp or distort, I'd just replace it another piece of tooling plate. If I get >3 years use out of a bed plate that costs <$15 (random rack at Howard Precision Metals), I can live with that. How many dropped, cracked, or chipped glass plates, and at what cost, will the average printer be replacing in that time?

The only place I've seen anything about thicker first layers is a tool-tip in slic3r. I believe that is aimed at people with the typical, sheet metal and PCB beds that bow when heated and are "leveled" with 4 screws. I always print with the same first layer thickness as the rest, typically 200 um, which works reliably from edge to edge. But you're right, I'd probably have some trouble printing a 100 um first layer over the entire bed surface. Thank goodness I've never needed to do that!

I see you use reddit as turn_n_cough. Fact is I disagree because a bed costs me around 40 dollars on my 300x300 and it isn't even the real Mic6 it is from Alpha metals so costs about 1/3 to 1/2. 40 dollars every 2-3 years is not acceptable to me and is to you so we differ in what our pain threshold is at.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.

Let's see if I understand. You paid $40 for something that isn't cast tooling plate which is 1/3 to 1/2 of something (?) else, and it's warping because of the heat, therefore, my MIC6 tooling plate must be warping even though I've been using it for 3 years and prints are sticking like the first day it was on the machine. I'm having trouble following your logic.

Ebay has plenty of vendors selling MIC6 and other cast tooling plates. Here's the very first one that comes up on a search for MIC6 aluminum: [www.ebay.com] $25 for the plate + $15.32 shipping gets you a 12"x12" x 1/4" MIC6 plate for $40.32 all in. Yes, it's a bit steep, but not everyone lives close to Howard Precision Metals, and you only have to buy it once.

---

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

Quote
the_digital_dentist
Let's see if I understand. You paid $40 for something that isn't cast tooling plate which is 1/3 to 1/2 of something (?) else, and it's warping because of the heat, therefore, my MIC6 tooling plate must be warping even though I've been using it for 3 years and prints are sticking like the first day it was on the machine. I'm having trouble following your logic.

Ebay has plenty of vendors selling MIC6 and other cast tooling plates. Here's the very first one that comes up on a search for MIC6 aluminum: [www.ebay.com] $25 for the plate + $15.32 shipping gets you a 12"x12" x 1/4" MIC6 plate for $40.32 all in. Yes, it's a bit steep, but not everyone lives close to Howard Precision Metals, and you only have to buy it once.

40 dollars (includes shipping) and IS cast tooling plate just made by another metals company. Here it is [www.midweststeelsupply.com] ATP5.

---

_______
I await Skynet and my last vision will be of a RepRap self replicating the robots that is destroying the human race.