Beer Can mod for Prusa MK3/S/+
Wrote up a step-by-step guide for bed leveling using shims cut out of a beer can:
https://arcady.genkin.ca/2021/01/beer-can-mod-for-prusa-mk3-s/
RE: Beer Can mod for Prusa MK3/S/+
I really love the idea, but I don't like that you have placed these sharp pieces of metal below the heated bed.
I think it is safer to place them between the space and the Y carriage.
All in all I will give this a try soon. I didn't like the nylock fix. It is just cumbersome.
Thank you 👍
RE: Beer Can mod for Prusa MK3/S/+
I will probably be doing this once my printer comes in. Looks to be a nice and simple mod. To avoid thermal expansion issues though I probably will put the shims underneath the spacers. Aluminum has a somewhat high rate of thermal expansion, so putting it further away from the bed is probably better.
RE: Beer Can mod for Prusa MK3/S/+
@pathia
@dragon1291
Placing the shims under the spacer will work. I considered this myself, and thought about the potential issues with damage to the print bed and thermal expansion.
For thermal expansion - I am not entirely sure what makes more sense. The spacer is also made of aluminum, but it's a thicker piece of aluminum, which can expand even more than the thin strips used as shims. The shims are not perfectly flat, and are very pliable, so, I figured, that it is possible that they can "absorb" some of the thermal expansion by deforming and filling the air gaps, rather than applying the upward force on the print bed. Also, potentially, they can even act as a bit of a heatbreak between the bed and the spacers, because they are not perfectly meshed together. Not sure if any of this makes sense, but I am not convinced that placing the shims above the spacer creates a thermal expansion issue.
As far as sharp objects next to the print bed, I am also not concerned, because the underside of the heat bed is not very important.
This said, placing the shims under the spacers, instead of above, will definitely work just as well. It will be a little bit harder to do it that way, because you'll be holding the shims with one hand, and will have to lift the spacer with the other, probably using another pair of tweezers or needle nose pliers. Placing the shims above the spacer is a lot easier, because the bed can be lifted very easily by pulling up on the bed's edge.
RE: Beer Can mod for Prusa MK3/S/+
@antipode
My thought though is that at least all the spacers will expand at the same rate wherein where if the shims are acting as a heat break they might affect the overall expansion. Might be an extremely minor thing tbf.
RE: Beer Can mod for Prusa MK3/S/+
@dragon1291
Yeah, I don't want to make any certain claims one way or another, only guesses. I would bet that if there is any difference, it's very small, but I may be proven wrong.
RE: Beer Can mod for Prusa MK3/S/+
Wrote up a step-by-step guide for bed leveling using shims cut out of a beer can:
https://arcady.genkin.ca/2021/01/beer-can-mod-for-prusa-mk3-s/
Did you ever consider using this method instead?
It's more adjustable and, as demonstrated, can achieve an even tighter variance.
RE: Beer Can mod for Prusa MK3/S/+
@dimprov
Not gonna try the Ender-3 technique of using knobs and screws. The auto bed level is the reason I keep ordering Prusas.
--------------------
Chuck H
3D Printer Review Blog
RE: Beer Can mod for Prusa MK3/S/+
@dimprov
Hi, there is that, and there is also the silicone spacers method (same idea).
The advantage is that these methods allow raising *and* lowering an adjustment point. With shims you can't lower a point after you removed all the shims.
The main disadvantages of springs, IMO:
- I don't think that the springs or silicone spacers can be as stable as a torqued down screw with solid shims. Your adjustment will drift, and you'd have to readjust from time to time.
- You need to buy stuff (springs or silicone spacers), whereas a beer/soda can can be had for free.
- You need to take the build table apart to install or to uninstall the mod with springs.
I strongly believe that there is absolutely no reason to adjust the bed beyond 0.1mm variance. The table and the build plate are going to buckle from heat every time, in a different way depending on the temperature, and depending on the build plate, and depending on what side of the build plate you place up. So, any super-fine adjustment will be negated by this buckling. I think that 0.1mm variance across the whole bed is good enough for 3D printing.
Of course, you can achieve higher accuracy with shims, too, if you wish, you just need to get a set of shim stock of various thickness. But I'm definitely not doing that myself.
RE: Beer Can mod for Prusa MK3/S/+
@cwbullet
The prusa MK3 still holds the advantage: it's equipped so that you can easily upgrade to more than 4 springs. 😉
RE: Beer Can mod for Prusa MK3/S/+
- I don't think that the springs or silicone spacers can be as stable as a torqued down screw with solid shims. Your adjustment will drift, and you'd have to readjust from time to time.
I'll concede the point, but let us consider more than just the initial setup: in both our cases if we were to change the build plate and/or the temperature, we may face the need to make adjustments. In those instances, which method is easier to adjust? Wheel adjustment or add/removing shims? Wheel adjustment, right?
RE: Beer Can mod for Prusa MK3/S/+
@dimprov
Interesting, I will watch. I do not see the need.
--------------------
Chuck H
3D Printer Review Blog
RE: Beer Can mod for Prusa MK3/S/+
If you are planning to readjust often, then, yes, wheels are at your fingertips. Does it make sense, though, to readjust for various temps and build plates?
There are two issues:
- Print surface being not flat or not parallel to the printer's idea of the XY plane, due to manufacturing and assembly imperfections/tolerances. The variance here can be quite large, on my printer it was 0.5 mm, but I've seen the variance of more than 1 mm.
- Print surface distorting differently for different combinations of temperature settings and build plates chosen. In my observation, this distortion is usually within 0.2 mm of variance across the bed.
My position, is that I'm prepared to deal with (1), but I am going to ignore (2) as unavoidable and insignificant. I am quite willing to spend 30 minutes to fiddle with the print bed once and get it to some reasonable degree of flatness and parallelism to the XY plane. But I am not going to tram the bed on a regular basis, because I don't think that the benefits are worth the trouble. There is very little gain, and a lot of pain involved; diminishing returns, and all that.
RE: Beer Can mod for Prusa MK3/S/+
After spending a year with my Artillery Sidewinder, the last thing I want to do is introduce springs and any sort of manual leveling to my rock-steady Prusa.
- For one thing, corner springs are properly used to tram the bed to ensure perpendicularity between the axis. From all I've read, using 4 corners to level a plane is counterproductive, yet seems to be the method every manufacturer of cheap printers uses. 3 is ideal. More than 4 would be a nightmare.
- Regardless of how many shims or springs I introduce, there are still going to be inconsistencies that automatic MBL will correct for on every print with a much higher density than anything mechanical I can add. Like @antipode, I'm happy to do any procedure once, but not regularly. That's why I purchased a Prusa with automatic MBL.
In order to get anything near the 1st layer consistency of my Mk3 on my Sidewinder, I've had to:
- Adjust the 4 corner springs, usually any time I load filament or sneeze near the hotend. This gets my bed mostly perpendicular to the nozzle. I keep this intentionally high to avoid ramming the nozzle into the bed. I've tried stiffer springs, but it always drifts.
- Compile my own firmware, replace the dopey touch screen with a proper LCD with a knob, and use manual mesh bed leveling with a 5x5 grid. This takes 5-10 minutes every time I (frequently) have to relevel. This gives me consistency between all of the 25 test points. I'd use 7x7 but I go insane doing that many adjustments each time.
- Print a 75x75mm "Life Adjust" square to approximate Prusa's automated procedure. The previous steps adjusts the nozzle to a consistent height across the bed. This step ensures I get that good squish and a consistent first layer, essentially mimicking Live-Z the hard way.
- Print a 5x5 test grid pattern and make individual tweaks to the level settings at each test point as needed.
This gives me a level bed that is almost as good as my Prusa's in 15-30 minutes before any print. Not much of a bonus at all. My opinion -- formed after an admittedly limited sample size of 1 -- is that manual mesh bed leveling of any sort is tedious and not particularly accurate. In this case, the grass is definitely not any greener.
The only thing I'd like on my Prusa is a way of making minor adjustments during printing. I don't know if M421 can be used at print time, but if it were possible to select a test point while printing and make minor +- adjustments to Live-Z for that point on the fly, it would be very useful and make a lot more sense when using removable print surfaces that are regularly flexed and moved. I may try coding an OctoPrint plugin for this since it would be useful for all printers.
and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan
RE: Beer Can mod for Prusa MK3/S/+
After spending a year with my Artillery Sidewinder, the last thing I want to do is introduce springs and any sort of manual leveling to my rock-steady Prusa.
- For one thing, corner springs are properly used to tram the bed to ensure perpendicularity between the axis. From all I've read, using 4 corners to level a plane is counterproductive, yet seems to be the method every manufacturer of cheap printers uses. 3 is ideal. More than 4 would be a nightmare.
Even if the build plate starts flat, there's going to be some warping when it heats up. Using a dial indicator in one of his earlier videos, Tom Sanlanderer proved that the magnitude of the warping can be quite significant.
- Regardless of how many shims or springs I introduce, there are still going to be inconsistencies that automatic MBL will correct for on every print with a much higher density than anything mechanical I can add. Like @antipode, I'm happy to do any procedure once, but not regularly. That's why I purchased a Prusa with automatic MBL.
In the OP the distinction is made between using an ABL vs making the bed flat. The OP's use case, as stated in the OP, requires the difference between min-z-height and max z-height to be <= 0.1mm.
--------------------
Since thinner build plates allegedly warp more, I'd be curious as to whether using a much thicker heated bed (say, 8mm?) might serve to might mitigate against warping.
RE: Beer Can mod for Prusa MK3/S/+
@bobstro
On a Prusa, physical bed leveling is not about first layer adhesion. Mesh bed leveling works pretty well without any kind of physical bed tramming. You'll have a successful print without any of this fuss.
Bed tramming we are discussing is needed to ensure that printed parts have a flat bottom, and square walls.
RE: Beer Can mod for Prusa MK3/S/+
I used thin washers made of steel feelers. A friend machined them for me. I guess it should have worked. I have up and could never get them adjusted right. That was on my ender. Maybe the results would be better on an MK3S.
I be frank, I have had nera perfect results with my Mk3S abd Minis. I am fearful of messing with them. The enemy of good is better.
--------------------
Chuck H
3D Printer Review Blog
RE: Beer Can mod for Prusa MK3/S/+
@cwbullet
I envy you, it must be nice to have a machinist friend! 🙂
Steel washers may be a challenge to get past the bed magnets on a Prusa. 🙂
There is no chance of messing up the print quality by making your bed flatter, but I agree that there is no need to chase absolute perfection.
RE: Beer Can mod for Prusa MK3/S/+
What's the repeatability tolerance on the Prusa Pinda probe? If it's, say, 0.1mm or worse, then were already at the limit. The BLTouch version 3.1 supposedly is 0.1mm. On the other hand, my Clockwise Tools digital dial indicator has better repeatability than that.
RE: Beer Can mod for Prusa MK3/S/+
@dimprov
I remember reading somewhere that it's a lot more accurate than that. Don't want to quote any numbers from memory. Good point, though, PINDA accuracy is definitely an important part of the process.