Have you Measured your Glass?

Hmmmm, here's something that has been driving around the twist the last couple of days.
I have been trying to print something with a large base area, that's just refused to print. I print on plain glass BTW, and don't normally have problems.
So I washed it, scrubbed it, and still I had problems. And to confuse matters I had to use "THAT" dodgy reel of filament because I wanted it that colour...(We all have one don't we)...
Nope it refused to print. I got close, but then one corner would lift.... Warping.... I use a Big Blue, no external fan, so how? I couldn't see a reason for it....
I'm at work, so turned the air con off to see if that helped. Nope.... I turned the print 90'... Well that nearly worked, the corner wasn't lifting now....
But still having BIG problems. I turned the glass over, that made it worse.... So I was close to throwing the whole machine over the side I was getting that nuts with it.
Then I thought about it some more and got the DTI mount out and rigged it up.. Yep it had a HUMP in the middle of the glass.
So I turned it back over, and ran the DTI over it again... Still got a hump, but not so bad... How come... The glass is flat isn't it I said to my self.
So I measured it..... 4mm both sides, and 4.2mm in the middle... So that's the problem... All that time I was thinking the glass was greasy, the filament.... THAT filament was doing it....
The temps weren't right.... Grrrrrrrrrrrrrrrrrrrrrrrrrr....
So now I've altered the design of the base, and put a wide 0.15mm trench where the hump is and just waiting for Slic3r to finish the code so I can test it.
So next time you have a problem with a print, try measuring your glass as well. 0.2mm is nearly one whole layer of print....
It's never showed up before, because my prints might have been wide or long, but either not on the base, or not both wide and long at the same time.
I'll let you know if the 0.15mm trench does the trick... It should do... But before you throw out that dodgy reel of filament, check your glass is uniform first.
I'll be replacing mine when I get home, if I can find some that's better...
Kim..

---

Please send me a PM if you have suggestions, or problems with Big Blue 360.
I won't see comments in threads, as I move around to much.
Working Link to Big Blue 360 Complete

I have been fighting the exact same problems. Here are a few bits of my experience.

Have you tried to check the flatness of the glass. Generally I have found that the edge of an extruded aluminium profile is extremely straight.
I had about 0.5mm lift in the middle, I thought. Until I checked my glass plate for straightness with several aluminium profiles - and also checked them against each other and against other very straight things. The glass plate was perfectly flat!!!

It turned out to be the x-axis acrylic plate that was bending. I could see that when checking it with my profiles.
Then I ordered and installed the aluminium upgrade....Down 0.3mm....so still 0.2mm to go. I got rid of the last 0.2mm by putting small pieces of stainless steel tape between the two aluminium x-plates - and keep checking the straightness of the plate, where the x-axis bearing runs.

I think my z-height is within 0.1mm on the whole plate now. This is probably as low as you can get with a machine of this quality.

I can now print PLA on the whole surface every time. I'm still struggling with ABS, though.

My suggestion is, that you check the straightness of the acrylic/aluminium plate, where the x-bearing runs.
I might be wrong. This is just a very short resume of the about 2 months I spent hunting this problem

Please let us know, what you find.

Bets regards,

Carsten

Carsten, did you check the flatness of the glass with the aluminium extrusion when it was on or off the bed? And did you also measure the apparent flatness of the glass in the Y direction, at different X positions? When I did this, there was an apparent bulge at the centre of the glass, about 0.1mm AFAIR. I believe it's caused by a slight bending of the glass due to its being clamped at the corners. Using 4mm glass instead of 3mm should reduce it by more than 50%.

---

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].

That is an excellent point you note there, getting the x axis bearing runner surface as linear as possible is vital, I too have packed my rib with the shim tape at two points to do this. It is worth spending a lot of time on this matter as the positioning of the x axis runner in relation to the nozzle means every tiny undulation is amplified.

The good point is that once it's done with aluminium parts, it should not need doing again unless you have a very bad crash.

Edited 1 time(s). Last edit at 07/12/2014 03:51AM by Davek0974.

---

Another RS Ormerod Mk1 meets the world

Retired now but I used to make....
CNC Machined Mk1 aluminium bed support plates for the Ormerod
CNC machined X-plates and ribs for Mk1 & Mk2 Ormerods
CNC machined bed support arms for the Mk2 Ormerod.
Dual Hot-End heatsink blocks.

@dc42, I checked everything and kept cross-checking back and forth until I was sure what caused my "bulging" - including on and off the bed, clamped - not clamped, my aluminium bed itself - you name it

What I forgot to mention was, that I no longer use the clamps, as they actually make the glass curve down at the corners. I have just taped the glass plate to the heating PCB with capton tape. It does not stick well, but as long as I have like 5 cm on all sides, the glass doesn't move. In this way, I can also use the whole bed - literally.

I have worked 14 years with building high accuracy radar systems from 3 kg to 5 tons, where we have constantly been fighting mechanical inaccuracies to the extreme.

I have skipped the z-probe, and just use a piece of paper for the z-calibration and a known spot for the x-axis. I have wired a microswitch ready to be mounted for x-homing, never got to the point of actually mounting it

Best regards,

Carsten

Quote
Davek0974
That is an excellent point you note there, getting the x axis bearing runner surface as linear as possible is vital, I too have packed my rib with the shim tape at two points to do this. It is worth spending a lot of time on this matter as the positioning of the x axis runner in relation to the nozzle means every tiny undulation is amplified.

The good point is that once it's done with aluminium parts, it should not need doing again unless you have a very bad crash.

I should not be necessary to shim the aluminiums parts, the x-rib is laser cut and straight and when mounted will straighten the x-axis-plate, if not straight, mill the x-rib before mounting

Erik

Quote
colsenfoto
..What I forgot to mention was, that I no longer use the clamps, as they actually make the glass curve down at the corners. I have just taped the glass plate to the heating PCB with capton tape. It does not stick well, but as long as I have like 5 cm on all sides, the glass doesn't move. In this way, I can also use the whole bed - literally....

I do not quite understand why clamps in the corners should make you glass curve, surely it cannot curve without something to curve over?

I have the PCB heater resting on 3 adjusting screws with the isolation plate hanging underneath in kapton tape
The 4mm glass rest on top of the PCB heater close to the adjustment screws, the weight of the glass makes the PCB sag slightly in the middle
I use a slightly modified version of the original clamps to hold the glass and the PCB/isolation plate together

Erik

Hi Erik,

I have a heavily modified bed - that's why it can curve

I didn't have the patience to wait for the bed to heat up to ABS-temperatures, so I bought a second heater-PCB. I have used thermal padding and paste between the various layers - whatever I could get hold of on my job. When clamping the glass-plate that sandwich acts like a pillow, pushing the middle of the plate up ever so slightly - but enough to give me problems.

I upgraded the PSU to a 650W, with two 12V outlets. One is driving the second heater-PCB through a MOSFET controlled by the voltage on the first PCB.

Besides 2 PCBs, the supplied aluminium heat-spreder and glass-plate, I also have a 5mm aluminium plate cut to a frame for stability. The whole bed is really heavy, but so far I have not seen any problems, I can relate to that. I have increased to current through the motors to 1000mA, though.

Regarding the x-rib: Mine wasn't totally straight, and the small stainless steel-pads were a very quick and easy way to fix it. It works perfect.

Best regards,

Carsten

Its interesting to hear peoples thoughts.
My glass actualy has a Belly front and back (a bulge). The belly adds 0.2mm which is nearly a printed layer thickness.
The first thing I will be doing is replacing the glass.
Packing out the X-axis Rib won't work, because the X-axis carriage doesn't run on it, apart from the back bearing to stop it spinning around.
The carriage runs on the linier bearing on the 12mm rod, so packing the rib will only raise or lower end of the rod.

What I think I might do is make the sub bed stronger, and use thinner glass. Then add another two clamps for the the center.
Once done, I can check for level, and shim accordingly using the center clamps to bend the glass flat.

I even thought of running bed compensation, but that won't work either as it assumes the glass is flat, and just compensates for an angle.
If the compensation used 6 points then that would be better, but that's life.

I'd like to use glass for the finish, but if I can't find a flat bit I'll use 4 or 6mm Ally.

@Carsten
Try rotating the 12mm Steel rod on the X-axis. Maybe the rod has a slight bend in it. The X-axis rib only supports the end of the rod, so shimming won't affect it.

Edited 1 time(s). Last edit at 07/12/2014 11:51PM by KimBrown.

---

Please send me a PM if you have suggestions, or problems with Big Blue 360.
I won't see comments in threads, as I move around to much.
Working Link to Big Blue 360 Complete

The linear bearing supports the head in a linear fashion, I.e. From X0 to X 200, the x axis arm in conjunction with the x axis runner bearing supports the head rotationally around the x axis I.e twisting.

Any variation in the shape of the x axis runner bearing track surface will lift and drop the head as it travels along the x axis.

As the rib width is thicker than the arm plate, any shimming, milling or filing of it's mating surface will gently pull the arm plate into a straight line, it's what it's there for - to give rigidity to the arm plate.

Now, due to the nozzle being further away from the linear rod than the x runner bearing, any movement of said bearing will be amplified at the nozzle - basic action of levers.

---

Another RS Ormerod Mk1 meets the world

Retired now but I used to make....
CNC Machined Mk1 aluminium bed support plates for the Ormerod
CNC machined X-plates and ribs for Mk1 & Mk2 Ormerods
CNC machined bed support arms for the Mk2 Ormerod.
Dual Hot-End heatsink blocks.

Nicely said Dave. But then that is how my Z-axis microswitch works.. By the lever moment... And it always works

Edited 1 time(s). Last edit at 07/13/2014 06:09AM by KimBrown.

---

Please send me a PM if you have suggestions, or problems with Big Blue 360.
I won't see comments in threads, as I move around to much.
Working Link to Big Blue 360 Complete

Yes, my glass also has a slight bulge in the middle, which acts to exacerbate the slight tilt caused by the twist in the middle of the X axis arm. I know it is the glass because a metal ruler will rock across the bulge. Bed compensation will not overcome the issue because it assumes an overall tilt of a plane flat surface - it won't compensate for a bulge.

Kim, to overcome it for large prints, I set a first layer height of 0.4mm and set the Z zero height at the lowest point of the glass (near the lowest edge), so the centre prints too close and so is squashed (but is still OK with the 0.4mm first layer height giving it leeway). I also recommend you use solvent cement for large prints. To help stop warping I paint some extra solvent cement over all the corners of the print *after* it has printed the first layer. My large prints are too big to have a raft (which would be another method to stop warping), but the additional cement stops the corners lifting (which otherwise they invariably do). OTOH I am printing with ABS, and the above may not work with PLA.

Dave
(#106)