Cast Tooling Plate Print Bed Flatness

I clamped a digital dial indicator to my printer's extruder carriage and wrote a gcode file to move the gauge around the surface.

The bed is a 12.5" x 12" x 1/4" cast tooling plate covered with 5 mil Kapton tape.

For this test the bed was heated to 105C and the everything closed up for about 1/2 hour to let the temperature inside the build chamber (and the dial indicator) stabilize. I found through experiments that the readings from the dial indicator drop the longer the probe is in contact with the hot bed.

https://vimeo.com/142670595

The first measurement at the center of the bed is -0.13mm and the bed sags about 0.10 mm away from the center and drops as much as 0.14 mm at the unsupported corners.

If the first print layer is set to 0.25 mm it should be possibe to print edge to edge.

Edited 2 time(s). Last edit at 10/16/2015 07:53PM by the_digital_dentist.

---

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

Good information!

Did you check the difference between cold and hot?

---

Printer I bought: 2015 Sunhokey Prusa i3
Printer I am designing: Another big CoreXY machine

Yes. It reads more consistently and with a little less variation across the bed when cold. I think heat creeps up the probe and causes shifts in the gauge readings because repeatability is poor at high temperature. I may have to lift the probe off the bed and let it cool between measurements to get consistent results, but that adds any backlash in the Z axis to the numbers.

The measurements include errors due to flex in the X and Y axes and extruder carriage, and slop in the bearing block that the carriage rides on.

Edited 1 time(s). Last edit at 10/16/2015 10:37PM by the_digital_dentist.

---

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

Not sure how heat is going to affect the alum extrusion frame and other structural parts. I am planning to have a heated chamber
After the whole thing is modeled in Solidworks, I need to some thermal simulation to see the displacement.

My heated bed is planned to be 400mmx600mm. length maybe longer (700mm) if I planned the Z axis well.

Edited 1 time(s). Last edit at 10/18/2015 12:39PM by bonmotwang.

---

Printer I bought: 2015 Sunhokey Prusa i3
Printer I am designing: Another big CoreXY machine

Someone suggested that the variation in surface flatness I'm seeing may be due to lateral forces on the bed support screws, especially when the bed heats up. I calculate it should expand about 0.7 mm in X and Y. If the screws are tight against the bed the lateral forces could cause the surface to deflect.

The deflection isn't severe, but is larger than expected. I am able to print almost edge to edge as it is, but since I now have some measurements it makes me want to see if I can figure out what's causing the deflection.

I did some work on the undercarriage at the makerspace today, and when I reassemble it I'm going to set it up so only the reference screw is solidly attached- I'm going to let the two adjustment screw mounts slide a little and see if that has any effect on bed flatness. I ordered a short linear guide via ebay for $9 and can add it to the bed support system- I'll bolt the bearing block to the undercarriage and put the reference bed support screw through the guide rail. That will limit motion very precisely to the Y axis.

Edited 1 time(s). Last edit at 10/18/2015 06:56PM by the_digital_dentist.

---

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

Thanks Mark for the information.

I will do a thermal simulation with the frame and other details when the design is done.
But I was planning to make the bed "floating" or "sliding", similar idea as the sliding mechanism on some bridge design.

Edited 1 time(s). Last edit at 10/18/2015 08:05PM by bonmotwang.

---

Printer I bought: 2015 Sunhokey Prusa i3
Printer I am designing: Another big CoreXY machine

I did a bunch of stuff to it today. I pulled the undercarriage and bed plate off the printer, chopped about 660 gms of excess aluminum out of the undercarriage plate, and checked the hole spacing for the bed support screws. They were indeed too close together, putting force on the bed plate that increased with temperature because of the expansion of the bed plate. I decided to try a simple approach before going to the trouble of mounting a couple linear guides under the bed plate- I simply enlarged the holes in the bed plate and the undercarriage plate which allowed me to get everything set up with a passive fit.

I reran the low temperature zigzag test with the dial gauge and found the variation over the bed surface was down to about 40um max from about 80 um previously. I heated the machine up and reran the 105C zigzag test and found the corners still droop a little, but now the absolute maximum is 70um where previously I was getting 140 um of droop. Over about 90% of the bed surface the variation is under 30 um.

I don't think I'm going to mount the linear guide on it. It's flat enough to print edge to edge. I'm running a test print now to make sure I didn't screw anything up. If the results are good, time to move on...

Original bed plate, top and bottom views:





New bed plate:

---

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

Error! The photos above are NOT the bed plate- they show the old and new undercarriage plate.

---

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

Quote
the_digital_dentist
Error! The photos above are NOT the bed plate- they show the old and new undercarriage plate.

Either way I do love looking at drilled, machined aluminium, with the price of this 6mm plate it makes no sense to use perspex, its the same price(almost) unless you cant find a decent machine shop to do the work, do you have your own tools for that side of things or you outsource?

I am a long time member of the Milwaukee Makerspace. We have a well-equipped machine shop where I can mill and drill and turn until I'm blue in the face. We also have laser cutters, 3D printers, a forge, pottery equipment, sewing and embroidery equipment, silk screen printing, a wood shop, welding and cutting equipment and a few other things I don't recall now. I learned to do some simple operations on a milling machine to build my printer, starting with milling the frame pieces square and to equal lengths, precisely drilling holes, etc. I'll be replacing the 3D printed X axis motor mount with a milled aluminum part soon- I will do all the milling myself.

---

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

Cool, the closest facility I have doesn't have such a wide selection of tools, otherwise i'd set my camp bed up there.
They only have 3 decent machines, 3D printers (too slow) Laser cutter (much faster but only perspex n ply) and a larger router for a bit of wood. I've yet to find anywhere I can get time on a mill or lathe, even the local college is struggling to get a decent finish out of the machines they have, but I would have to enrol for a term and pay £3000 to get a little time there.
I've been making the case that every town should have a public makerspace but maybe in another 20 years government might wake up!

Start up a space yourself! Start looking for like-minded people and start meeting somewhere (pubs are good!). As you get more people, you'll find that more stuff just magically appears. People have machines that are sitting unused, etc. Eventually you'll find a space to work in- one of the interested people will have an unused garage or building that can be used, etc. It takes a while to build up the momentum, but once started, it will start rolling on its own. Put up notices on bulletin boards at schools, pubs, offices, etc. Set up a web site for info and notices. Go!

---

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

Yes I'm trying, suggested to the local unemployment office to provide a list(or setup meets) of unemployed engineer types(though would be nice to get a little help from those in industry), but data protection prevents it, but they should be asking them individually if they would be interested in such a club/group, (but they won't) next step would be to run some add's for people with tools or skills, but so far I just been trying to build my portfolio of things that could be made, when we have the tools & space, have a little cash, but if I bought a machine I'd have nowhere to put it, would rather have the help of a little matched funding & a few contracts for training, to get things going, slowly making progress on my own but I'm sure it would be faster & more fun with others. If I had a little success selling printers, I would plough funds back into such a space(assuming there's was any change from $300), but making printers to sell is full of it's own issues. Making machines just to make more machines is much more fun.

Edited 4 time(s). Last edit at 01/20/2016 03:43AM by MechaBits.

Amazing. There are guys like you that talk craft co-ops and community involvment and then there are #@*&heads like the punk selling our Thingiverse drawings on ebay. Can't undrstand how humanity can vary so greatly. All of the givers should be proud. Thanks.

it's just a logical extension of my BeatSkool... the idea was to teach more things that are relevant for today, Beatskool was never really just about music, but computer tools & media, creation, and now encompassing 3D printing & machining...if you had 40k to spend (zero interest, dont pay it back till you can afford to) would you spend it on an education, or 40k worth of tools... I know what I would pick.

Edited 1 time(s). Last edit at 02/26/2016 07:30AM by MechaBits.