avoid warping with large, flat object

Do you have an enclosure? I've done 7" parts in ABS on glass, bed at 110C, no brim, no warp.

The thing that has worked best for me is using a sheet of glass over the heated print bed with ABS glue covering the surface that will contain the part.  I made ABS glue out of previous failed prints added to acetone in a glass container.  Let the ABS melt into the acetone until it forms a thick paste, then apply it to the glass and allow the print bed to reach temperature.  It essentially acts like a raft/brim for the entire print.  I have now made separate batches of ABS glue for every color filament I have so that the bottom of the print isn't a different color.

+1

What's a brim? I know no uses for such things. Even with bed at 100C, I only use glass and hairspray. I never have issues.

I've never used ABS slurry,too messy.This trick is just as good.

Let the printer make the brim and first layer.While it's printing,take a 1/4 in wide artist's brush,dip in acetone and go around the entire outside of the brim.This will melt the brim to the bed as good as slurry would.

With an open part like that I suppose you could try going around the inside of the print with acetone as well but I'd leave that as an option.

thank you for all your overwhelming support guys!

Yes I have an enclosure and my first layer is calibrated, as well as the bed. I used the paper method, I hate it because it almost destroyed the printer, the limit switch doesn't limit z-travel with manual commands...

To be able to print without the brim would be a dream!

To sum it up, here's what I'm going to try, and keep you updated:

- warm the bed at 100-110C and give the glass time to warm as well. Will check with IR thermometer even if I know it's not a reliable tool (Elmoret said it countless times hehe)
- increase extrusion multiplier (calibration cube has 0.39mm thickness instead of 0.42)
- lower first layer height for a more squished first layer
- increase brim width to 7mm
- try the ABS slurry or a different brand of hairspray - sorry I don't think they sell Aquanet in Europe

Other tests:
- try the artist brush trick to melt the brim
- replace glass with ceramic tile (but the brick and mortar store sells only 20 tiles per pack haha)
- use PLA (but this introduces other issues compared to ABS)

105C with no brim didn't work, but 110C with 4mm brim worked!

Unfortunately no one warned me about the MAXTEMP safety switch, which tripped in the middle of print when it reached (after 2 hours) 110C. I've also noticed adhesion improves when you already printed over a couple of times on fresh coat of hairspray

So this is the result... Next evening, next try

I will consider your first layer advice Chuddly for other objects, but for this one I think the brim is the best approach, because it gets removed very easily and leaves a sharp edge. With a smashed first layer I guess you have to sand it a little.

My stock bed took 2 hours to reach 110C with an enclosure, I guess you pimped your heated bed

Hey agentsmith here's another trick that no one mentioned, glue stick!

There's several benefits over hair spray, you can put it much more accurately (for instance just in the edges of your print), you can apply it with the glass already inside the printer, it wipes of with just water and because it is REAL glue, it sticks to the glass all the time.

The only disadvantage really is that you have to reapply on every print but if anything else fails or it's too complicated it's worth a try.

I have 2 suggestions:
1. I have posted this around this board before, so it is starting to sound like a skipping record.  Use an ABS slurry over the hairspray, but make the ABS slurry using ABS pipe cement solvent available at the hardware store.  The MEK in the solvent welds the ABS to the hairspray.  Acetone does not.  Since the hairspray sticks to the glass really well, you now have a complete bond from part to bed.  I print large parts like that all the time on a 95°C bed.  This may be all you need to solve your problem.

2. You can put 1cm counterbores all over the inside of your enclosure that leave ~1mm wall thickness.  This will break up the linear shrink stress that causes the warping.   

Look at the sliced model and iterate the counterbore depth to leave complete extruder paths so that there is no zig-zag infill in the remaining wall.  If you are successful then you won't see them on the outside of the part.