Glue Stick Works ABS & Nylon

There have been some posts appearing for over a year now about folks using glue stick to print directly on glass. I'll share my observations and would like to get feedback about what other folks are doing.

My setup is an E3D hot end with standard Robo3D heated bed. When I was running ABS with stock Robo hotend @225C, I was experiencing shredded wheat layer delamination on larger prints. Running now at 235-240C ABS seems to resolve the delam issues. My Ramps board had a bad solder joint on the heater connector which eventually started to smoke and delaminated the PCB trace. In repairing it, I also changed out the heater power FET from a 60V to a 30V FET. The lower rDSON of the 30V fet allows it to run much cooler to where it might not need a heatsink but gets more power to my heated bed. Now my bed will reach 110C and will hold 105C during prints with the dual cooling fans at 75%.

UHU is not available locally so I looked at Elmer's Washable School Glue (purple) and Elmer's Extra Strength Glue Stick from Office Depot. The EWSG had limited success on Nylon but the stick from EESGS was amazing! Had to chisel Nylon the part off the glass. Alternatively folks use water to redissolve the glue. Nylon was dryed, printed 0.15 LH, 250C first 245C remaining with 3mm brim, bed 105C. I am just getting started so far with Nylon and only have experience with small parts.

For ABS, favorite setup so far is 245C first layer, 240 remaining, 0.3 layer height, 3mm brim 105C. I apply a layer of glue stick to the bed then take a plastic baggie that has been spritzed with water and swirl the EESGS glue on the bed until it turns into a creamy layer, almost like a lotion. Just need enough of the stick lotion to make sure there are not dry spots on the glass. As it dries, the glue reforms in to a smooth uniform print layer on the glass. So far, thickness of the glue layer doesn't seem to make much difference as long at there is a dull haze on the glass.

Two strange observations printing ABS:
-Maintaining a uniform bed temp is vital. After printing, as the bed temp drops to about 30C there are loud snapping sounds as the ABS delaminates itself from the glass. The glue film is very thin compared to using Kapton, so there is nothing to spread out the shear forces as the ABS cools and the part just pops off. Have not observed this with Nylon yet.
-I have see some strange 0.25mm puckering patterns on the first layer as the filament is applied to adjacent layers, with measured first layer height of 0.27mm. The pattern seem to be somewhat random and propagates roughly normal to the filament direction.

I have printed some large 75% of bed ABS parts have not seen edge curl up when using a 3mm brim. Have seen some traditional edge warping when printing at lower temps 235C without brim which might be a function of stick lotion thickness. However there are some strange indentations on the first layer where 2-3 mm sections have pulled away from the glass in corners, but the brim has not detached. Looks more like an gas bubble than traditional warping.

Haven't tried PLA and it would be interesting to see what other people are seeing with ABS 225C which is more reasonable for the stock Robo3D hotend. I would be very nervous about pushing the stock hotend to to 235C because of its Teflon contents. My setup is calibrated using a thermocouple and I am running with custom thermistor tables.

Side note for people running with cooling fans directed at the hot end - there is an ifdef FAN_KICKSTART_TIME in Marlin that lets you run at lower fan speeds. It punches the fans at full power for nominal 100ms to get them started and then throttles back to your specified PWM setting. Able to run my fans reliably down to 29%.

 

Thanks for the FAN_KICKSTART_TIME ifdef! It will be very useful.

The bubble-like ABS warping on the corners is typical and will vary with the number of bottom layers and and the infill percentage. As usual with ABS warping if there is more material to contract there will be a larger force to pull off of the bed. Those corners are the max stress points.

 

After printing, try heating the bed to 120 and try using a spatula to get under one side of the part. Work your way around the print till it comes off. Wait till it's soft underneath the print then work your way around the print trying to find the weak spot.

I found using the glue stick worked good but it left a thick residue behind. You would have to clean it every print otherwise it actually left some artifacts on the print and led to possible uneven height issues.

I switched to hairspray. It too has to be cleaned once in a while but one application last a lot of prints.

Have you tried ABS with hairspray? I use the following here in Canada



Sent from my iPhone using Tapatalk

 

SteveC - I am inclined to agree. Could be max stress points as the filament wraps around corners forming the bubbles but the brim is remaining attached. Still seeing the scalloping lines. I'll try post some microscope pics.

Peter - Glue stick by itself will leave bumps. That is why I spritz it with some water and and foam it up into a bit of a lotion or cream. Turns into a glass smooth surface by the time the bed heats up. No need to remove the Elmers after a print. Add a touch more of glue stick and re-spritz and foam it up. I just put a sandwich bag over my hand and swirled it around for 30 seconds. Peter can you really get your bed to 120? I am 110 with a tail wind. Wonder if there was a design change. Any idea what your bed current is?

I had been successful using AquaNet Super Extra Hold hair spray for PLA, but it softens up too much with 100C bed heat to useful for ABS. Might have to give Alberto a try.

Seems like only the Nylon remains stuck with the Elmers back at room temp. Run a dozen ABS prints now with the Elmers without brim at 245/240 105 bed temp on smaller parts and not seeing any warping. ABS parts pop themselves off when the bed cools back down to 35C. Perfect Auto Eject system.

Another useful ifdef is to not turn off the z steppers after each move #define DISABLE_Z false. I was having some random z position issues that I made worse by turning up the Z current slightly. At the end of a layer the move Z command was being sent out to the Marlin buffer. From what I remember from the code is there is a look ahead that immediately re-enables power to the Z steppers if they have been powered down in anticipation of the up coming Z move. Drove me nuts trying to debug it, but there was some issue at the higher drive current that would make the steppers wildly jump. Could be ground bounce, but I never debugged it to that level. So at normal levels I would occasionally see a small jump. You might burn up a Stepper Driver module (not likely since they have internal thermal protection), but you're not going to burn up your steppers by leaving them on all the time. And even with this ifdef, they do power down after a while like X and Y. If it is ground bounce, it may be Stepper Driver lot or rev specific; some people may see it while other won't. My fix to add a uber large custom heat sink that fits over the A4988 Driver and have it put out slightly more current than the X/Y drivers which when it gets divided between the Z steppers is slightly less than what the actual X/Y steppers see. Also finally double sticky taped a fan pointed at the ramps board. Stuff can get warm in Robo base and all the stepper drivers were especially toasty.

 

Glue Stick update.
Here's a photo of the ripple pattern. Turns out it was cause by too high of a flow rate on first layer for hot end to glass spacing. Manually reducing the flow rate in Repetier or setting Z height correctly could make the problem disappear.

Now using a tooth brush to prep the bed for next print. Just dip the brush in a bit of water. Don't clean the bed between prints. Add a little more glue stick every so often. Since there is just a thin film of glue between the ABS and glass, it is vital to keep the bed at a constant temp. Fans blowing on my bed can drop the temp from 105C to <90C. Part is solidly affixed until temp drops to maybe 65C. The last loud snap crackle pop at around 35C means the part has completely freed itself from the glass.





Example Normal Bottom Surface finish: (B Gear has a bevel on the bottom; teeth did not pull away)

 

Still using a wet tooth brush (water) and a dab of glue stick between prints; no cleanup. However I am discovering that glue stick has some limitations w/ABS on parts with high internal stress. Picks below show a 30mm tall wall and the separation on a thin tall section of wall. If you look carefully at the pics, you can see the separation started directly below the wall and pried up the adjoining brim. No better success with a draft shield. One option would be to put a large hole in the side wall of the part that is lifting up so there would be no way for a large stress torque moment to develop. Still trying to use up my Bubbly Black Robo filament. Reprinted on Kapton with ABS juice and there was still some separation in the same area, but not quite as bad. 0.2mm LH 245/240C 105 bed no cooling. Need to be able to print large enough parts to build my Kayak, but stress releaving holes are not an option.

 

Would a heated enclosure help relieve the stress in the side walls? The side walls appear to be self delaminating because of the stress. You can see the layer separations in the pics. This was ToyBuilder Green ABS. Same cracks appeared in the Robo Black ABS.

 

What temperature is your room at? Looks like you could use an enclosure: http://forums.robo3dprinter.com/index.php?threads/mike-kellys-robo3d-enclosure.2712/