Saturday, 12 June 2010

Wooden overcoat

I can make raft-less ABS objects on a heated bed pretty reliably, but when I try to do a whole bed full I get corners lifting on the objects near the edge of the bed. I think the reason is that the air around them is not as hot. If you think about it, with a moving bed machine like HydraRaptor, if you have a single object in the centre of the bed then you have a buffer of hot air around it. The bed only has to move by the dimensions of the object, so if the object is less than half the size of the bed then it remains inside that buffer. When you make objects near the edge of the bed the buffer is smaller and the bed moves further, so you get a double effect. To mitigate this effect I have halved the size of my biggest build trays. This is less convenient as a full bed is about 8 hours giving three shifts a day.

For example, with natural ABS on Kapton I was able to get away with a full bed like this: -

... but with white ABS on PET tape I would always get the odd part around the edge lifting, so I have to do it as two builds now. I bought a new reel of natural ABS from MakerBot but I can't make it work with HydraRaptor. I made a couple of objects but then it started to always jam after a few layers. The reason seams to be that because it is undersized at about 2.8mm, and my barrel has an internal diameter of 3.6mm, molten plastic back-flows up the barrel as far as the cold zone, where it freezes and makes the filament hard to push. This causes the filament to buckle and jam. I don't understand what has changed. The last reel I got from MakerBot a long time ago was the same diameter and I only finished using it recently and had no problems in the same extruder with the same settings. I switched back to white ABS and it works reliably again, so it must be something to do with the plastic.

Another problem with ABS is the fumes. My Mendel extruder seems to give off more fumes than HydraRaptor's does, perhaps because the melt zone is much bigger, and the white ABS seems to smell more acrid than natural ABS. I did a build with a window open to get rid of the fumes but most of the parts then warped, presumably because there was no longer a buffer of warm air around them, but a cool breeze.

In an attempt to tackle both of these problems I built an MDF box around my Mendel.

The front of the box is held on by magnetic door catches. It is sealed by door draft extruder strips and has a window made from plastic from a picture frame. This is glued on with silicone sealant.

The box is tall enough to allow the filament to enter through a single hole in the middle of the roof with a felt gasket that catches the dust.

The fumes are extracted by a tiny fan mounted in a chimney in the roof and piped out through a window vent.

I made a little pipe with a flange that fits into a slot in the vent and taped up the other slots with PET tape. I have another vent in the other window for fresh air in.

This fan is controlled by a spare output on my extruder controller and I have a thermistor to sense the air temperature in the middle of the chamber at the height of the top of the Mendel frame. Together with a small fan to cool the extruder heatsink and a large fan to cool the bed that uses up all the free outputs of my extruder controller, but not for the uses I originally envisaged.

I set the target chamber temperature to 40°C because that is as high as I dare to run the electronics and power supply. With the front closed the small fan cannot hold the temperature down and I have seen it go as high as 50°C without any ill effects. The extruder stepper was then too hot to touch though. Note there is no chamber heater. All the heat comes from the uninsulated bed, extruder and the motors and electronics, so I have actually reduced the total power consumption slightly and gained a heated chamber. To maintain 40°C I have to leave the front open at the bottom. I will add some vents at the bottom of the sides to allow cool air in and perhaps use a bigger fan.

Even with a gap at the bottom of the door I cannot smell any fumes. Since using the chamber nothing has warped provided the first layer outline went down properly as discussed in my last post. It also makes the machine very quiet although it was already much quieter than HydraRaptor.


  1. Any way that you could be convinced to post your blueprints and/or BOM for the cover? Pretty please? :)

  2. That bed full of parts is an inspiration, a beautiful thing to see. My Mendel build is complete and functional, but my knowledge of Skeinforge is incomplete and an unheated 3M #2090 bed precludes decent parts.

    I'm glad that you brought up the vapor and have created a solution for venting. The styrene monomer in ABS has effects on the central nervous system that are very nasty. It is also carcinogenic. My experience working in high concentrations of styrene vapor long before workplace safety became the norm taught me this in 1965-6. Some high school summers were spent building sailboats with polyester resins and a host of other toxic chemicals and particulates without any safety equipment.

    You can eat ABS without danger, but really don't want to breath the vapors released.

    Stratasys' own MSDS says wear a respirator, but their MSDS is rather benign. Link to the Stratasys data, 6 pages :

    The sobering MSDS is this single page, just 2 paragraphs whose address is self explanatory. I have personally experienced the symptoms listed in the second paragraphs and witnessed others people falling down. None of us want to go there.

    My favorite workstation sits in my bedroom on a hand made oak desk from wood that I logged and sawed myself. The Mendel is going into a self heated box outside, just as soon as I master Skeinforge. :)

  3. Hi Nophead,

    I too have noticed less curling on prints within an enclosed chamber versus prints made in the open air. With the acrylic makerbot-inspired design I use, it seems like the difference between a 20C ambient temperature, and the 30-40C the chamber gets just from warming up the heated table makes quite a bit of difference. This is probably especially true in the winter months.

    The lack of fumes is also super important, especially if you're running in an apartment (like me). I think I remember reading a post some time ago about enclosing a Darwin's build envelope with a temperature resistant bag to try and reduce warping -- and that by the time the print was complete and the bag was opened, there didn't appear to be any bad smells. Venting is good, but perhaps a poor man's extractor (that doesn't involve outdoor vents) is simply enclosing things and waiting for the fumes to settle.


  4. dnaworks,
    I will try to add some more detail soon and upload the plastic parts to Thingiverse. It is only a wooden box made from 6mm MDF, some battens, nails and glue. I actually made the mistake of not making it deep enough as I didn't realise the Mendel bed moves outside of the frame. I had to trim the bed.

    Thanks for the links.

    Presumably the fumes condense onto the bag. I do get tarry deposits on the metal bracket above the extruder on HydraRaptor, so perhaps that kills most of the fumes.

    I need to have ventilation in the Mendel box because it would get too hot for the electronics otherwise. A better arrangement would be to have the electronics outside.

    I suppose it could vent into the room through some sort of filter, perhaps just charcoal.

  5. Great! So this his the heated chamber :-)

    I also had that problems when I had window open, and I use that same idea to remove the pieces, blowing cool air with the air dryer :-)

    Now we just need support material :-)

  6. I just started a wiki page to document the heated room:

    Thanks for sharing, you did a great work NopHead!

  7. "To maintain 40°C I have to leave the front open at the bottom. I will add some vents at the bottom of the sides to allow cool air in and perhaps use a bigger fan."

    Why not turn off the Heated Bed on that time? Maybe the heated room will make print still working ok with heated bed at lower temperature (after the start of printing, for sure).

    Maybe we can save a lot of energy with the Heated Room + Heated Bed.

  8. I tried turning the bed off during the build. The parts come off due to contracting when doing ABS on PET.

  9. I really find the blog interesting and informative, Thanks for sharing.

  10. Nice work, but you are going to kill the motors quite fast in a chamber like that. A pity that we can't have the best of both worlds...

  11. Stepper motors can generally handle about 100C so not a problem.

  12. The problem you had with different diameter of filament in Makerbot ABS. I believe i have also experienced this problem with PLA. I had the PLA flow up and into the insulator and seal up the entrance. I could not heat the barrel enough to get it to flow again.
    What do you see as a solution other than getting filament that is the right diameter? Can this be prevented?

  13. How far up it flows depends on the temperature gradient as well as the diameter. If you are getting PLA going all the way to the top perhaps the insulator is too hot at the top end. I had that problem when started using a heated bed, which is why I have a big heatsink and fan on my extruder.

    The is a limit though to the difference in barrel diameter and filament diameter. I can run the 2.8mm Makerbot ABS in my Mendel as the bore is 3.3MM IIRC. Perversely, the closer the fit the easier it is to push.

  14. Thats great to know! Ill have to measure my filament often as well as the internal diameter of the nozzles im using. It should be interesting to find out what the +/- tolerance is. Graphed out. At what point should we really think twice before trying the filament. :)

    Thanks NopHead

  15. Hi Nophead,

    I have the same problem with wraping and smelling, are you still using a heated room for your printing?

    ( and is not only a problem of smelling, studies carried out in the 1970s and 1980s suggested a possible increased risk of lung cancer among workers exposed to acrylonitrile. ).

    Thanks for sharing.

    1. Yes I run my production machines in boxes with extraction fans.

      The natural ABS I get from doesn't produce much smell compared to others I have tried. Npt sure if that means it is less toxic though.