Thoughts, news and information about the world as seen through RV windows
RSS icon Email icon Home icon
  • Hacking the 101Hero: Testing the Head-Mounted Cooling Fan

    Posted on January 17th, 2017 admin No comments

    Earlier, I wrote about experiments I did with building a head-mounted cooling fan for the 101Hero.   It's in place and running now, so it's time to finish writing up the project.

    I use HVAC (Heating, Ventilating, Air Conditioning) terminology here.   The plenum is the upper part, which contains the fan, and gathers and directs blown air into the duct.  This is the narrower tube that guides the air to its destination.  The vent is the aperture or opening at the end of the duct that directs air forward and down around the print nozzle.   These parts can perhaps more clearly be seen in the earlier article.

    Making it Airtight

    Because I had printed this to be lightweight, I used a print height of 0.1 mm, with side/top/bottom layers of 0.8 mm (along with a slow print speed, very light support, and 20% infill overlap).   The result was that the plenum (the part that holds the fan) and duct (which leads the air around to the nozzle) were basically a mesh that leaked air everywhere.  I tested this by blocking the vent with my finger and blowing into the plenum.

    Fan housing with fan temporarily attached

    Fan housing with fan temporarily attached

    The inside surfaces were a bit rough, and I wondered about mixing white glue and water and pouring it inside to smooth the surface and improve airflow, but the thing is just so dang small!   I was afraid the glue would block off the duct.  The best I could do was cut a little strip of sandpaper to run back and forth inside the duct, and scrape the inside of the plenum with the square end of a small file.  After cleanup, tThe two parts fit well together; I glued them with CA glue.

    Then I mixed up some five-minute epoxy (JB Weld brand) and carefully smeared it all over the outside of the assembled fan unit.  I used this epoxy because it is both heat resistant and fireproof.   It took two coats to completely seal all the leaks.  So much for light weight!  It might have been better to print a sturdier part in the first place!  Or maybe use spray paint...   Then I sanded it a bit, especially the top of the duct that had to rest against the bottom of the print head.  I might have to sand more for a better fit.

    Another view of the fan and housing

    Another view of the fan and housing.  The "ears" had no epoxy.  You can see sanding on top of the duct (lower left) and the tape sealing the fan in.

    The fan dropped nicely into its bracket atop the plenum, but there was still a little air leakage around the fan frame, so I roughly sealed it with  transparent tape for the trial runs.  I also snipped off the tiny electrical connector and soldered on a two-pin job to hook up to the 5V power supply unit (PSU) that had been running my LEDs and print-bed fan.   While I'm using the head-mounted fan, I'll have no lights, and the little head-mount unit will be the only cooling fan.

    Printing Results

    With the fan running, there's a reasonable amount of air coming out, directed at the recently laid layer, about 3 mm back of the nozzle.  This was about where I'd wanted it.

    Since I had 101Hero gold filament in the printer and wanted to print some gold vases, that was where I started.  The purpose of these tests was to determine if

    1. the duct would clear the print
    2. not interfere in any way with the print process
    3. provide reasonable cooling

      The print from the front; the red arrows indicate the location of the vent

      The print from the front; the red arrows indicate the location of the end of the duct; the vent is on the bottom

    There was no off-switch in the experimental setup, so I unplugged the PSU for the first three (bottom) layers.  Once the walls of the vase (50 mm diameter, 95 mm high) started to print, I plugged in the PSU and left the print to run.   No issues, and fewer hairs than usual, despite the number of times the head criss-crosses the diameter of the vase (I don't know why it can't just go around the outside, but it doesn't, it does part of a wall, then crosses over and does part of a wall on the other side, then comes back and does more on this side....)

    Duct circled in red.  The air blows to the right and down in this image.

    Duct circled in red. The air blows to the right and down in this image.  The nozzle is at the left edge of the print, about 3 mm from the vent.


    Success So Far

    The duct is narrow enough that it clears the print, and did not in any way interfere with the process.  The print result was as good as those done with the more powerful 40 mm bed-mounted fan.  There were a couple of holes in the side walls.  This had also happened even worse with the BotFeeder translucent white, but not with the RepRap silver.  I think this is just a difference in filament quality.

    Here you can see the entire fan unit.

    Here you can see the entire fan unit.  It's not actually touching the print.   And yes, that's a hole in the wall of the print.

    The second vase in the series was printed oval, 60 mm x 80 mm x 50 mm high.  The extra width was to see if the suspension arms would hit the fan unit (they didn't) and the reduced height was just to make sure I didn't run out of filament and to keep the print time down.  This print showed considerable stringing and gobbing, suggesting that the cooling was not sufficient.

    The Next Series of Tests

    There was no heat damage to the vent area, so I might re-position the duct closer to the nozzle.  I'll also better seal the fan housing into the plenum so that all air blown by the fan goes down the duct and out the vent.

    The real test, of course, will be something tall and spindly like an Eiffel Tower, or a good test print like a Benchy.  I'll run those tomorrow using the silver RepRap PLA.




    Leave a reply

Supporting your fulltime RV adventures and aspirations