E3D Water Cooled Mod

Discussion in 'E3D-v6 and Lite6' started by BurningDownTheHouse, Mar 25, 2014.

  1. BurningDownTheHouse

    Joined:
    Mar 25, 2014
    Messages:
    10
    Likes Received:
    1
    Hi there boys and girls,

    For those of you interested here are 3 versions of a quick, cheap and easy mod I did on my E3D hotend to turn it into water cooled.

    All three versions use the same principle but just vary in their degree of difficulty and desired result.

    Each can be made in about an hour(apart from waiting for epoxy to cure)

    They all run very cool even with a chamber temp of 75 degree C

    Version 1

    This version is the easiest in terms of difficulty and uses a 24mm water cooling jacket from an rc boat motor and is an almost perfect fit for the 3ED hotend. It can be found here http://www.aliexpress.com/item/SAITE-24 ... 81810.html or here http://www.aliexpress.com/item/Free-shi ... 44250.html

    [​IMG]
    [​IMG]

    The only modifications that need to be done are to the fins of the heatsink. They should be filed down as in the diagram below.
    Discard the O rings that come with the water jacket, they are not needed.

    [​IMG]

    Fins 1 & 10 should be filed enough to be able to get your favourite epoxy into the gap with a syringe and applicator needle ( I use J-B Weld) I would not advise using five minute epoxy.

    Fins 2 & 9 are the important ones. They should be filed just enough for the water jacket to slide over the fins and to be a close fit with the water jacket (this stops the epoxy from running into the water chamber) They fit pretty much where the o ring groove is.

    Fins 3,4,5,6,7 and 8 should be filed just a little to help water circulation The easiest way file the fins is to put the heatsink in an electric drill and use a small flat hand file to remove the material. Be careful not to file too much off. They file down pretty fast.

    [​IMG]
    [​IMG]
    Version 1 in situ

    Version 2

    This version works on the same principle as the first version but uses a 33mm length of standard 25mm x 1.6mm aluminium tube . It also needs more material to be filed off the fins than in the previous version and also a couple of holes have to drilled and tapped for the water nipples and so is slightly more difficult than version 1

    The advantage of this version is that the tube diameter is 25mm the same diameter as the fins of the unmodified heatsink. This is very handy if you already have printed parts that require the heatsink to be the same diameter as the original.

    [​IMG]
    [​IMG]

    Fins 2 and 9 are the important ones. They need to slide into the tube with a minimal gap.

    Drill and tap the water nipple holes before glueing the tube in position

    The  90 degree brass M6 threaded water nipple can be found here

    http://www.ebay.ca/itm/PZ152-1pc-90-deg ... 1cb&_uhb=1


    Version 3

    This version is the most difficult and requires some precision drilling and a piece of 25mm diameter x 1.6mm x 18mm long aluminium tube. But if you want a great, efficient, lightweight and compact water cooled hotend then this is it.

    [​IMG]
    [​IMG]
    Step 1.
    [​IMG]
    Insert a 5mm drill bit into the heat break hole in your heatsink as far as it will go. Carefully wrap tape around the drill bit at the opening (to use as a depth guide)

    Step2.
    [​IMG]

    Cut your heatsink in the position shown.

    Step 3
    [​IMG]

    Carefully drill the shortened heatsink with the 5mm drill bit to the same depth as the original using the tape as a depth guide. The hole must also be centered, so this is best done on a drill press. This is the most critical part of the process

    Tap the hole you have just drilled with a 6mm bottom tap

    Step 4
    [​IMG]
    Insert your heat break. Feed some 1.75 filament through from the top. If you have drilled correctly it should feed smoothly through.

    Step 5
    [​IMG]
    File the fins of the heatsink as above. Fins 2 and 5 are important, these should just slide into the tube with as small a gap as possible.

    Step 6

    Drill and tap the holes for the water nipples. Found here http://www.ebay.ca/itm/PZ152-1pc-90-deg ... 1cb&_uhb=1

    Step 7
    [​IMG]
    Insert epoxy resin into the gap with a syringe. When it has cured turn it over and do the other side.

    Step 8
    Insert the water nipples using a thread sealant/locker.

    [​IMG]
    Version 3 in situ


    Here is a PDF version of the above tutorial (this is a corrected version Sun April 13, 2014)
     

    Attached Files:

  2. Eaglezsoar

    Eaglezsoar Administrator

    Joined:
    Mar 7, 2014
    Messages:
    267
    Likes Received:
    0
    A very unique and inventive cooling solution.
    Thanks for sharing with us.
     
  3. mhackney

    mhackney Well-Known Member

    Joined:
    Mar 4, 2014
    Messages:
    53
    Likes Received:
    11
    Very cool (pun intended). I am completely sold on water cooling now that I have a Kraken running. I'm going to mod my E3D for water cooling like your 3rd option for a Mini Kossel I'm building.
     
  4. Paulnd

    Paulnd Member

    Joined:
    Mar 26, 2014
    Messages:
    4
    Likes Received:
    0
    very interesting, what pump are you using?
     
  5. BurningDownTheHouse

    Joined:
    Mar 25, 2014
    Messages:
    10
    Likes Received:
    1
  6. Paulnd

    Paulnd Member

    Joined:
    Mar 26, 2014
    Messages:
    4
    Likes Received:
    0
    Thanks for the information and links

    I'll be interested in details of your system, if it's not too much trouble.
     
  7. BurningDownTheHouse

    Joined:
    Mar 25, 2014
    Messages:
    10
    Likes Received:
    1
    [​IMG]

    My aim in building a printer was to be able to make large thin walled parts that could then be glued together to make much larger objects. High quality parts with minimum distortion was also an important objective.

    The kit I chose was a German RepRap X400 which has a print volume of 400 x 400 x 350 mm
    When constructing the printer I modified the volume to be 650 x 400 x 450 mm

    [​IMG]

    After some testing I decided the original steppers were under powered and skipped steps too easily. So I upgraded all the steppers to 60mm Hi-torque steppers and doubled the number on the X and Y axis
    I now have 2 on the X axis, 4 on the Y axis, 3 on the Z axis and 4 on the extruder.
    I have a heated bed and a 650 x 400 removable glass tray.

    [​IMG]

    It also did not take me long determine during initial testing that to reduce warping and layer delamination a heated chamber was necessary. So I enclosed the printer in 6 mm acrylic and proceeded to design a heating system that could heat the chamber to between 70 and 80 degrees C

    For a heater I modified a small 1500W domestic heater from the local hardware store and placed it in a box in the base of the printer. The temperature is controlled by a PID temp controller.

    With the chamber temperatures I was aiming for, water cooling of the steppers was going to be needed and so I used water cooling blocks from Syscooling http://www.syscooling.com/
    http://www.syscooling.com/products/Water_Blocks/89.html
    http://www.syscooling.com/products/Water_Blocks/88.html

    These fit nema 17 steppers pretty well. I have clamped them to the sides of the steppers without any thermal transfer compound or adhesive and still the heat transfer is excellent. The temp of the steppers I have measured at about 63 degrees C. If I remove the water cooling they shoot up to over 120 degrees C ( 80 degrees C is the maximum temperature a stepper should get to)

    [​IMG]

    water cooling of the extruder steppers

    [​IMG]

    X axis stepper cooling

    [​IMG]

    Y axis stepper cooling

    [​IMG]

    Pump, radiator and reservoir

    Extruder System.

    I have developed an extruder system that works extremely well with elevated chamber temperatures. One of biggest issues I faced with higher chamber temperatures was filament flow problems. When the temp in the chamber started climbing over 50 degrees a number of extruders I tried all had issues with filament flow, like intermittent flow or the extrusion stopping in the middle of a print for no apparent reason. Other issues were jamming and filament stripping.

    So I decided to design and make my own extruder to overcome these problems. My idea was that as the higher temps were softening the filament and thus causing distortion which then led to stripping, slipping and jamming. The reason for this I figured was the pressure of the idler wheel on the drive gear due to the higher pressure needed to stop slipping. By having two driven drive gears on each side of the filament I hoped to reduce the pressure needed on the filament and like a 4 wheel drive vehicle as opposed to a 2 wheel drive there would be more traction.

    Another aim was to design a way to to be able to feed a new filament through the extruder and into the hotend without having to retract and remove the filament already in hot end. This is a time when seems a lot of jams can start!

    I decided to use 2 motors in my design rather than 1 motor and a gearbox to get the dual drive. This would be much simpler to make.

    [​IMG]

    [​IMG]

    These images show the final product and it works extremely well – simple and effective.

    I also experimented using it in a Bowden type set up but I found the lag and long retract times I needed to use to prevent ooze were unacceptable. I think this was due to the long Bowden tube I needed. Also as an experiment I then tried setting it up as a Bowden and conventional together – worked like a charm!

    I now have a dual motor Bowden extruder in the base of my printer feeding the dual motor extruder at the hot end. I now have zero filament feed problems at 75degrees C!

    [​IMG]

    This set up has allowed me to build a simple auto spool changer in the base of the printer that is accessed through a pull out drawer. Now there is no annoying waiting around for the spool to finish on long prints or getting up at 2 am to change a spool. The change system over uses a micro switch, timer, small solenoid and DC motor. There is also an alarm that sounds at the change over to alert you if you are in the vicinity of the printer that a new spool can be loaded.

    [​IMG]

    [​IMG]

    [​IMG]

    If anyone wants more info on any of this I am happy to give details.
     
    Steltzer likes this.
  8. twicx

    twicx Well-Known Member

    Joined:
    Mar 6, 2014
    Messages:
    161
    Likes Received:
    0
    very interesting setup! Impressive to say the least! How do you heat such a large printbed?
     
  9. BurningDownTheHouse

    Joined:
    Mar 25, 2014
    Messages:
    10
    Likes Received:
    1
    The glass plate sits on top of a 5 mm piece of aluminium. On the underside of aluminium I have attached a custom 590 x 380 mm silicone rubber heating pad. The pad is glued with high temp silicone. The pad is 240V 800W. This is wired into the ramps heat bed circuit using a solid state relay. The pad I ordered online from China for the ridiculously cheap price of 38USD. So I bought two!
    I was very skeptical about the quality of the pad I would receive but as it turns out it looks OK and been working without any problems.
     
  10. Mike Kelly

    Mike Kelly Volunteer

    Joined:
    Mar 7, 2014
    Messages:
    439
    Likes Received:
    89
    Damn man that setup is amazing. Can you share more details about your heater control? What controller are you using?

    How do you keep your belts from loosening during heating?

    My next plan is to heat my enclosure but need to figure out the details before I dive in.
     
  11. twicx

    twicx Well-Known Member

    Joined:
    Mar 6, 2014
    Messages:
    161
    Likes Received:
    0
    That's very impressive. This is a textbook case of things i love about the 3d print industry. Someone has an idea to make a really big printer, and instead of people saying "uh that can never be done!" instead, people are like "no I did it, and here's how."

    excellent job buddy!
     
  12. Josh

    Josh Administrator
    Staff Member

    Joined:
    Feb 26, 2014
    Messages:
    183
    Likes Received:
    38
    Where did you get it from? I would be interested to have a play with them..

    On the topic of heated beds, Sanjay and I made this bed to print FEP on an early version of v6. It was quite exciting, we had it up at 415deg, and were running this bed at around 220deg! :)
     

    Attached Files:

  13. Eaglezsoar

    Eaglezsoar Administrator

    Joined:
    Mar 7, 2014
    Messages:
    267
    Likes Received:
    0
    I think you and Sanjay were trying to burn down the house for sure! 415 degrees? I am surprised the aluminum didn't start
    to melt!
     
  14. BurningDownTheHouse

    Joined:
    Mar 25, 2014
    Messages:
    10
    Likes Received:
    1
    This is the controller I use http://www.tcdirect.co.uk/Default.aspx? ... t_id=100/2

    But I now see they have a smaller cheaper one here http://www.tcdirect.co.uk/Default.aspx? ... t_id=100/9
    model 309-100

    I bought the relay version and just use on/off control

    You have to enter a number of parameters into the controller like thermocouple type etc. which is a little tricky to figure out from the very basic instruction sheet.

    The sensor is this one http://www.tcdirect.co.uk/Default.aspx? ... _id=230/35
    model 515-620

    I have it mounted at a slightly higher level than the hotend on the back wall. I have also removed the plastic cover. The response time is much too slow with the cover on. The sensor its self is very small.


    The timing belts I use are 9 mm wide HDT belts and I haven't noticed any loosening. They are quite tensioned at room temp though. But I will check that during the next print.
     
  15. BurningDownTheHouse

    Joined:
    Mar 25, 2014
    Messages:
    10
    Likes Received:
    1
    Yes it's the classic case of if somebody says it can't be done then there's a high probability that it can or at least given a bloody good try! :cool:
     
  16. BurningDownTheHouse

    Joined:
    Mar 25, 2014
    Messages:
    10
    Likes Received:
    1
    I can't find the original link for the pads but it was through searching Alibaba.com, but the email address is [email protected]
    and the persons name is Jim Zhang.

    The name of the company I sent the money to by bank transfer is called
    Honeywon International (HK) Limited. which if you do a Google search does show it as a heater manufacturer.

    You need to send him a drawing with dimensions and location and size of any holes you might want in the pad. Also give him the required voltage and wattage and indicate on the drawing the location and length of the power cord. They will make it in about 5 days.

    I actually got an email from him a couple of weeks ago asking if I needed anymore heating pads :)


    Wow that's hot!!! really pushing the boundries. You didn't mention whether the print was successful or not.
     
  17. Mike Kelly

    Mike Kelly Volunteer

    Joined:
    Mar 7, 2014
    Messages:
    439
    Likes Received:
    89
    Thanks for the info. I found this yesterday: http://www.amazon.com/gp/product/B00J9B ... 7P1BBKCZM8 and it seemed like a really good price, especially since it includes a thermocouple. Doesn't appear to do PID but bang bang is good enough for what I want to do.

    I'm curious about the belts. Before I built my enclosure I was looking to print with ABS so I pointed a ceramic space heater and the build. It ended up getting the belts so hot they skipped. Though I might need to look into a belt that's better suited for elevated temperatures.

    Here's a picture of my enclosure I intend to cut a 4" hole in the back part and plumbing in the ceramic heater seen in the background for heat.
     
  18. Josh

    Josh Administrator
    Staff Member

    Joined:
    Feb 26, 2014
    Messages:
    183
    Likes Received:
    38
    Thanks for the info about the pads, will keep that in mind next time Sanjay and I have a crazy idea. :)

    We printed an egg poacher to take full advantage of the fact that FEP can easily hand boiling water and that it has extremely low surface energy. Plus we are English, so it seemed like the perfect object.

    It printed successfully twice! Getting it to work involved using turkey-bags to create a heated chamber. The layer adhesion was perfect up at those temperatures and the material very translucent - I will try and lay my hands on the print to take a photo to show you guys.

    The only drawback in the system was the bed - we used silicone-backed FEP tape from 3M, which is sacrificial but when peeling the print off, the bottom of the cup was left behind.
     
  19. Eaglezsoar

    Eaglezsoar Administrator

    Joined:
    Mar 7, 2014
    Messages:
    267
    Likes Received:
    0
    :lol: How did the cup work with no bottom? :lol:
     
  20. Josh

    Josh Administrator
    Staff Member

    Joined:
    Feb 26, 2014
    Messages:
    183
    Likes Received:
    38
    The print worked.. the cup not so much! :p
     

Share This Page