Liquid Heatbed: A Concept

Discussion in 'General' started by twicx, Apr 7, 2014.

  1. twicx

    twicx Well-Known Member

    Joined:
    Mar 6, 2014
    Messages:
    161
    Likes Received:
    0
    A concept. Discuss.

    I've been doing a bit of research into a larger heatbed recently with a view to creating a really large format (1m x 1m x 1m minimum), printer with a low enough current draw for the heat bed. From what I can tell, currently, large print beds have fairly massive current draws when using things like kapton/pcb/silicon heat beds. However, I think I have a theory for making a large one with much lower current draws.

    My current printer uses 4 x power resistors to heat a 2mm aluminium print bed, and it does require quite a bit of power. Firstly, I'm taking this concept PURELY as a heatbed, without considering how it would be applied to X, Y and Z motion. Now, as far as the bed goes, what if we considered liquid heating. I know there's liquid cooling knocking about, but liquid heating is used in architecture quite a bit, and it can be quite efficient.

    The idea is to heat liquid in a reservoir and pump it through an aluminium print bed in order to heat it. But, in order for this to work efficiently, you need a number of things: 1. A liquid that doesn't turn to gas below approx 150 degrees. 2. One with a considerably lower specific heat capacity than water (ie., one that doesn't need as much energy as water to get hot). 3. One which is not corrosive. 4. One which is not toxic.

    I don't know enough about chemistry to pick a liquid, but it does seem to me that this could be done.... I think! Even with water, you could heat it to about 90 degrees without it turning to steam, but it does need quite a bit of energy, and keeping it hot over the length of time of a long print could be just as much energy use as a normal heat bed.

    In anycase, it's a concept. Anyone have any ideas on it?
     
  2. Mike Kelly

    Mike Kelly Volunteer

    Joined:
    Mar 7, 2014
    Messages:
    444
    Likes Received:
    89
    Based on the temperature you'd need something that's an oil. Peanut oil is a good choice because it's non-toxic, has a very high flash point, and is pretty cheap. I also don't believe it has much corrosion issues, but oil can be.

    If you could pressurize the water you could theoretically get it up to 150C without it flashing to steam.
     
  3. twicx

    twicx Well-Known Member

    Joined:
    Mar 6, 2014
    Messages:
    161
    Likes Received:
    0
    Oil is interesting indeed... especially with those kind of characteristics. I'd imagine something like peanut oil is probably viscous enough that it wouldn't need much more pressure than water to get it through a system? it would also need a lot of temperature monitoring and a good shut-off/failsafe system I think. What do you make of the concept?
     
  4. Mike Kelly

    Mike Kelly Volunteer

    Joined:
    Mar 7, 2014
    Messages:
    444
    Likes Received:
    89
    It's a good concept in principle and is actually pretty common to use oil as a heat transfer fluid. We use peanut oil a lot since it's so much cheaper than actual transfer oils. Most pumps that work with water will also work with oils, but you should check before you buy.

    You could do it with two stainless braided flex hoses, preferably insulated. Then you could use something like water heater inserts to heat the reservoir. It'll take a really long time to get to temperature but will maintain it very well. It'll also take a while to cool back down afterwards.

    I'd be interested to see it done but you'd need a really large build volume to make it justifiable.
     
  5. twicx

    twicx Well-Known Member

    Joined:
    Mar 6, 2014
    Messages:
    161
    Likes Received:
    0
    Yeah I know what you mean. Personally, it wouldn't bother me having a long heat up time, and there's a twitch of excitement at the thought of a printer that's a kraken (or 2, 8 nozzles because screw budgets) and a 2m x 2m x 2m build volume. Revive the art of coach building hahaha!

    But you're right, without thinking of the actual method of implementation, you would need a big print volume to justify it. I know there's a few buildings that use hollow-core concrete and heat the air/liquid within it to heat the building, but that's temperatures of 18-40 celcius. it's a different story when you're looking at 60-100!
     
  6. AlexC

    AlexC Well-Known Member

    Joined:
    Mar 25, 2014
    Messages:
    46
    Likes Received:
    0
    That's an interesting idea,. but when you go big you start dealing with more and more belt stretch, on my printer, at just ~307mm square, traditional reprap belts start to become to flexable,. and of course whatever weight on that axis multiplies the stretch,. so obviously you'll need much beefier belts (yah those are standard on mine, but better are on order),. but,. i start to wonder if the energy savings is worth the extra weight of the fluid on that axis carriage movements,.

    Personally if i where you id use Nichrome resistance wire, with a decent power supply,. the fluid option would technically be more energy efficient,. but the added weight you'll have to fight with fluids themselves, and the blocks needed to efficiently transfer said heat from fluid to plate, will result in slower max print speeds for quality, to gain that extra efficiency. IMHO, id rather have print speed than save a bit of energy. Plus slow heat up speeds may negate some of that efficiency. not saying it would not work,. im sure it would,. and id love to see it as an experiment,. just skeptical on overall performance, dependent on design.

    This is a hot plate that i built, it can get up to 60c at 12v, 4 Amps (55c is all iv ever needed for ABS with hairspray),. and the resistance wire barely adds any weight,. If I wanted to i could have easily made this 24v @ 2 Amps (its currently 2 parallel traces), and then you could multiply that out by however many square feet (it would just need higher amperage plugs/wires/supply),.

    And you can easily up the max temp by simply shortening the length of the resistance wire, sourcing resistance wire with a hair less resistance per foot, or upping the voltage by a little bit. (though that changes the math in the previous paragraph)

    Obviously when playing with experimental heating elements, don't leave them un-attended till they are properly tested over time, and get a fire extinguisher!
     

    Attached Files:

  7. Eaglezsoar

    Eaglezsoar Administrator

    Joined:
    Mar 7, 2014
    Messages:
    267
    Likes Received:
    0
    I don't know if your heated liquid bed will catch on the rest of the 3D crowd but it is a fascinating idea and if it works for you
    then all is well. I must admit that I would have never thought of doing that but that's why I have never invented anything.
    Best of luck to you.
     
  8. Creo

    Creo Member

    Joined:
    Apr 3, 2014
    Messages:
    13
    Likes Received:
    0
    Very interesting idea. Agree with Eagle though, not sure it will catch on...would make a printer more complex, you are only adding components(hoses, pumps, fluid...), you still need a resistive heat element.
    Using a fluid will make a stable system, holding a very steady temp on the bed which could not be desirable sometimes. I like to print first layer hot and turn down bed after....if I dont turn down bed, next layers warp horrendously as they stay above the glass transition temperature from close proximity to nozzle and heat bed.
    And, seems like your goal is to reduce needed current....this will not solve that goal as you will need more energy to heat up more mass and you will still be using a resistive heater.
    My thermo is a bit rusty but pretty sure things to consider for comparing energy needed for fluid heating compared to bed mounted heaters is the total mass of the system(parts to be at elevated temperature), thermal conductivity, heat loss to environment, and conversion efficiency(electricity --> heat) of heating element. Thats purely for the thermal consideration, you have an additional current requirement now for a pump. But, on thermal side, you've increased system mass so you will need more energy to heat to same temperature....more power if heating to same temp in same time.

    Fluid heat transfer would be the way to go though for multiple printers. One large central heat sink(fluid resevoir) with constant temp then flow control to individual print beds for needed temp. Or a printer with a heated chamber...again, one resevoir/heating element, then flow control seperately to bed and chamber radiator.

    Personally, if I planned to use in my house, I would choose mildly toxic over fire hazard and look for a non-combustible fluid. Really only have to worry about toxicity when charging system, but if a resevoir of peanut oil bursts into flames inside your house....you can probably imagine the horrors.
    Do like mkelly suggested, pressurize system and use water(distilled). I would also add a corosion inhibiting chemical that increases heat transfer like Redline Water Wetter. Or use the red type of coolant for automobiles with water.

    Definetly has merits. Just do it!
     
  9. twicx

    twicx Well-Known Member

    Joined:
    Mar 6, 2014
    Messages:
    161
    Likes Received:
    0
    Yeah I think you're right. The idea has something to it, but the execution could be very tricky! Still, its always good to have some ideas!
     
  10. ddseeker

    ddseeker New Member

    Joined:
    Mar 17, 2014
    Messages:
    1
    Likes Received:
    0
    A little off topic, but you can get some DIY ideas from Build it Solar

    I was contemplating building a solar water heater, and as I recall the system maintains 140° water tank temp (even during winters in Colorado) and If I am not mistaken it can get hotter but it is limited to 140° because of the EPDM Liner used in the storage tank.

    Now this would be overkill to make for a printer, but if you had/built one for your homes heating, all you would need to do is add another smaller Heat Exchanger in the tank, a small pump and then port that to your printers heated bed.

    Note: the tank water is only to store heat and does not go into your piping, so you can use peanut oil or whatever other liquid you want in your heat exchanger.


    While your at it you could even make a small solar space heater or a water to air heat exchanger for a heated build chamber.
     
  11. twicx

    twicx Well-Known Member

    Joined:
    Mar 6, 2014
    Messages:
    161
    Likes Received:
    0
    Hahahah! Well you know the saying, go hard or go home! See, I know you were saying it might be overkill for a printer, but it's like when you see overclockers using liquid nitrogen. Sometimes, the attitude has to be 'Well why do it?' 'Because I can.'

    It's a bit like what people were pointing out about mine, probably impractical for small format 3d printers, but if you were using solartherm for a heated print bed, and heated chamber, that was running nearly all day, and large format, then yeah, it probably makes sense. you could even integrate a small heater on a PV storage for times when it's not at the required temperature.

    Nice idea ddseeker! I approve! :p
     
  12. gixxer

    gixxer Member

    Joined:
    Dec 19, 2014
    Messages:
    2
    Likes Received:
    0
    Have you considered heat pipes?

    http://www.thermacore.com/thermal-basics/heat-pipe-technology.aspx

    They can be custom made to suit the bed. Just heat the ends and you have instant transfer across the length. Acetone filled ones would work at that temp. Ill dig out the company I used in the past if your interested. Best do some research as to how you'd heat the ends first though.
     

Share This Page