Discussion in 'General Electronics Chat' started by jjanes, Aug 11, 2013.

  1. jjanes

    Thread Starter New Member

    Aug 2, 2010
    I am looking at designing a circuit that will have a metal tube wrapped in nichrome wire I want to heat this to say about 120C. I know that thnigs change when dealing with different gauges and lengths of nichrome wire.

    My voltage source is going to be a 24V 400w Power supply and it will be doing other things as well.

    What I need help with is building a circuit that will maintain that temp say to with in 10C and I really have no idea what components are out there that can do that. I know I can get a thermister to get the temp back to the circuit but I do not know how to translate it.

    As far as I know the heat is maintained via turning the power off and on based on the current temp recieved from the thermister.

    I actually thought by adding a large resistor(pot) based on some type of calculation I could maintain it at a given temp while it is turned on but I am not sure if it is that simple. I would not have to readjust except if I wanted to make it a bit hotter or cooler at design time then it gets sealed up and packaged.

    Would something this simple actually work. If not what kind of circuit would I have to put in place to do this job

  2. LDC3

    Active Member

    Apr 27, 2013
    You are going to have a design problem. The nichrome needs to be electrically insulated from the metal tube, but heat needs to be transferred. I don't know what you can use.
  3. #12


    Nov 30, 2010
    The LM35 chip is rated to 150C and simplifies temperature measurement immensely compared to a thermistor or a thermocouple. Outputting 10mv/C will result in 1.2 Volts DC when you get to 120C.

    This chip makes temperature measurement trivial compared to the thermodynamics of the system. Know this: A perfectly insulated device with a tenth of a watt of heat being generated in it will eventually burst into flames or melt. You must do the calculations for heat loss in the finished product to find the power you will need to maintain the temperature. It isn't about how big the pipe is, it is about how well you contain the heat. You can't calibrate your circuit, then "package it up", and expect it to work the same. Think bigger. You have to consider the whole system and even the outside air temperature to get it right.

    Now that we have that behind us, +/- 10C is very doable. I dare say you won't even need a PID (proportional-integral-differential) controller. Just a fairly stupid, "on-off" circuit.

    Now I task you with telling us details about the size, flow rate, of what substance, in what package, in what atmosphere, how fast, and some things I didn't think of already.
    • LM35.pdf
      File size:
      303.7 KB
  4. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    Fibreglass tape. Or asbestos tape, although that may be harder to find these days.

    The big problem is like #12 said, you will have issues from variable heat loss.

    If your "pipe" has a flowing liquid etc (like a water heater), then the temp will change a lot depending on the initial temp of that liquid and it's flow rate.

    A good technique is to have dual closed loops with dual sensors and dual heaters. The first heater brings the temp up to about 8 degrees, then the second heater adds the last 2 degrees. The big benefit of this is the first heater brings the liquid temp up to a relatively stable, known, temperature. The second heater then has an easy job and can produce quite an accurate temperature.

    Obviously, guarateeing a constant flow rate will also help stabilise the system and provide better temperature accuracy.

    You did not say what this was for?
  5. jjanes

    Thread Starter New Member

    Aug 2, 2010
    Here is what I am trying to do and forgive any gross errors you see while reading this I have no idea what I am really doing and that is why I am asking these questions. I have a 3d printer and it uses a filament called ABS or even a nylon one. The problem is that wehn these filaments are taken from their sealed package they begin to absorb moisture from the air and when they do that develop undesirable aspects when trying to printas teh temp the filament is extruded at is about 180-230C.

    One of the resolution is to place the entire spool of filament in the oven for about 2 hours and just let the heat dry out the filament but I want to try a different route. In stead of doing the entire spool at once I want to have a tube the is just larger than the filament and I want to heat that tube to about 100-130 +/- 10C it is not super critical I get an exact temp just that it is significantly warmer than the evaporation point of water. I will run the filament through this tube and as it goes through it will be dried by the heat but not be in danger of melting as it is far from its melting point.

    My brains design went something like this a 2mm tube made from brass
    is covered in Kapton tape to electrically insultaed it from the tube Once this tube is electrically insulated I want to wrap it in the nichrome wire so it will heat the tube and then cover that whole thing in a hitemp RTV silicone caulk to eliminate and heat hazard on the outside as well. I will pass the filament through this tube assembly to remove the moisture.

    So I will have a tube assembly with nichrome wires ends sticking out of it. I plug those into a circuit with a pot on it and once the wire is getting the inner tube tube ~120C then I can enclose the this in a small case with holes going into the inner tube and coming out the other side dried significantly.

    Now that is the BASIC design in my head and I know it is full of problems
    I currently do not have ANY material on hand. I know that nichrome wire acts like a resistor and that the guage of the wire and the length have to be taken into account and that is where I get totally lost.

    I will be running this in parallel on a new printer I am building and it will be utilizing as 24V 400W Power Supply connected directly to the PS and as different things will be running and not running all the time I am not sure how much is going to get to this device. To save on room I was planning on bending the tubing into a elongated circle almost like a running track with maybe 8 inches to the long side and 4 on the short and 2 loops from entrance to exit so about 48 inches total length of tubing and probably 4 times that of nichrome wire.

    Since I have such a large area for the temp range I was hoping to set it and forget it. Basically turn it on and in a few minutes start erunning filament through it. so if it can be regulated with resistors or a pot then it is simple if I need to constantly turn it off and on toe maintain that temp the I need some help with what I need to have in place. The place I was going to order the nichrome wire in had several different gauges but I didn't know what to get so I came here in order to try and get assistance.

    I hope this is detailed enough to help me build something that will work. I know there are still a lot of blanks but at least it is a start. I will need some general pointing in the right direction but will give as much information as I can to help you help me

  6. atferrari

    AAC Fanatic!

    Jan 6, 2004
    Do you think that if the spool requires 2 hours, the filament moving through the tube will do in maybe seconds? Is there any sure way to check that?

    Maybe you could need a long path to ensure the results.
  7. atferrari

    AAC Fanatic!

    Jan 6, 2004
    First time I hear of a solution like that! Interesting.
  8. GopherT

    AAC Fanatic!

    Nov 23, 2012

    You are making this way too complicated. ABS does not absorb enough moisture to interfere with a print. It will also soften at the temperatures you are talking about. This causes problems because it will kink the plastic and change the diameter. Your hot end will have trouble pushing the filament through the extruder tip if the filament is inconsistent diameter or kinked.

    People who claim he ABS moisture is the source of their problem are chasing ghosts. They have other problems that they cannot figure out.

    Nylon is, in general, much more difficult to print using melted string and the standard is laser sintering. Nylon does absorb moisture. I would strongly suggest getting your printer up and running with ABS before getting distracted with nylon with the higher temp, small moisture issues and higher viscosity.
  9. gerty

    AAC Fanatic!

    Aug 30, 2007
    If you have an old coffee maker take a look at it. It have a heating element attached to a tube, you will however need to throttle down the power. Either by reduced voltage or PWM , to lower the temp.
  10. wayneh


    Sep 9, 2010
    Before embarking on this, I'd check the basic assumptions:
    1) Use an analytical balance to confirm gravimetrically that the filament does indeed gain weight due to water absorption. I'm skeptical.
    2) Confirm that your proposed treatment actually drives off the water gained. Drying an object is not always so easy - it can require long exposures at temperatures well above boiling, and dessicated air. Unless someone has done the hard work to define a process that works, you have to experiment for yourself. The usual approach is to set up multiple samples in a drying oven, pull them for weighing at timed intervals, cool them in a dessicator, and weigh them quickly. This develops the drying curve under the conditions tested, and of course you want to dry to constant weight. It's drudgery.

    If you really need a drying chamber, I'd forget the complexity of the nichrome wire tube. You just need a temp-controlled chamber with an inlet and exit port. Having a bit of thermal mass in the chamber would make it much easier to control. I'd be tempted to use a TEC (peltier) for the heating, since they're so easy to control and build around. A light bulb is another good choice, since you get visual feedback. Simple thermostatic control (on/off, nothing proportional) with a comparator and a LM35 will get you within 1°C.
  11. GopherT

    AAC Fanatic!

    Nov 23, 2012
    ABS is rarely dried before using for injection molding - it just doesn't absorb that much moisture.

    Nylon beads are usually dried before using nylon in injection molding. Various theories and temperature profiles are used (low heat for long time, higher heat for a shorter time, ...). In the end, it is all about surface finish and size tolerance for injection molders. For 3D printing, the surface finish will never be as good as a polished mold so you can stop worrying about that. I will bow down before you when you are able to measure size differences based on moisture content in Nylon for 3D printing (and prove they are not because of a stuck slide bearing, a clogged nozzle or a mis-alignment of x, y and z axes.

    Unless there is documentation somewhere that nylon will not stick to the print bed or stick to itself when 3D printing, then skip the drying step. Also, ABS is so easy to print and surprisingly strong. I don't know when you will need nylon - we haven't even thought about Nylon yet and we have been printing with a rep-rap for 2 years now.
  12. shortbus

    AAC Fanatic!

    Sep 30, 2009
    Having worked for two different companies that did plastic injection molding, for over 35 years, I don't agree with this statement. Both companies dried every thing before molding. ABS, Nylon, Polypro, all of the glass filed stuff, didn't matter it got dried before molding.
    Metalmann likes this.