Reflow oven - temp sensing question

Comments

• Franklin Posts: 4,747

  2014-02-01 17:07 edited 2014-02-01 17:07

  I just want to keep it simple,

  Probably the simplest method is a thermocouple unless you can find another temp chip that can handle the elevated temperatures needed for reflow. You might look into contactless IR thermopiles.
• localroger Posts: 3,452

  2014-02-01 17:13 edited 2014-02-01 17:13

  For under $100 you should be able to find a serial-enabled multimeter with a high temperature probe. Using a thermocouple not only requires the special wires, it requires a non-thermocouple standard for the other end of the thermocouple because the thermcouple only tells you the difference between its ends, not the absolute temperature, so you also need to know the temp of the cold end by other means. A DMM will do all this for you. Example:
  
  http://www.amazon.com/TekPower-TP4000ZC-RS232-Interaced-Ranging-Multimeter/dp/B000OPDFLM
• TC Posts: 1,019

  2014-02-01 21:05 edited 2014-02-01 21:05

  xanatos wrote: »

  Just curious - is a thermocouple the only valid method for sensing the temp inside the oven? Or is there another preferred method that doesn't require special "K" wires and wire lengths? I just want to keep it simple, something I can control with a stamp, and that I can reasonably make conform to the "recommended" reflow soldering curves.
  
  With the ever increasing load of chips I am finding that I need that are no longer made in through-hole packages... I have to make this transition. I've been teetering on the edge of it for a few years now... gotta do it.
  
  Thanks for the info.

  
  Thermocouple is the easy route, with respect to temperature range. I am doing the same as you, just with a prop. Look up Maxim's MAX31855, it is simple 3-wire hookup, and only 3 output values.
  14-bit thermocouple value 0.25C per bit.
  12-bit cold end (internal temp) value 0.0625C per bit
  3-bit error value. OPEN, SHORT TO GND, SHORT TO VCC
  
  You can find them ready to go for under $20USD, and thermocouple probes (not wire) for under $5USD.
• SRLM Posts: 5,045

  2014-02-01 23:09 edited 2014-02-01 23:09

  I use a hand held IR thermometer to measure the temperatures. I have one of these ovens, and I can do some pretty fine lead free SMD. The trick is to bake for 2 minutes, then broil for 45 seconds. This will reflow a 2"x2" 4 layer board with QFN components (many with center ground pads) reliably.
  
  I've considered making a controller, but for a few reasons decided not to. The above formula works, and it's easy. I can spend 20-30 hours making a proper controller and testing it, or I can do other things. I don't think a controller will save me 20-30 hours of reflow monitoring, so I do other things instead. That's my value equation...
  
  ps: Don't you just love it when "Tech Specs" for a product don't include such basic information as size and weight? Or outlet voltage and frequency? Or output range? The Breville oven web page is pretty much useless.
• xanatos Posts: 1,120

  2014-02-02 11:48 edited 2014-02-02 11:48

  TC, that MAX31855 wins hands-down. Ordered 2 of them and a pair of k-type thermocouples. Everything else I'd seen so far had separate cold-junction sensors, amplifiers, all sorts of stuff. The MAX31855 is an all-in-one that'll do exactly what I'm looking for.
  
  SLRM - normally I'd agree about time spent in making vs. time saved - but this is just a project I want to make A personal satisfaction thing to get an oven working that conforms to the recommended temp curves and actually solders my stuff. I also do expect that it may actually see some volume if I can perfect my photo expose/etch process to get really tiny little traces without any burn or bridging. I'd really love to be able to reliably go full-tilt surface mount, and leave the through-hole stuff to breadboard prototyping exclusively.
  
  Thanks folks!
• nathantrantham Posts: 5

  2014-04-14 18:26 edited 2014-04-14 18:26

  Xanatos and TC -
  
  I'm in the process of making my own Parallax reflow oven as well, so it is great to see these recent posts about them.
  
  I too have found that the MAX31855 is easy and affordable and seems to work well. I use Jonny Mac's object for it.
  
  My current situation is I have everything up and running, but the specific reflow temp profile PID algorithm is giving me trouble. I have tons of overshoot on the temperature.
  
  Have either completed your project? If so, can you post your code for the temp profile please?
  
  Thanks!