gfxgfx
 
Please login or register.

Login with username, password and session length
logo
 
gfx gfx
gfx
19852 Posts in 1275 Topics by 5182 Members - Latest Member: charbuild March 29, 2024, 03:17:06 AM
*
gfx* Home | Help | Search | Login | Register | gfx
gfx
Breaktru Forum  |  eCigarette Forum  |  Modding  |  Topic: Has noone tried PID controlling for temp control yet?
gfx
gfxgfx
 

Author Topic: Has noone tried PID controlling for temp control yet?  (Read 7295 times)

0 Members and 1 Guest are viewing this topic.

Offline norcalreballer

  • Sr. Member
  • ****
  • Joined: Jul 2014
  • Location: USA
  • Posts: 63
  • Karma: +4/-0
  • Gender: Male
Has noone tried PID controlling for temp control yet?
« on: January 26, 2015, 11:55:31 PM »
I have some experience designing an arduino PID Rework station controller in the past.  I'm still a noob, but I think I could get it to work in a mod.  The only issue is running the thermocouple up through the center post and to a coil, but I honestly think it'll work.  I used to have some 36g Omega TC's that would fit right through with a modified 510 connector and atty.

Any thoughts?  People seem to like temperature control.  I dont' think it could get any better than that.  Any wire will work etc.

Offline CraigHB

  • PV Master
  • *******
  • Joined: Nov 2011
  • Location: Reno, Nevada
  • Posts: 2023
  • Karma: +246/-1
  • Gender: Male
Re: Has noone tried PID controlling for temp control yet?
« Reply #1 on: January 27, 2015, 02:37:18 PM »
If you can do some micro-electronics, you may be able to add a small unit to a customized atomizer to provide a wireless link to the sensor.  Otherwise, you need an extra contact in the connector for a sensor.  That would eliminate the ability to use a 510 connector.  Maybe you could add a simple jumper plug that you stick in the atomizer after screwed in to a standard connector.  I don't know, just ideas off the top of my head.  It would be a lot better than being limited to a particular type of low resistance wire.

Offline norcalreballer

  • Sr. Member
  • ****
  • Joined: Jul 2014
  • Location: USA
  • Posts: 63
  • Karma: +4/-0
  • Gender: Male
Re: Has noone tried PID controlling for temp control yet?
« Reply #2 on: January 29, 2015, 12:34:41 AM »
 It's only a matter of a max1855 and a k type thermocouple.  Omega has 36g thermocouples that will fit right through a Centerpost that's modified. 

Tuning will be the only hard part.  I'll test when I get some time.  I already have all the parts from the previous project and I even know how the popular PID library works. 

edit..  Wireless will require too much bs

Offline CraigHB

  • PV Master
  • *******
  • Joined: Nov 2011
  • Location: Reno, Nevada
  • Posts: 2023
  • Karma: +246/-1
  • Gender: Male
Re: Has noone tried PID controlling for temp control yet?
« Reply #3 on: January 29, 2015, 01:13:09 PM »
I don't know what software you're using, but tuning a feedback loop for something like a thermocouple is typically pretty easy to do in software.  Doing it in hardware is where it becomes more involved.  You only need the proportional and integral part of the feedback loop, the derivative doesn't really do much for you and you can disregard it.  In any case, a proportional-integral feedback loop can be done in software with just a few lines of code.  It's actually surprising how easy it is when using a minimal approach.

For example, here's the formula in discrete terms for one I did recently,

Vc+1 = Vc + ( Vref - Vfb ) / (5 * Vref -2)

Vref is a constant, Vfb is the feedback voltage and Vc is the control input.  You can see it only takes a few lines of code for the math and a few lines above and below to grab and send the data.

The interesting thing is, in hardware this is a huge task to accomplish requiring the use of differential equations and advanced mathematical methods to find component values.  Doing it digitally in software greatly simplifies the task.




« Last Edit: January 29, 2015, 01:29:54 PM by CraigHB »

Offline norcalreballer

  • Sr. Member
  • ****
  • Joined: Jul 2014
  • Location: USA
  • Posts: 63
  • Karma: +4/-0
  • Gender: Male
Re: Has noone tried PID controlling for temp control yet?
« Reply #4 on: January 30, 2015, 12:36:37 PM »
There's a popular arduino PID library that I've messed with before.  It has many features and it's easy to use.   That's what I was referring to,  though.  I tuned it before with a toaster oven for reflows and it took work to even get that right.  Just moved,  but hoping to grt some experimenting time soon.


Offline CraigHB

  • PV Master
  • *******
  • Joined: Nov 2011
  • Location: Reno, Nevada
  • Posts: 2023
  • Karma: +246/-1
  • Gender: Male
Re: Has noone tried PID controlling for temp control yet?
« Reply #5 on: January 30, 2015, 03:24:00 PM »
I'm not familiar with that library.  The times I've had to do it, I just stick a few lines in code and adjust gains on the bench to get the system behaving the way I want. 

You can probably just set the derivative gain to zero and adjust the proportional and integral gains.  That should simply things quite a bit.  Derivative compensation is typically not required unless trying to tune systems that are inherently unstable.  Most systems are what's called "first order" as opposed to "second order".  First order systems only require proportional and integral compensation.  In fact adding derivative compensation to a first order system can destabilize it.  Even some second order systems can be stabilized without derivative compensation.

Offline norcalreballer

  • Sr. Member
  • ****
  • Joined: Jul 2014
  • Location: USA
  • Posts: 63
  • Karma: +4/-0
  • Gender: Male
Re: Has noone tried PID controlling for temp control yet?
« Reply #6 on: January 30, 2015, 08:21:07 PM »
Awesome!  Thanks for the tips.

Offline Zanderist

  • Full Member
  • ***
  • Joined: Feb 2015
  • Location: SI, NY
  • Posts: 40
  • Karma: +5/-0
  • Gender: Male
Re: Has noone tried PID controlling for temp control yet?
« Reply #7 on: February 07, 2015, 06:02:00 PM »
This something I've been thinking about for sometime now. My idea though was to do it without having to modifiy a RDA.

The idea would be to monitor the temp of a mosfet in series with the coil and then interpolate from reading the mosfet temp what the coil temp would approximately be.

Also I even thought as an experiment to create a transfer function by having one sensor on the mosfet and one sensor on the coil which would be 1 ohm.

The only draw back is that the mosfet would heat up at a rate different from the coil which would heat up almost instantly. So in addition you would have to factor in that lag.

Offline norcalreballer

  • Sr. Member
  • ****
  • Joined: Jul 2014
  • Location: USA
  • Posts: 63
  • Karma: +4/-0
  • Gender: Male
Re: Has noone tried PID controlling for temp control yet?
« Reply #8 on: March 18, 2015, 12:23:18 AM »
You can get some super light gauge(40awg +)k type thermocouples.  Even thermocouples with flat beads.  It would be the same as a bottom feeder.  I would try this, but I'm not into temp control.  Still a great idea.  I'm not sure how the mosfet idea would work.  Using feedback from the coil itself would be awesome.

Offline Zanderist

  • Full Member
  • ***
  • Joined: Feb 2015
  • Location: SI, NY
  • Posts: 40
  • Karma: +5/-0
  • Gender: Male
Re: Has noone tried PID controlling for temp control yet?
« Reply #9 on: March 18, 2015, 12:48:05 PM »
Mosfet wouldn't work, I've did an observation with a temp sensor.  What happens is that the mosfet heats up at a rate from the coil and cools down at a different rate than the coil.  Imagine two parabolic curves one for the coil and one for the mosfet, the X axis is Time add Y axis is Temp. Now one curve will have a steeper rise to the max temp point then  the other one because of heat capacity. The mosfet will take longer to heat up and longer to cool down, while the coil will heat up almost instantly and cool down with an equal amount of time it took to heat up.


Breaktru Forum  |  eCigarette Forum  |  Modding  |  Topic: Has noone tried PID controlling for temp control yet?
 

gfxgfx
gfx gfx
Powered by MySQL Powered by PHP Valid XHTML 1.0! Valid CSS!