Author Topic: roaster control widget  (Read 56858 times)

milowebailey

  • Guest
Re: roaster control widget
« Reply #285 on: November 15, 2009, 09:10:45 PM »
John

Look at this thread too... it will tell you what you need to do.  An AC fan will act as a light from a control standpoint.

johnr

  • Guest
Re: roaster control widget
« Reply #286 on: November 16, 2009, 08:09:35 AM »
Great working example - much appreciated. :) It works pretty much as I was thinking, which means I'll either have to write a kernel mode driver or interface with another board (maybe an Arduino) to do the high-precision timing stuff. Hey... then again, maybe I'll dust off my copy of DOS and write a VGA version of the roast controller in Turbo C... I wonder if there's a DOS driver for my Phidget I/O board... :P

Offline Mlee

  • Standard User
  • *****
  • Posts: 1212
  • Roaster-Hottop b Espresso-Gagggia espresso
Re: roaster control widget
« Reply #287 on: November 16, 2009, 10:05:14 AM »
Ok, so let me see if I understand... in a nutshell, you clip the sine wave by employing a 60hz-based PWM period (16.67ms) in which the 'off duty' cycle begins on the zero cross boundary, which is indicated by a distinguished INT input value/range? Do I have that right? If so, I may have to explore hw alternatives as it will be almost impossible to achieve the necessary timing in my non-realtime, software-only setup (I've got 10ms timing granularity to work with under the best of conditions).
What the heck ??? ???
Prov 3:5-6-Trust in the Lord with all your heart
                Lean not on your own understanding
                Acknowledge Him in all your ways
                And He will make your paths straight

johnr

  • Guest
Re: roaster control widget
« Reply #288 on: November 16, 2009, 01:07:48 PM »
Sorry, I have a tendency to be terse... PWM=Pulse Width Modulation is (in this context anyway) just a fancy term for a technique that's used to flip a digital switch on and off in order to drive some analog device (like a heating element or a fan) at less than 100% output. Terms and details vary but the basic idea is that you have some fixed time period that is broken up into two parts: an 'on duty' cycle and an 'off duty' cycle. During the on duty cycle (or just 'duty cycle'), the digital switch that controls the analog device is ON and during the off duty cycle, the switch is OFF. If we want to average 60% power over a 10 second period using this technique, for example, we'd  just use a 6 second duty cycle (10 seconds * 60%) - the switch would be on for 6 seconds, off for [the remaining] 4 seconds, then back on for 6 seconds, and so on.

The period used is selected based on the device you're trying to control. Generally, the more slowly a device reacts to the input, the longer the period should be - heating elements, in particular, are notorious for taking a relatively long time to heat up and cool down (thermal inertia) and PWM periods on the order of seconds are usually appropriate.

Naturally, there's always a better way to skin a cat (is there really a bad way?) and that's where zero cross detection apparently comes into the story. It turns out that some analog devices react badly (or at least perform sub-optimally) to being switched on/off at arbitrary points in the AC cycle, as is necessarily the case with the simple PWM technique whose period/duty cycle is independent/ignorant of the AC signal powering the analog device. By anchoring the period/duty cycle to the moment in time when the AC signal is 0V (zero cross), the analog device can be switched on/off in a more optimal/appropriate/phase aware manner for that particular device.

Let's say that you're a resistance heating element and you do your best work near the positive and negative peaks of your power signal but while your power signal is near zero volts, eh not so much. Using PWM+zero cross, the logic driving you can be mindful of the phase of your power signal and switch you on during the most 'productive' portion of the cycle, while switching you off during the least productive portion. Now you're a happy and efficient heating element, rather than one that's being turned completely off for four seconds at a time and causing coffee beans to put on little parkas.

Hopefully someone will correct me if any of this over-simplified, narrow-perspective description is wrong... I'm just a hobbyist at best with the electronics stuff, my area is primarily software.

Offline Mlee

  • Standard User
  • *****
  • Posts: 1212
  • Roaster-Hottop b Espresso-Gagggia espresso
Re: roaster control widget
« Reply #289 on: November 16, 2009, 05:30:01 PM »
Me thinks North Korea has a job opening for you.....
Prov 3:5-6-Trust in the Lord with all your heart
                Lean not on your own understanding
                Acknowledge Him in all your ways
                And He will make your paths straight

yorel23

  • Guest
Re: roaster control widget
« Reply #290 on: November 16, 2009, 05:34:49 PM »
 ??? ???  I put it through GOOGLE TRANSLATOR and it couldn't make heads or tails of it either.  ;D

milowebailey

  • Guest
Re: roaster control widget
« Reply #291 on: November 16, 2009, 07:26:05 PM »
??? ???  I put it through GOOGLE TRANSLATOR and it couldn't make heads or tails of it either.  ;D
Through the milowebailey translator it says that if you can syncronize on the AC current you can control the speed of the fan by turning off part of the AC and limiting the power to the fan instead of just turning it on and off really fast.

JohnR

thanks for the above posts!!   I'd like to see photos of your setup!

baylanger

  • Guest
Re: roaster control widget
« Reply #292 on: January 02, 2010, 06:29:35 PM »
NOTE: I moded my message to reflect reality from my readings ... it's now alsmost 6AM and didn't sleep yet!

I went on the sourceforge and didn't find much information to interface a Hottop with this board as well as schematic.  Note that I have a B model and I read about the controlling of the heating element different than the older models.

Are you soon planning to release notes on your kit?

I'd love to contribute to this project, I've done programmation for many years so I might be able to help, I've done electronics but not enough to safely create the PCB and interconnect everything together.  I'm missing knowledge to do everything myself without perhaps frying a few things.  Now that I've been doing C programmation since years, I guess I can somehow help.

All I'm asking to jump in this project is the schematic + how to use/implement the latest source code.  From there, I'll order the parts and jump in the code!

If you don't mind, PM me your number and I'll call you (or I'll PM your mine).  After all the readings that I did, I'm positive with a short conversation I can get a clear picture of what I need to jumpstart!
« Last Edit: March 13, 2010, 02:33:50 PM by baylanger »

milowebailey

  • Guest
Re: roaster control widget
« Reply #293 on: January 06, 2010, 08:56:03 AM »
Good news folks!

The milowidget is alive... very alive.

baylanger (fellow GCBCer) is a software programmer.... we are planning the completion of the milowidget.  He's going to work the software and I'm going to work the hardware.

first version will be for the Hottop

2nd for either the sonofresco and/or Ambex

3rd SC/TO an other popcorn poppers (I have to get some to toy with).

stay tuned!!

yorel23

  • Guest
Re: roaster control widget
« Reply #294 on: January 06, 2010, 08:58:42 AM »
Good news folks!

The milowidget is alive... very alive.

baylanger (fellow GCBCer) is a software programmer.... we are planning the completion of the milowidget.  He's going to work the software and I'm going to work the hardware.

first version will be for the Hottop

2nd for either the sonofresco and/or Ambex

3rd SC/TO an other popcorn poppers (I have to get some to toy with).

stay tuned!!

First in line for Sonofresco beta testing!!  :icon_salut:

Offline YasBean

  • Standard User
  • ****
  • Posts: 539
  • Lever that sucker!
Re: roaster control widget
« Reply #295 on: January 06, 2010, 09:14:36 AM »
Good news folks!

The milowidget is alive... very alive.

baylanger (fellow GCBCer) is a software programmer.... we are planning the completion of the milowidget.  He's going to work the software and I'm going to work the hardware.

first version will be for the Hottop

2nd for either the sonofresco and/or Ambex

3rd SC/TO an other popcorn poppers (I have to get some to toy with).

stay tuned!!
I am very excited!  The more I use my HT-P, the less confident I am in my profile programming.  It seems to be hit and miss.  The ability to "design" and "control" the profile is enticing!
Londinium L1, Bullet R1, Compak E8, VBM DB, Vario, Hario Vac

yankeeNH

  • Guest
Re: roaster control widget
« Reply #296 on: January 06, 2010, 11:02:42 AM »
...and the idea of having control over the Sonofresco would finallly vault it into the realm of a "legit" roaster. On my wish list: an "on/off" ability to use the pre-programmed profile/roast levels or bypass them. Of course If given a choice any control is better than none...

Tex

  • Guest
Re: roaster control widget
« Reply #297 on: January 06, 2010, 11:42:18 AM »
...and the idea of having control over the Sonofresco would finallly vault it into the realm of a "legit" roaster. On my wish list: an "on/off" ability to use the pre-programmed profile/roast levels or bypass them. Of course If given a choice any control is better than none...

That's my dream too - a Sono with true roast profiling. But I'm not convinced it's doable with a large gas heater & a small roast chamber - too much up & down range in temps to permit small variations in temp profiles.

yorel23

  • Guest
Re: roaster control widget
« Reply #298 on: January 06, 2010, 12:20:06 PM »

That's my dream too - a Sono with true roast profiling. But I'm not convinced it's doable with a large gas heater & a small roast chamber - too much up & down range in temps to permit small variations in temp profiles.

I don't know.  Sonofresco already has a reasonably controlled ramp.  I don't see there being a problem with slowing or accelerating that ramp at certain temperature ranges.  In my mind it seems that it's just a matter of when to turn that heater on and off.  Maybe I'm making it simpler than it really is.

baylanger

  • Guest
Re: roaster control widget
« Reply #299 on: January 10, 2010, 06:27:32 PM »
Good news folks!

The milowidget is alive... very alive.

baylanger (fellow GCBCer) is a software programmer.... we are planning the completion of the milowidget.  He's going to work the software and I'm going to work the hardware.

first version will be for the Hottop

2nd for either the sonofresco and/or Ambex

3rd SC/TO an other popcorn poppers (I have to get some to toy with).

stay tuned!!

Just like it says below my username on the left...  I'll make sure the program stops roasting "just before the third crack".  >:D
« Last Edit: March 13, 2010, 02:34:38 PM by baylanger »