I need to buy or build a PID controller to run a roaster project I started a couple year ago.

I started a project to build me a fully automatic roaster and have decided to resurrect this project. I worked on it for several months and a lot of the individual pieces I have done but when it came time to start putting all the individual pieces together and making the work, I got tied up with other things and never got back to it.

Actually never decided if I wanted to go with a drum or fluid bed roaster. I've heard pro's and cons of both. Actually considered building both and make the controls interchangeable.

I used an electronic gas valve from a furnace to turn gas on and off and to light the flame.

I have a modified manual gas valve with a high torque, heavy duty 5 volt servo to modulate the gas to the burner for burner control. The burners are both my own design and will put a one s**t load of heat. I built a SS heat burner chamber to inject the flame source directly into the air stream for fluid be and another burner to heat a SS drum if I go drum roaster.

I built a gate valve to control air flow by also using a 5 volt servo. This can be used with either one.

I built a heating element from a cloths drier heater to heat the air going through the drum if I make it a drum type or if I decide to go fluid bed, it will be part/all of the heat source for the fluid bed. Either way, I'm thinking if I go with gas burner for main heat and electric to supplement it, the electric will be a lot quicker response time than the gas for maintaining precise heat control.

I will plan to have thermocouples in the input heat side, the output air, and in the bean mass. I even considered using laser for infrared reading on actual drum temp if I go drum.

I already have the metals and air flow to build either one

What I don't have and need help with is the interfacing, controlling and data logging the thermocouples and servos.

I have an extensive background and a degree in electronics, even though I'm a little rusty with it now, and have anything and everything I would need to design, and build a controller, even the circuit board if need be, but hoping there might be a source for just buying one.
What I don't have is a clue just what I need to build since I've never worked with PID controllers before.
I know I can buy individual PID's that will control each function I want to control, but I would like it to be and integrated system and that would not do anything for data logging.

Getting my dream machine PID integrated and a method of data logging is where I'm needing help without having to spend the next year relearning a lot of my electronics and computer programming.

Artisan should do what you need easily with either a TC4 ( which can control the roast ) or Phidget ( which only logs ).
Just search on those names and you will find heaps of information on this forum.

John

It's hard for me to imagine you not wanting to use a programmable micro-controller. Sounds like you want an off-the-shelf one, although many have had success with something like an arduino, some solid-state relays and writing a program.

I have a six channel micro controller that I can run off the USB port on my laptop. What I'm hoping for is one system that will integrate the servo's, (has built in micro controller) temp sensing and motor speed controls and do data logging all into one system.

I ordered the TC4 Kit, the LCD Adapter, and the ZCD. Since I don't know squat about PID's, I figure this will get me heading somewhere in the direction I need to go. The LCD Adapter is probably $60 for really nothing because I figure the laptop should give me everything it will and then some, figured I would get it just in case.

If this doesn't work, I guess I will try the TC4C-Built unit. I didn't go with that to start with because I don't have enough understanding at the moment to tell him how I would need it programmed. Hopefully, with what I've ordered and have, when I start getting it all together and trying to get it working, I will have a better understanding of what I really need. But for now, I've gotta start somewhere and learn what the hell I'm doing with a PID.
The fun part, I haven't done any kind of programming since the mid 80's and then it was C and C+, and long forgot that.

Let me also say thanks to JMC for pointing me in direction of the TC4.

I think the simplest solution would be a Fuji PXG4 to modulate your gas valve via PID, and log to Artisan. The tricky part is finding a driver for the valve, or you can just buy a valve with it's own driver like Clippard valves. There are a couple of documented projects like that on homeroasters.org, one for a fluid bed and one for a drum.

Making the servo's do what I want them to do is no problem. I was operating it with my bench tester and then my laptop back when I first built the valve. With the drivers and program for the micro controller I have, I can manually change them and program them how I want them to operate. Servo's and controlling them is about the only part of this process I do know, I've been messing with those for years with my RC stuff and even have a bench tester I built for them. That's why I initially designed my gas valve to be operated by one. However me telling it what I want it to do or having it preprogramed to do what I think it needs to do is not what I want. That would be fine I wanted to be the one doing all the knob twisting or button pushing. I want it so I can program in a roast profile and the program makes the servo do what it needs to do to keep the roast on track based on the bean temp at the different stages I program in. I want set a roast curve and then have the bean temp follow that curve within a few degrees.
Basically, a white bean temp line and a red programmed line should be a pink line when done. I know that is probably not achievable, but you get the picture, they should be very close to each other. To do this, there are several other functions that are going to need controlling and it looks like the TC4 will do those. It's cheap enough, I'm not out much if it doesn't work out and I will get one hellava education out of what I want and need to do trying to get it to work. Then again, when it all said and done, I may end up having to do more hands on than I'm planning on.

BenKeith wrote:If this doesn't work, I guess I will try the TC4C-Built unit. I didn't go with that to start with because I don't have enough understanding at the moment to tell him how I would need it programmed. Hopefully, with what I've ordered and have, when I start getting it all together and trying to get it working, I will have a better understanding of what I really need. But for now, I've gotta start somewhere and learn what the hell I'm doing with a PID.

Based on a recent message from Jim Galt at homeroasters.org, you won't need to do any PID programming:
http://www.homeroasters.org/php/forum/v ... post_50250

It sounds like PID functionality will be available this summer (after some testing).