Process Controller for home built roaster

[sea330]

I am looking for a good Process controller to install on my home made coffee roaster.I need at least the ability to setup 10 roast profiles and be able sample two temperature control zone so I can get a good average of the roast chamber. I have built my own enclosure and installed two 1400 watt convection oven fans a 500 watt heat lamp to stabilize the dead band. I also installed a 50 rpm variable speed dc motor for the drum.I was kinda excited when temp ramped up to almost 500 deg in about 5 min.Now the fun part is to get it to control. I built the drum out of heavy gauge perforated aluminum thinking it would work better for heat transfer.also I found a air conditioning unit that came out of a electrical cabinet that seems to cool the roast chamber down fast.I probably have a little over $200.00 into it so far.My background is I retired as a Process control Technician with 30 years in this field of process control systems. I have a lot of experience with Honeywell and Foxbro controllers but there to expensive to do what I want to do, also I need RS232 or RS422 for data collection and configuration to use with my software. I am thinking I can do around 2 to 3 lbs with this setup.Any help with finding the controller would be greatly appreciated. I know there is a ton of expertise out there that can make a recommendation.When I get threw I will configure a build out list if anybody else would like to build one.Thanks in advance for any advice Rich

[peter]

Sorry I can't help with what you're asking. 

But I have one question about the a/c; do you want to do back-to-back roasts, and why cool off the roast chamber instead of cooling the beans once they're out of the drum.

I'm also wondering about the aluminum drum, since every drum I've seen or heard of is steel.

[Tex]

If you find all that you'll be the king!

[mp]

If you find all that you'll be the king!

What ... you thinking of giving up your title?

 

[milowebailey]

There are several groups working on controllers for home roasters.

Look here at the Milowidget and homeroasters.org is working on one too.  Both use the Arduino and a custom sheild (which is an interface to the arduino).  It has a serial output that you could program for RS232. 

you might have to work on an interface board for your roaster but it can be done without too much work.

[oldgearhead]

My first choice would be a PLC. Why? You can use it to automate everything...

[sea330]

I can do ladder logic with ease,c++ or advanced basic a little tough for me. I have been looking for Allen Bradly or any PLC with at least 4 I/O 110 volt ac, a couple 4-20 ma I/O and 4 -1 to 5 volt mv inputs.and at least 4K of non vol Mem.It has been about 9 years into retirement so I don't know whats out there for micros but if I could find a used slick 100 or 500 should do the trick and stay in my budget.I have everything put together to start running things manual so I can get a feel for temp ramps, soak stability and cool down time in my insulated enclosure.I have a dc variable speed drive for the drum rotation. I don.t know what the best rotation speed yet I am thinking any were from 20 to 50 RPMs.Thank for your input and sorry for rambling on about my project.You folks turned me on to the SC/TO method 2 years ago and it has worked great the only problem I have had is getting repeatable profiles I hope this will help and if it don,t I am having a blast Rich 

[rasqual]

Rotation speed: depends on whether you're concerned with conductive or convective heating (though, personally, I think the latter term is a bit of a misnomer in coffee roasting. Slower for conductive, faster for convective. With drums around radius 6", 60 RMP is nice for convective. I'd halve that for non-perf drums where conduction is the principal heating method (and where, I'll understand, drum temp does not surpass MET).

Personally, I shoot for a bean trajectory that crosses from up- to down-side of the drum just over centerline, for convective.

[oldgearhead]

..Like Rick, I'm a retired Electrical Controls Engineer, and it might be fun to hook up an SLC505 to
something to cook coffee. However, for now, I prefer 100% manual control with the roast being controlled
by an accurate bean mass temperature probe.

Yes, the roasting heat can be controlled using PID math, but BMT does not always follow the roasting heat. The task would be to come up with a 'statement' to tie BMT to 'cooking temperature'. I think 4
temperatures would be required:
__Environment
__Inlet temperature
__Outlet temperature
__Bean Mass Temperature