Posted Sat May 16, 2009, 11:19pm Subject: (Another) Poppery PID controller project
I have been using my Poppery for the last 4 years or so, profiling my roasts the hard way. First, I separated the fan and heater curcuit so the fan was on all the time, and heater controlled by the switch in the front. This allows rudimentary control over the ramping speed, and I achieved some quite good roasts this way. The drawback is that you have to continually monitor temperature and learn when to switch on and off the heater to maintain a constant ramp rather than letting the temperature fall off. Using this method, you're basically a human PID controller!
The next step taken several months ago was to route the heater circuit through a router speed control, like the one sold at Harbor Freight. This one allows me to set the heater power and gradually increase it to achieve the ramping -- Much easier and smoother. Still it requires a significant amount of monitoring temperatures and mental arithmetic to keep the ramp stable.
I recently sourced the parts to construct a PID controller for the roaster. It'll allow me to set the ramping and make the process less labor intensive. I don't expect the PID control will result in better beans, but it will be a whole heck of a lot easier to repeat roasts, adjust profile parameters to see what the effects are, etc. I considered buying a Behmor (about $50 more than the parts for this project) which would have the advantage of larger batch sizes, but I wanted the more precise control and customizability of the profiling that the PID will provide. I might someday consider a Hottop programmable unit, but that's another $750, which is out of my price range at the moment.
Here's the parts I picked up to start the project (all sourced from Auber Instruments -- good quality and a good price, and less hassle than trying to source multiple parts from multiple sources and decoding the PID controller parameters on eBay)
Auber SYL-2352P PID controller with 30 ramp/soak steps, fuzzy logic, auto tune, dual set point temp & actual temp indicators $90
Auber HS25 Heat sink w/mounting screws and thermal paste $10
I considered getting a SCR unit instead of an SSR so that the heater output could be variable, but the Auber unit does not appear to output variable voltage. Also, I again had trouble deciphering the various parameters of the affordable SCRs on eBay, so I decided to go with the Auber PID and SSR for now. There may be another project in my future. :)
When I get the parts, I'll have to pick up a project box at Radio Shack that can hold the PID & SSR mounted on the heat sink. I imagine it will need to have decent airflow to help the heat sink do its job and not heat up the PID unit.
Any suggestions and pointers and ideas appreciated. I'll post to this thread as I go along in case someone else might be trying the same thing. I have marginal electrical skills, so I used the homeroasters.org Poppery modification thread as a rough guide so far for my mods - I am grateful to Mike for posting that.
Dana Leighton - Espresso hack and CoffeeGeek moderator
ronnie_b Senior Member Joined: 7 May 2005 Posts: 308 Location: New Jersey Expertise: I love coffee
Grinder: Bodum Antigua Drip: Melitta pour over Roaster: PID'd P1
Posted Sun May 17, 2009, 3:54am Subject: Re: (Another) Poppery PID controller project
Good luck! I use the Auber also, but it isn't the easiest to use. The instructions are a bit dense. This is my setup. "My popper" The PID uses the heater temp for input rather than the bean temp like the homeroasting article, but you can use whichever you want. A Dremel is invaluable in constructing it. Also, the booster transformer really helps with increasing the amount you can roast. I started with one and now use 2 and can roast 240 grams at a time.
Posted Sun May 17, 2009, 8:56am Subject: Re: (Another) Poppery PID controller project
For the popper controller I built awhile back, I used a Delta PID controller like this from factorymation.com (also available here from B&B Electronics). When coupled with a USB to RS-485 serial interface like this or this from B&B Electronics, and this downloadable software, then programming the PID, including ramp/soak profiles, becomes pretty simple. (I can't imagine the hell you'd have to go through to program ramp/soak profiles from the front panel with the user interface I've seen on some PID's).
For this application, the only serious criticism I have of the Delta controller is that the LED display is dim. It is fairly difficult to read if roasting outdoors. Otherwise, I wouldn't hesitate to recommend it for popper-roasters.
Yes, I was thinking about that -- I will be considering the relative advantages of both, and the PID with a computer interface would be definitely a bonus. A cheap old Pentium Windoze laptop to run the software and I'd be stylin'
Dana Leighton - Espresso hack and CoffeeGeek moderator
Posted Sun May 17, 2009, 9:47am Subject: Re: (Another) Poppery PID controller project
While you have the popper apart, it may be worthwhile to go ahead and install a type K sensor where the heated air stream enters the roast chamber. Even if you don't use it for control, at some point you'll want to know your MET. I say this with confidence because putting a PID on a popper indicates the disease is already well-advanced :-)
The problem I had with my popper setup when trying to use bean temp as the control variable was that the PID would drive the heater full blast to get the bean temps moving. As a result, I was seeing MET's over 500F. I worried about overheating the popper since I had bypassed its thermostat.
My PID project was not a complete success on the P1. Those issues were among the problems I never solved completely.
Tuning, i.e. setting the right P, I, and D parameters, for a bean mass probe is a challenge because the thermal properties are changing even within a single roast session. So you can imagine the challenge managing changes from session to session with different beans.
I had a little success by running the autotune process during a sacrificial roast. IIRC, I tuned the system at a point where the beans were nearing first crack, thinking this would represent average conditions.
Tuning for a heater air probe is a little easier because it is relatively less affected by the changes in the beans.
I eventually lost my motivation with the popper because I had so much trouble coming up with ramp/soak steps that would automagically give consistent results (for exactly the reasons you identified).
In hindsight, I would probably have had better success by making periodic MET setpoint changes on the PID, manually. That's what I do now with my Hottop: 3 different ET setpoints (at most) are adequate to manipulate the BT profile fairly predictably. If I ever go back to the popper, that is the strategy I would try first.
Both! :-D The instructions at homeroasters are pretty good. I gather that you're mechanically and electrically inclined so you shouldn't have any problems. We're here if you do.
Once you have the PID hooked up, run a few test roasts to determine the lag your roaster has. The ramps that are required for roasting beans is steeper than the rate of increase that the popper can provide and the difference can vary by about 60F or more. I autotuned it at 400F. As for beans of different densities and humidity, I have 2 profiles programmed, one for regular beans and one for decaf. Decaf beans are much lighter than regular beans and I adjust by reducing the fan speed. All you need to do is to make sure the beans are moving with adequate agitation. There's no way to adjust for humidity because it varies every day. My profiles do include a minute at 100F to have a consistant starting point for the roast and I cool everything down to 100F after the roast.
Symbols: = New Posts since your last visit = No New Posts since last visit = Newest post
Forum Rules: No profanity, illegal acts or personal attacks will be tolerated in these discussion boards. No commercial posting of any nature will be tolerated; only private sales by private individuals, in the "Buy and Sell" forum. No SEO style postings will be tolerated. SEO related posts will result in immediate ban from CoffeeGeek. No cross posting allowed - do not post your topic to more than one forum, nor repost a topic to the same forum. Who Can Read The Forum? Anyone can read posts in these discussion boards. Who Can Post New Topics? Any registered CoffeeGeek member can post new topics. Who Can Post Replies? Any registered CoffeeGeek member can post replies. Can Photos be posted? Anyone can post photos in their new topics or replies. Who can change or delete posts? Any CoffeeGeek member can edit their own posts. Only moderators can delete posts. Probationary Period: If you are a new signup for CoffeeGeek, you cannot promote, endorse, criticise or otherwise post an unsolicited endorsement for any company, product or service in your first five postings.