Nice data Jim. That shows some of the roasting character I have observed about the poppery roaster. A couple observations: The relation between MET and BT, and how, beyond a certain MET-BT offset, you do not get a faster rate of bean temp rise. and how the BT rate of rise is directly related to the difference from MET. (....obviously) This is why I choose to use a heater control setpoint linked to BT; MET = BT + offset. This gives a steady rate of rise for a given offset.
I found that pushing the heater harder beyond a MET-BT offset about 100F did not increase the BT rate of rise. Once browning begins (above 300F BT) these 100f+ offsets will tend to produce uneven roasts, and I believe are detrimental to roast quality. Somewhere between here and above stalling is the 'reasonable range of ramp rate' for my roasts. Note also how less heat (lower MET-BT margin) is required to finish the roast than is needed to get to first crack.
Other thing to note on your second graph is at the end where BT meets MET. ....that was my first clue that I wasn't reading the 'real' MET with my probe position (similiar to yours)
Well done, this tells alot about the roaster character.
I have read this also. I trust this is probably correct, considering that Jim is the source, but I am not sure I understand why yet. Since my bean probe is sleeved, and has a 1/8" diameter, there will be some "stem" effect. That's probably a good thing in this situation, since it results in the t/c reporting sort of an average temperature over the lower 1" or so of the probe.
dana_leighton Said:
What are you using to datalog the temperatures? Does the software from Delta for the PID do that? What device are you using as a transducer for the bean mass temperature? I didn't want to spend much more money, but dammit, Jim, that dual probe plot would sure be handy for profiling...
The PID setpoint trace is not logged data. I just plotted the times and temperatures when I made manual changes to the setpoints.
I don't think the Delta software will support capturing temperature data (but I'll check to confirm as soon as I have a chance). If capturing MET's is important, then probably a good idea to install a second t/c.
(EDIT: the above is incorrect with regard to data capture by the Delta DTB controller software. Turns out the software captures setpoint, measured temperature, and time stamp every second and writes the data to a text file, which can be read by Excel.)
I captured the bean temperature data on a Fluke 54 II digital thermometer.
A couple observations: The relation between MET and BT, and how, beyond a certain MET-BT offset, you do not get a faster rate of bean temp rise. and how the BT rate of rise is directly related to the difference from MET.
Some sort of annoying differential equation, no doubt :-)
Frost Said:
Other thing to note on your second graph is at the end where BT meets MET. ....that was my first clue that I wasn't reading the 'real' MET with my probe position (similiar to yours)
Yeah, I noticed that too (and was a little surprised). I chalked it up to exothermic reaction in the beans. But the simple, and perhaps more correct explanation would be the failure of the sensor to capture the true MET. Better start calling it the AMET (almost maximum ....).
I use a sheathed BT probe for durability. It seems the probe mass slows the response quite a bit and I have found it fairly insensitive to position in the beans (within reason..., 400-405F is consistent for FC)
Jim, sorry I can't help on the differential equation as you can see I'm using addition and subtraction ;^) My test was about exploring the limits for ramp rates that the beans could handle. A range where the system heat transfer was stable and even from air to bean (and outside to inside the bean). And what happens when pushing the upper limit. I was somewhat surprised that the beans seemed to refuse to take on additional heat after a point.... (But really wasn't interested in wasting more beans to explore it!) The roasting system went unstable.
....sorry if I'm spinning off topic here on the PID, but it is about roast temperature control. ....I was hoping someone might put together a controller to integrate MET and BT probes to design Linked Temperature Profiles, so I don't have to do the math in my head and turn the variac periodically... ;^)
On the exothermic response, I have noticed it is more pronounced when I have lowered the fan speed. This makes sense. Since relocating the MET probe, I haven't had the BT meet MET temp.
That would be the advantage, I think, of being able to datalog MET and BT probes for a particular PID profile. The PID profile could be adjusted to try to bring the BT in line with the desired roast profile curve, right?
Of course, auto-tuning the PID with MET and BT both considered? well, that may be sing too much, but I am not a process control engineer. Where's Jim Schulman when you need him? :)
Dana Leighton - Espresso hack and CoffeeGeek moderator
I would put the PID in charge of MET control. (this should tune up fairly straightforward) and put some frontend software to monitor bean temp and adjust the MET setpoint accordingly. Drying phase, ramp, and finish would have separate (and could be fairly simple) algorithms to control MET setpoint based on how the beans are responding.
There are a couple places where bean behavior still needs to be verified: finishing the drying phase and how the first crack develops. ..... but these things are fairly predictable (using temperature alone) once a bean has been dialed in.
The electronics in the panel draw a pretty small amount of current, but adding the transformer might make it overload the circuit. I did not think about that. Did you have problems with that, Ron?
No. I originally had 1 transformer and had no problems, but when I added a 2nd transformer I decided not to overload 1 circuit so I got a longer extension cord for the controller and the fan.
I have received most of the components. The Delta PID is awesome. The cheap temperature meter I got from eBay jumps +/- 1°F every half second or so. I emailed the supplier, and they are sending a replacement IC to reduce the sampling frequency which they say will reduce the fluctuations. We'll see. The PID is much more stable -- I presume it uses some kind of smoothing/weighting algorithm...
The other cheap components I got from eBay were the two volt meters. One worked fine, the other not. It's going back for a replacement. They also did not come with documentation, but the supplier emailed me the info I needed to wire them.
Lesson: It may benefit you to buy higher-cost components from companies that have a good reputation and provide lots of documentation. I could have picked up a basic used Watlow controller for not much more than the cheaper temperature meter I got.
I bought good-quality thermocouple jacks and connecters from Omega for the back of the enclosure, and 25 feet of thermocouple wire from an eBay seller (only $17!). That way the Poppery will be a self-contained unit, with leads for the heater, fan, and three thermocouples that will all plug into the back of the control box. There will also be a plug in the back of the box to plug my bean cooling fan into. That way, one plug from the wall to the box, and everything else plugs into that.
I had trouble finding a project box that would accommodate the front panel. So I had to order an aluminum enclosure from a Ham radio operator in Pennsylvania who has a side business building antennas and enclosures. Bonus: hand-made in the USA. But it's $47. Yikes. It has a great 30° slope on the front where the controls will be mounted. I also plan to put a handle on it so it can easily be transported from my storage room to my porch where the roasting happens.
Speaking of cost, I have not totaled up the cost of the components yet, but I believe I have exceeded $200 (and quite a bit of it was shipping costs), which is more than I wanted to spend. Having a strict budget at the beginning would have helped me not keep adding capability and cost. But I don't regret the cost or design -- It's been a fun project so far.
I bought a variable speed Dremel from an eBay seller, which should arrive tomorrow, and I should have the enclosure and the rest of the components next week, so then the construction will begin.
Dana Leighton - Espresso hack and CoffeeGeek moderator
I tried a fan control today. It wasn't the Lutron and wasn't totally variable, (it had 3 steps) The article from fancollectors is concerned about low speed control of the fan motor and that's not needed in a coffee roaster, all of my use of the fan is at or near top speed. I need fine control of the fan speed at the top end not the bottom. I don't have an air conditioner on the same circuit but there are air conditioners in used in my building and I've had no problems. Make sure the motor controller is continuously variable, like I said, you need control at the top end.
Posted Sat May 30, 2009, 5:45pm Subject: Re: (Another) Poppery PID controller project
ronnie_b Said:
I tried a fan control today. It wasn't the Lutron and wasn't totally variable, (it had 3 steps) ... Make sure the motor controller is continuously variable, like I said, you need control at the top end.
Good point Ron. I did search for one that was continuously variable, and it's easy to get mixed up with the 3-position rotary controls. I tried hooking it up to the Poppery fan as a mock-up and the variable control works great. :)
Dana Leighton - Espresso hack and CoffeeGeek moderator
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 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.
= No New Posts since last visit 