Posted Sun May 17, 2009, 8:22pm Subject: Re: (Another) Poppery PID controller project
OK -- After a brief consultation with Jim Schulman, I believe I'll have to delay my project and rethink the temperature control situation vis-a-vis bean probe temperature and inlet temperature. There are a couple of profiling issues I am going to have to think about. The main problem being the fluctuations that will be inherent with bean probe monitoring and the profiling issues with inlet probe monitoring. I'll keep using my router power control until I can get these issues ironed out.
Dana Leighton - Espresso hack and CoffeeGeek moderator
I emailed Jim about this privately, but thought I should have posted it here also:
OK - I am working up my courage to install two thermocouples, one to measure bean mass temperature, and one to measure the temperature of the heated air entering the roast chamber PS: (what does MET mean?)
The bean mass temperature probe is easy to install. For the probe measuring the temperature of the air entering the chamber, is the probe to be placed in the area of the cast aluminum , just above the heater coil, next to the fins (there is a little air space there it could measure from), or is it best to have it inside the roast chamber just above the fins (doesn't it interrupt the circulation of the beans if it's there?)? If the former, it looks like it has to be only a nub, about 1/2inch long (assuming I mount by drilling into the cast aluminum and using a tap to create a threaded mounting). I ordered this thermocouple from Auber.
Dana Leighton - Espresso hack and CoffeeGeek moderator
Hi Dana, been meaning to reply here, but haven't had time.... I've got some experience in what you are working on (but no PID.... just manual variac.) MET is Maximum Environment Temperature. The highest temps the beans see during the roast. Hard to get this temp on some roasters but a popper is at least possible.....
I mounted the MET probe in the space you describe on the heater housing. I mounted a bare probe as the space is so limited, drilling a hole only big enough to thread the probe wire, sealed with High temp RTV and strain relief with aluminum tape so the probe tip does no wander around in there. (the probe tip is between the ceramic disk and the vent holes entering the roast vessel.) This worked out well until I realized the MET data was too low, something must be wrong....
I ran a bunch of tests and various temps and air flow rates, (no beans in the popper) and measured the temperature using the probe tip (and an analog mercury ref thermometer) and found the temperature at various points around the vents was + - almost 25F (about 45F variance). There's several reasons; the coil overlaps, coil spacing, and it appears some parts just get hotter and glow brighter.
....So you can put the probe there and try to 'guestimate' what the 'real' MET is, (and not to worry about blowing hot and 'not so hot' air on your beans through the roast....!) So I put the 'mixing chamber' at the bottom to even out this temp gradient. No doubt though it does effect the nice swirling air flow and does not mix the beans as well (requiring more stirring) Accurate temp control was more important for me, I'll do automatic bean agitation for my next roaster ;^) I left the probe wire through the original hole in the chamber and extended it through one of the vents, secured in the 'mixing chamber' with a bit of bailing wire. (the answer here may be to make a 'cone' in the bottom and force the beans to spout up through the center)
In a separate thread I described how I've been running the heater control. (roasting 'general principles') If I were to install a PID, based on my manual control experience, it would directly control the MET with the PID setpoint/ramp. (....hey, that's all it really has control of anyway!) but the bean temp would be nice to use to guide the MET setpoint (a 'linked temperature' profile) Not to repeat my long winded method, ....it's in the 'general principles' around page 8) EDIT: link "Roast profiles -- general principles?"?Page=8
For the probe measuring the temperature of the air entering the chamber, is the probe to be placed in the area of the cast aluminum , just above the heater coil, next to the fins (there is a little air space there it could measure from), or is it best to have it inside the roast chamber just above the fins (doesn't it interrupt the circulation of the beans if it's there?)? If the former, it looks like it has to be only a nub, about 1/2inch long (assuming I mount by drilling into the cast aluminum and using a tap to create a threaded mounting). I ordered this thermocouple from Auber.
You've probably received my email, but I'll rehash some of it here for others to read and disagree with :-)
I put a probe just on the outside of the chamber, in-betwixt the vanes. It is a type K bare bead probe, epoxied in place. It works.
Then I added a second bare bead probe sticking around 3/8" into the roast chamber. Oops. On each roast, this piece of wire traps a bean and carbonizes it. If I were to repeat this project, I would not install the second probe, i.e., I would use the probe between the vanes only.
Finally, with regard to the probe you ordered: it might have too much mass to effectively measure air temperature. And since it will be threaded into the aluminum, it will be heavily influenced by the temperature of the casting. A bare bead K probe, insulated with glass/glass, is probably a better choice.
with regard to the probe you ordered: it might have too much mass to effectively measure air temperature. And since it will be threaded into the aluminum, it will be heavily influenced by the temperature of the casting. A bare bead K probe, insulated with glass/glass, is probably a better choice.
I really like manual control, a BT and MET TC, a variac with a kill a watt meter and count-up timer. The beans act and I can react. I datalog some roasts and record the voltage readings. I can adjust to the temp. the bean is handling the best as the roast progresses.
Well put. I agree with you. Especially in those periods where the beans go between endothermy and exothermy. Unfortunately (or fortunately), it is summertime and I got the tinkering bug, so I'm going to try it out with PID and computer monitoring and profiling the inlet air temperature to try to match the beans.
Dana Leighton - Espresso hack and CoffeeGeek moderator
I have my poppery II PID'd and the way I did it was I drilled a hole through the plastic body, through the iron roast chamber and inserted my thermocouple in there.
It sits about halfway as high as the beans and it really doesn't effect the beans spinning (i thought it would)
I have spent a lot of time talking to Jim Schulman and he recommends mounting it about a half inch above the beans and going by MET.
Also he recommends autotuning and then modifying the constants:
On detuning: on the Fuji, halve the I and D, double the P range (this halves the amplification). You want the PID sluggish, so it doesn't drive the supply air temperature at the bottom of the chamber up and down. If it's always lagging a little behind the ramping set point, this won't happen. Classic PID control is based on the premise that the quality is based solely on the sensed temperature, in this case that of the beans. But in a roast, quality also depends on the roasting environment, i.e. the air comoing into the roast, not oscillating in temperature. One way to do this is to PID the inflowing temperature and ramp that up, then just monitor the temperature at the top and stop the roast at the desired temperature. I used to do that, but detuning makes the whole process easier and more automatic.
Thanks Ilya. I had a similar email consultation with Jim. With both thermocouple probes, I will be in good shape to try both methods and see which I prefer. Just switch the TC on the PID and it's good to go. At what temperature do you do the Autotune, in order to get the starting P, I, & D values that you then double/half?
Dana Leighton - Espresso hack and CoffeeGeek moderator
Placing the TC above the beans will not give you the MET. The MET is the hottest air the beans are exposed to which will be on the inlet, below the beans. It will give you a usable ET once you get used to it.
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 