I've given up on getting consistent readings with a TC in the bean mass; so I took some readings of the exhaust air. The TC is mounted in the Freshroast's lower lid, wrapped around one of the bars of the chaff catcher cage, with the probe sticking out towards the middle of the roast chamber diameter. I did this very roughly on two lids, the upper lid still sits properly as the TC wire is thin enough to fit. I've done 10 roasts and have gotten consistent readings with Brazil, Kenya, Indian Pearl Mountain Peaberry, Yauco Selecto, and Mocha beans. The first crack started at 360F and ended at 410F, the first pops of the second start at 440F, and one gets a "rolling second" at 450F. Decaf roasts read about 5F higher at each stage. Altering the input air temperature with my variac affected the readings initially, but they stabilized at close to the same level as the previous settings (takes about 20 seconds). I would guess the readings are some average of 80% to 85% bean temperature, 15% to 20% inflow air, and some constant for the ambient. In any case, the readings correspond fairly well with the standard crack temperatures, and they remained consistent roast to roast. It was much easier controlling the roast profile between the cracks with this temperature reading than by judging the bean color. Is this a peculiarity of my roaster, or the FRs; or is exhaust air a good measuring point for airroasters in general? Anyone willing to try this on theirs? Jim Schulman homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast
Actually I have the little dial thermometer in the beans on my FR and seem to get repeatable results. My readings may be lower because of the bean temp being cooler than the air temp. I variac also, with my fan driven off a 12 volt wall transformer. Also, the consistency of the reading is important, not the actual reading. However, it does seem that miKe McCoffee and me do seem to agree about temperature points. Ben Jim Schulman wrote: <Snip> homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast
Jim: <Snip> or is exhaust air a good measuring point for airroasters in general? That's a dandy question . . . and I think that there's not going to be a simple answer. In the FR there just aren't any beans that are not "in the airstream" .. . . so measuring bean temperature alone (by any method other than IR) is always going to be difficult (if not impossible). It may be that the exhaust air is close enough to equilibrium with the beans to be accurate enough . . . and if the air input temperature and air flow are constant it should be reproducible regardless accuracy and that gives what we're looking for anyway . .. . a consistent way to determine state of roast. For my next round of Rosto experiments I'll look at exhaust temperature too, and see how that tracks with what I see at the beans (in the Rosto the beans at the outside wall of the roast chamber are pretty much out of the airflow, so maybe, just maybe, I'm actually measuring mostly the beans there). There is a potential opportunity for the Z&D in this regard . . . since there is so little airflow to worry about a probe at the top of the bean mass could be assumed to be seeing pretty much all bean . . . which would make it relatively easy for them to incorporate *real* endpoint control (instead of just a timer) in the next iteration of their machine. That, and something to accelerate the "quench", would dramatically improve the Z&D machine. <Snip> easier controlling the roast profile between the cracks with this temperature reading than by judging the bean color. I think one has to acquire quite a body of experience, with roaster, with bean, in general, to master that without temperature. I often cannot hear second crack until it's too late (if I hear it at all) . . . and color is, as you note, "uncertain". I can stop . . . I *always* stop . . . at the first hint of oil, but even that varies from bean to bean and is itself open to interpretation . . . ("sheen"? "spots"? "darn it's getting dark but it still looks dry?") . . . and often leaves me with something other than the roast I set out to produce. It's always a problem with my (unistrumented) HWP. It's sure hard to "profile" that first-to-second transition when I don't know when second is going to happen, or how close the end is. With my Rosto I've come to depend on the thermometer more than anything else, and my roasts have improved considerably for it. One cannot control what one cannot measure . . . Deward homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast
Jim: Finished my gas grill roaster last night and did 3 roasts. The thing has a very high quality temperature gauge in the top of the grilling chamber which is just above the roasting drum-it very accurately measures the boiling point of water. I recorded the time and temps of the three roasts. Now since I have horrible hearing I could not hear the first crack and so kept on roasting the first roast till what I thought was the first crack was actually a screaming second crack. Took the drum out and emptied it at 21.5 minutes and the final temp was 500F-well when I opened the grill top to pull out the drum-smoke and more smoke and the beans were covered with shiny oil. Next roast I stopped at stopped at 17.5 minutes and 480F. Beans looked great and the coffee was very good-had the smooth full flavor I like-beans are el cheapo col supremo-cost $2.50 /#. Batch #3 I stopped at 16.5 minutes and 460F. This one looks like it was just getting into the second crack. Again I am measuring the temperature above the drum and the temp in the drum maybe a bit less-not much though since the drum occupies a signifigant bit of the space inside the grill. Each batch was about 1.5#. Cold tested a 10# batch-beans were agitated very well and should be no problem to roast 10# or maybe more beans at one batch. Also no scorch or burn marks on beans from batch 2 & 3. The first batch was universally over roasted-beautiful black shiny beans and what smoke. Again I am just getting started but also think the below numbers you have seem to agree with what is going on with my roaster. I am going to build another drum-this one built so that I can hopefully insert a remote temperature sensing device in the drum. Will do this by suspending the drum on the axel a few inches inside the roaster and making the hole in the end of the roaster for the axel a bit bigger to permit the insertion of a temperature sensing device. Any suggestions as to what kind of remote temp sensing device. Sources and brands would be helpful. Will also try to develop a means of using a bean thief to pull some beans out during the roasting process-that too appears to be possible. Ah roasting-what fun and am now getting coffee I really like-spresso maybe next.
I would think that measuring exhaust air would be a good indicator of roast progress. If it was compared to input air, a clearer picture of bean temp would be possible. At first, assuming the input temperature is constant, the output air should be quite a bit cooler than input air, since the cool beans are absorbing heat quickly. As their temperature becomes closer to the surrounding air, I would assume that the output air and input air would be very similar. A plot of the difference between the two should give a clear indication of bean temp change. I realize there will be heat loss to the roaster itself and loss through the glass, but that loss would be fairly constant after the parts all heat up. Ed Needhamhttp://www.homeroaster.comed **************************************** **********************************************
My .02 cents: as other people have concluded, IR is going to be the only way to get accurate readings in a rapid air flow, but that doesnt mean readings arent useful. Figuring out a good place to mount it and get consistent readings. like you did, is very useful, even if the numbers don't indicate empirical data about all coffee in all roasters (but your crack temps look accurate anyway). In my Diedrich, i NEVER got "correct" temperature readings for bean temp, and I drilled quite a few holes in that poor thing to reposition the probe. But I did get consistent readings, logged them, and found them very useful for optimizing the results from that roaster (the 12 kilo with IR burners). The Coffee Kinetics roaster uses an RTD probe in the exit air as its only data for turning the gas burner on and off to moderate the heat. It works well because in a sense it accounts for a change in airflow due to a greater charge of coffee in the roaster .... if you overload the roaster with coffee, the initial RTD readings are too low and the roaster shuts off with an error code. Tom <Snip> -- "Great coffee comes from tiny roasters" Sweet Maria's Home Coffee Roasting - Tom & Maria 1455 64th Street Emeryville CA 94608 http://lists.sweetmarias.com/mailman/listinfo/homeroast">http://www.sweetmarias.comhomeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast
It works for me, with my popper. The temp may not be the exact bean temp, and may not agree with someone elses, but it's very consistant. That's what really helps, having another reliable reference to consult. It is important to keep the thermometer in the same place, moving it somewhere else in my own popper results in a different, and unfamiliar reliable indicator of roast temps! It's also helped to fine tune my nose to degree of roast. John <Snip> Protect your PC - get McAfee.com VirusScan Online http://lists.sweetmarias.com/mailman/listinfo/homeroast">http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid963homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast
John: <Snip> Ain't that just the way it is . . . my thermometer is teaching me how to roast without a thermometer . . . Deward homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast
I ended up doing this the other way. My nose taught me what the thermometer means at different readings so I can now have my son-in-law roast by thermometer because he insists that he can't smell the difference. Jim Gundlach On Thursday, January 16, 2003, at 11:28 PM, dewardh wrote: <Snip> homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast
To all: Thanks for all the information. I guess there's lots of right ways to measure temperature consistently (along with all the wrong ways I've discovered). I'm glad I finally got one that works for me. Floyd: I think a thermocouple is fine, although a more expensive RTD won't have to be replaced as often (TC sensors cost about $10, and mine seem to last about a year before the insulation goes. Maybe if they're dedicated to one use, they'll last longer). This link has been posted before, it seems to be a reliable and very cheap K-type TC reader with one probe: <Snip> Perhaps you could attach the TC (its a long wire with the sensing at the end) to an L shaped rod with some wire wrapping, and adjust the position until it's in the bean mass. However, i'm hardly the person to give anyone mechanical instructions. Ed: I tried the inflow/exhaust temperature system. But with switching between TCs, and tweaking the variac, all the while setting up the next roast batch and cooling the last, my brain overloaded. I'll stick to one temperature until I get graceful with it. Then I'll try this extra wrinkle again. Steve Ackman of Two Loons recently started a thread related to this on alt.coffee (just in case you missed it) <Snip> Deward: I'd be interested to hear comparisons between exhaust and bean mass measurements on the Rosto. The latter seems to work well on it, so it would give some idea what sort of discepancies are introduced by measuring at the exhaust. Jim Schulman homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast
Thanks for the info Jim. Dan Nathan sent me info on a TC and it is called Atkins #39658K and his company has had their unit for 7 years-will get details and post it on AC or here or both and get input. Hope u enjoy the party-is it this weekend-having to make that trip alone-oh I drive so much and another 10 hours in my car is more than my mind can take-anyway thanks again.
Jim: <Snip> roast by thermometer because he insists that he can't smell the difference. I guess that's my point . . . that temperature is the most consistent and conveyable measure of state-of-roast. It's the least "subjective", and is relatively invariant across variety of bean and (properly measured) across method of roast. Once temperature is established it is relatively easy to learn the associated "non-instrumental" clues . . . (and how they vary from variety to variety). Over time he may well "learn" which changes in smell to "notice" (that correspond to the temperatures you provide). Deward homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast
<Snip> I have two holes drilled in my Popcorn Pumper, and I used two thermometers as an informal experiment the other day. No data logging, just seat of the pants stuff. My initial indications are that you are exactly correct. I want to try some experiments with no beans, to see if position of the thermometer affects the measured air temp significantly, and to reverse the two thermometer's position to see if they are calibrated equally. But with one thermometer immersed in beans, and the other somewhat above the beans, I got the results you predicted. I have a cheapo thermocouple thermometer on order, so my next stage is to use three at once, to monitor coil temp, bean temp and exhaust temp. homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast