drgary wrote: To follow one of your questions, does ET falling below BT matter when the beans are generating their own heat (exothermic)? As long as they're continuing to generate enough, especially at that phase where they're mellowing out acidity, why would it be a problem? You might offer a hint by telling us what was special about that particular roast. What flavors were enhanced?

The beans going exothermic may be part of the reason why the BT can advance above ET. I wonder if another factor may be that the temperature of the drum itself is hotter than the the exhaust air the ET thermocouple is reading. If so even when ET falls below BT, the temperature of the honkin' heavy metal drum drives the BT forward. If this is happening, knowing the temperature of the drum would help in deciding how long the roast could coast with low gas pressure without risking a stall. I haven't tried to get "under the hood" of the TJ-067 to try to place a probe that gets closer to measuring the drum temperature (would this be Maximum Environmental Temperature, "MET"?). Has anyone else tried to measure MET on a TJ-067? By the way, my ET probe is snaked down the bean chute to the top of the opening to the drum.

None of the graphs in Rao's book show the BT/ET curve flip-flop and I haven't seen it anywhere else, either, besides in postings in this forum. I don't think I can get a roast to Rao's full 25% development time and keep it at a City+ drop temperature unless the curves flip-flop.

Rao's commandments, as best I could implement them, seemed to have worked well for the Panama bean. I had done smaller batches ending at different drop temps and used the one I liked best (adjusted for the inaccuracy of my BT probe on small batches) before doing the one posted above. I hit similar EoD and 1Cs times, but I didn't slow down the ROR of the smaller batch enough to get into the 20-25% development time; it was at about 17%. Both batches were very fruity, sweet and had good acidity. The larger batch with more development time had caramel and other flavors I struggle to describe. I would just call it more complex. I am still experimenting with Rao's commandments.

Dregs

JK wrote:I have been experimenting with leaving the gas off longer.. I charging at 360* for 24oz roasts and leave gas off till a 1:30 to hit my drying times for roasts I want 6min drying time.. This way there is more stored heat.. I always try for a 175* turn.. I read that several places on the Roasts Guild site and Scott Rao also shows 175* turn in his graphs..

I'm not sure if 175* is a good TP temp but its what I use..
Maybe someone knows if it matters and can comment

Charging at a higher temp and turning the gas on later is an interesting experiment. It is consistent with Rao's advice to apply adequate energy at the beginning of the roast. I'll give it a try as well. Hopefully, we can compare notes.

I don't know if there is anything special about the 175F turning point. Isn't turning point temperature primarily a function of the thermocouple you are using?

Dregs

Dregs wrote:None of the graphs in Rao's book show the BT/ET curve flip-flop and I haven't seen it anywhere else, either, besides in postings in this forum. I don't think I can get a roast to Rao's full 25% development time and keep it at a City+ drop temperature unless the curves flip-flop.

Rao's commandments, as best I could implement them, seemed to have worked well for the Panama bean. I had done smaller batches ending at different drop temps and used the one I liked best (adjusted for the inaccuracy of my BT probe on small batches) before doing the one posted above. I hit similar EoD and 1Cs times, but I didn't slow down the ROR of the smaller batch enough to get into the 20-25% development time; it was at about 17%. Both batches were very fruity, sweet and had good acidity. The larger batch with more development time had caramel and other flavors I struggle to describe. I would just call it more complex. I am still experimenting with Rao's commandments.

Jim Schulman is skeptical about Rao's commandments as being anything more than generalizations based on the equipment with which he's familiar. I'm a beginning roaster so can't offer useful personal opinion on this, but I'm less drawn to Rao's book than I might otherwise have been, because I respect Jim's observations for being empirically based. I'm not saying this to be controversial but am more drawn to learning from other criteria than "Rao's rules."

another_jim wrote:What Rao says may hold empirically for the range of profiles on the range of roasting machines with which he is personally familiar. But it would not be for the reasons he gives, since he has these misapprehensions about chemistry and mathematics. This is not a problem if he's just giving instructions on how to roast well on the gear and circumstances he knows; but it does not bode well for applying his adages more generally.

Gary,

I found with higher charge temps I get no cross over.. High heat at beginning
With lower starting temps. or lower gas setting at start my et and bt are close together through out roast I get cross over.. Lower heat at beginning needs more heat through ramp..

Also this could just be my different placement and type of probe..

I'm guessing one gets a feel for one's particular roaster and probe read-outs and what's been tried to get to roasts that taste good. There's a big rabbit hole here and years of trying different approaches. Could you post a profile or two and let us know what probes you use and where you've placed them? Also do you have a solid drum TJ-067? I wonder if any else has a perforated drum.

First a disclaimer - I am on the low and steep portion of the learning curve so I am giving a description, not a recommendation . Sorry for the length...

My lowest loads have been 300g and I have been charging them at 250F or 275F, lately I have been working on 600g loads.

My ET probe is in the lower right screw hole on the bean chute - I have had a few instances where my ET dropped below BT at the very end of the roast, but not as drastically as in Dregs' example.

I have been trying a different startup approach after reading Rao's book and Chris Schooley's post in the October Roast and Learn. Starting with airflow at 100% and then gas on after TP and held at one setting for most of the roast. I reduce the airflow at 3 minutes. Trying to use the timing of gas on and air reductions to control the Dry/Ramp intervals in a predictable manner.

For my first roast of the day, I have been careful to preheat pretty consistently by holding at 50F above my planned charge temp (measured at the Bean Temp probe) for 20 minutes then dropping to 5-10F above and holding there for another 10 minutes with my air at 100%. Then I set my gas to the setting I want for most of the roast - usually around 2.7 - 3 kPa for 600g loads. Then I turn off the gas and charge when the temp drops to my target temp.

I like how this approach helps me manage the pre-heat at the lower charge temperature because I find I can't hold my temp as low as 250F without a lot of airflow and it makes the start sequence a little easier because my air and gas are both preset and my first gas adjustment is just hitting the on button and I'm good to go until I get close to 1Cs. Controlling RoR with air adjustments vs. gas adjustments is a little easier too because I sometimes get strange responses during gas adjustments where pressure drops when I am adjusting it up and it takes a little more fiddling to get it stabilized where I want it whereas an airflow adjustment is simpler to make.

I am still playing with this approach and would love some input on the difference between a high airflow start vs a low airflow start as BDL uses and also the value/detriment of a "stall" between the TP and start of the RoR climb which is what I get if I delay the gas on for very long after TP.

Here's a pic from my last roast today which went almost exactly as I planned (the white dashed line is my target - another bit of BDL's advice The gas was set at 3.0kPa initially and reduced to 2.6 kPa ahead of 1Cs. I have been doing similar roasts at 300g but haven't quite settled on the timing and gas settings yet.

That's very systematic. I like the way you keep some of the variables fixed. Thank you for the detail on how you do this.

No Fan No Gas till you see red delta line on graph..
This is the only profile I have saved at this time..
his was a full 2 lbs and a 3 KPA and I mostly did 3.2-3.5 KPA higher gas roasts.
this was trying to get my drying times longer buy starting gas later..

Gary, try not to make any adjustments to gas and minor adjustments to fan..

I use cheap Procon probes from eBay
They read about 0* in ice water and 212* in boiling water I order about four or five and ask for combined shipping after I place order..
My probe that checked exhaust temps I changed and bent it into a L shape and its now 3" in roasting drum with bare bead exposed..

Beans are only exothermic for a brief phase before they become endothermic again, but they're still a mass that conducts heat energy more readily than that of gas.. The divergence of temps is several things, probe location, total available system heat and overall heat and mass of the system, which is likely the greatest contributor (larger commercial roasters don't generally show this phenomenon). Once the beans are equal-thermic to the environment they're contained in, you begin to see the differences in how each material ( gas vs bean) react to heat input to the system more clearly, meaning the beans are no longer massive heat soaks with their starting temperature, moisture and internal density; their heat capacity shifts significantly.

I imagine the biggest reason why we see this in smaller (< 5 lb) roasters and not larger units is likely due in part to the turbulence of the coffee in larger systems is likely creating a slightly more homogeneous blend of beans and the gases flowing around them (Jim's hornets nest description) where nearly all the heat transfer is via convection, where as smaller roasters the bean mass might spend a greater proportion of time either against the drum or as a more condensed mass, where their higher mass reads differently (higher). Higher drum RPM won't help, but better, larger and more vanes do.

What looks like a stall on one system likely will not translate to another, so the more important matter is understanding how your roaster works and correlating what you see with what you get in the cup and avoiding stalling the roast. I'd pay much more attention to the BT ROR than the ET, as long as I knew how my roaster and coffee reacts to heat/ airflow modulation. A descending ET relative to the BT isn't the end of the world.

There's quite a bit to chew on this thread as a number of good topics are covered that are worthy of separate threads. Rather than quote text all over the place, I'll try to belatedly respond (I've had little time recently to feed this hobby) to the various topics:

First, I've seen this BT/ET cross-over on my roasts as well a few times on the North. This tends to happen when I've applied too much power in the ramp phase (to make up for too high charge temp or too latent power application in the drying phase). I suspect (similar to JK's and Tom's comments) that there is build up heat stored in the capacity in the system (drum, mainly), and falling rate of heat capacity in the beans, so the heat transfer continues coasting to the finish (perhaps from conduction or close radiation from drum to bean). If the coasting is too strong, I'll push up the air flow high to dump the excess head, and since the ET probe gets the full air flow, whereas the beans are clumped together, this will also tend to exaggerate the BT-ET inversion phenomenon.

Regarding the comments about MET, I've held similar thoughts myself. It would be nice to gauge BT against drum temp, but it may not be worth the trouble with the North for three reasons:

• 1. There's enough head capacity in the North, where MET temps wouldn't get high enough to cause roast defects without other parameters getting really out wack (e.g. high-temp TS, short ramps, FC-SC melding )
  2. Installing a MET probe on the North may prove difficult, due to it's countersunk design around the drum;
  3. Finally, I'm not yet convinced that the burners are overbuilt like the rest of the roaster to blow up drum temps under a full charge (opinion on this my change with further testing).

Regarding Rao-Jim controversy, and although I haven't read Rao's book, I think this is perhaps the very case where Rao's rules of thumb may come in handy - with higher thermal-mass, higher-capacity, conventional rotating solid drum roasters. For example, his falling rate delta-BT rule may just reflect the characteristic of dealing with the heat transfer latency of a high thermal mass drum with the falling rate heat capacity of the beans, as described earlier.

Regarding the high-fan, constant temp roasting approach, this sounds like another take on the constant-ventilation/variable heat versus constant-heat/variable-ventilation roasting approach mentioned over the years on the forum. As to what approach is best to use with the North, all I can say at this point is that at least the North gives you a choice with easy-to-control (and automate) variable fan speed control versus a mechanical damper, especially if you switch out the speed potentiometer with and audio (logarithmic) pot.

Finally, Gary, much of what we see here may apply little to your perforated drum setup. Not only are the ventilation characteristics very different, but the heat delivery to the beans is via direct radiation from the burners (versus indirectly from the drum). At least you may not need to turn up the heat as much. Do let us know how your roast techniques progress.