Well, that was fun (and not pretty!) Did my first test roast ever on a gas drum machine on some throw-away beans. I learned a lot to analyze for the next test run. This was 8lbs of beans. The roaster has plenty of BTU. I under estimated its thermal capacity and BTU output and went through drying too fast. The TP was higher and a bit sooner than I wanted. The air flow drifted downward, which was the biggest issue I had. I made minor corrections which had a substantial influence on the ROR. The huge rise and crash was exacerbated by me opening the air as the manual suggested to clear chaff.



Next test batch I plan to keep consistent airflow throughout and not adjust it. I did not realize how much minor changes in airflow impact bean temp. I'll go in at a lower BTU and not adjust down until dry end then make a few downward adjustments in gas roughly 20-30s before 1C. Now that I know roughly where events happen on my BT probes, it should be a little easier to navigate. These ended up medium/dark (14 on roast vision) so I expect them to taste pretty roasty with that little of dev time and being that dark plus the flick.

Any tips are appreciated as always.

I'm shooting more for an 11min 410F drop with 25% DTR on the next one.

Are you planning on 8 pound batch sizes consistently once you're using fresh greens?

Your initial run-thru wasn't too far off, and I wouldn't jump around to too many changes unless you're decided on a batch size and try it with beans with a normal moisture content, then start making adjustments. But I agree, you can hold your airflow steady and see smoother, more predictable results. There's several ways to get chaff off of beans and increasing airflow at the end isn't the best way on a roaster like yours, at light to medium roasts.

turnaround @ +200° seems high. But then again it appears your TC are 3mm.

TomC wrote:Are you planning on 8 pound batch sizes consistently once you're using fresh greens?

Your initial run-thru wasn't too far off, and I wouldn't jump around to too many changes unless you're decided on a batch size and try it with beans with a normal moisture content, then start making adjustments. But I agree, you can hold your airflow steady and see smoother, more predictable results. There's several ways to get chaff off of beans and increasing airflow at the end isn't the best way on a roaster like yours, at light to medium roasts.

I was taking some basic initial targets from Rao's suggestions. He noted to shoot for 50-75% capacity depending on the roaster since most companies overstate their capacity. I went with the higher range of roughly 75%. I'd be pleased to do just over 8lb batches. With loss that would produce 9 12oz bags per batch. A good start for me I think.

I was hoping to make it through a roast without any mechanical issues or software glitches, so I'm very pleased to have something to work from here. I'm going to work on keeping air flow consistent.

Something I've thought about while looking at the graphs is that a lot of graphs I've seen have air flow reported as a damper setting. A discrete mechanical setting. I've noticed that the air pressure changes with temperature without moving the damper. This makes sense because air density changes with temperature. At first I thought I should either keep a consistent actual air pressure (which means making tiny adjustments to the damper as the roast progresses) or just set it with the cigarette lighter test and forget it. I would think for consistency I'd lean toward setting the damper and forgetting about it instead of chasing pressures. Thinking a bit more, it would make sense that a fixed fan speed would have lower pressure as temperature rises. Air density falls dramatically at higher temps. However, I don't believe flow would be noticeable impacted. The fan would still move X amount of CFM, but there would be less mass per CFM (fewer particles.) How this affects a system thermodynamically is the real key.

I'd lean toward coming up with a software correction for my pressure readout using ET or BT, depending on the relationship, to have the pressure reading better align with a flow reading as temperature increases. Something to explore at least.

mkane wrote:turnaround @ +200° seems high. But then again it appears your TC are 3mm.

I believe the relationship is the larger the probe, the higher the turn around temp. I agree the turn around surprised me which caused a fast dry. I warmed the system up for 20mins at charge temp + 40F, lowered to charge + 10F over 5 mins. Cut gas until charge - 40F. Then turned gas to 50% until charge temp was reached, then charged the batch. Ambient temp was around 80F. I think with the 8lb batch I could do a soak.

Milligan wrote:Then turned gas to 50% until charge temp was reached, then charged the batch.

The gas was on when charged? If so, shouldn't it have been off all the way to a minute mark after the charge?

I've read a few theories. Some say to charge at a slightly lower temp and run the gas without change through to maillard while others soak. If I soaked, I wouldn't turn the gas completely off at this point. It isn't super responsive at turning the gas back on in a hurry. I need to get more confident in the timing of when the burners come on after being turned off before I make a plan around that.

Rao's book says the most common way is to go full gas until 300F (on an appropriate batch size.) An alternative way is to do the soak. He notes that he only suggests a soak on small machines (under 12kg) prone to scorching due to thinner single wall drum or overpowered burner or if the machine is hard to cool down between batches due to heavy insulation. Mine aligns with the first condition. The second is to be determined in the future. I've seen Mill City recommend a soak when doing half batches. I figure I'll follow Rao's suggestions to a T then deviate if needed.

I'm thinking I'll run 70-80% gas next time and not change it until drying is complete, roughly 300F. Then step down several times until 1C and again thereafter. Now that I have a rough 1C temp, I'll know when not to mess with gas changes. I've found the dwyer gauge combined with the needle valve are excellent on this machine. I can easily hover the BT at a constant temp and make tiny adjustments.

Tasted the coffee this morning. Wet aroma is very pleasant. Floral caramel. Taste has a nice caramel finish and medium body. It has a papery taste in there upfront. Of course, I never liked these beans to begin with so...

Milligan wrote: It has a papery taste in there upfront.

This papery taste may go away after a few more days.

Overall, your profile looks pretty good. But I don't think your dry is too quick. It looks a little over 4min, which is actually a tiny bit long. Your turning point temp is a bit high. I cannot tell if it is due to gas too high or too high of charge temp. With my 1K roaster I keep gas almost off until 1.5 min and then give it all the burner have. My turning point is about the same time as yours and range between 150 and 180F. It too high, my next roast would be charged a bit lower temp.

The roaster looks amazing BTW!

Milligan wrote:Tasted the coffee this morning. Wet aroma is very pleasant. Floral caramel. Taste has a nice caramel finish and medium body. It has a papery taste in there upfront. Of course, I never liked these beans to begin with so...

Papery taste with seasoning beans. I'll be darned.