EDIT:
Deleted first post as it was stupid, I now realize. First I need to learn how to understand the graphs though...I roast on an SR700 small fluid bed.

Sorry for the double thread going on, I didnt want to mingle up the idea of how to measure BT with the actual roasting help needed.

I am looking for any help I can get, fluid bed experience or not. I am currently forced to measure exhaust temp rather than bean temp. My software does its best job it can to maintain specified temperature settings by alternating low/medium/high cycles. It does so fairly well.

I have alternated between using as few temperature changes as possible, to using a bunch of micro changes. I am leaning towards the micro changes and attempt to gradually increase the air temperature as much as possible, with the idea beating to anticipate where the rates of rose will begin to fall of and increase temps slightly just ahead of those points. Thus the following graphs are based on my exhaust temperature and making changes by adding more smaller increments in the process. I am really tying to figure out, is Exhaust test 3 better than Exhaust test 2, and if yes/no why?

Exhaust Test 2:



Exhaust Test 3: (my probes have all started shorting out, I was just able to salvage the exhaust (bean) probe after starting the roast. I also messed up on the cooling cycle but, the main rise of data is accurate.)



My impression is that with some more refined micro adjustments to the 4.5-8minutes time frame I might create a pretty nice profile to start working/tweaking with? Am I way off?

The bean temp curve is much better in roast 3. You should try to regulate your heat source to increase ET as smoothly as possible, think of it as doing a manual PID. BT will then also follow a smooth curve since the delta ET to BT will stay relatively stable. Graph ΔBT and ΔET, they should both be close to straight lines on a decreasing slope throughout the roast. Problems to watch for and correct are seen as spikes or slope changes on the deltas.

Thanks a lot for your replies bill I was actually trying to get the deltas to show up but haven't looked up how yet. I will work on revising to try to create a smoother et-will only be able to reach a certain level though due to the heating relying on varying hi/med/lo heats by should be able to get a pretty close average