Roaster Automation - Drum Roaster - Page 4

#31: Post by **Brewzologist** » June 8th, 2023, 2:27 pm

LBIespresso wrote:I use DeltaSpan=10, Smoothing=0, Sampling=2 All other smoothing settings and check boxes turned off.

I know I put time and thought into making those choices but have since forgotten my logic and am always looking to learn from others. So I went back to one of my roasts from the other day and changed Delta Span to 6 just to see what it looked like and remembered that the line looked a bit more jumpy and that made me jumpy and quicker to make unnecessary gas adjustments. So I think I'm happy with my settings but point all of this out to either a) learn something from your explanation of why you use what you do (if you would be so kind to say). or b) point out that there are more than one way to skin a cat....who skins cats anyway??

I use Sampling=1 because Phidgets support it and I prefer as frequent a temp update as I can get. I agree DS=6 is more jumpy, but I did it again to minimize lag at the expense of jumpiness. I guess I've gotten used to it so I don't get jumpy myself anymore. So yeah, there is more than one way to skin a cat.

blondica73 wrote: Do you sample the TMP1200 RTD sync or async?

I use async. Why? Because Marko made that the default when I setup my Phidgets in Artisan!

June 8th, 2023, 8:47 pm

I received the Phidgets kit today and installed it. I set the sampling rate to 0.5s to see if Artisan updated temperatures and had no issues with late data. I'll run the roaster automation experiment in the next few weeks, I roasted enough coffee to drink for two weeks and I don't want to waste it.

Thank you, Brewzologist, for your help selecting the Phidgets RTD kit; it is much better than the stock BC roaster probes/kit. BTW, I connected my laptop to the stock configuration, and I still see a 3 seconds delay in data logging. However, once I switch to the Phidgets, it's like a different machine, with no more delay.

I'll manually roast some coffee next week to get a feel for the new temperature probes and sampling rate.

#33: Post by **Brewzologist** » June 8th, 2023, 9:55 pm

You're very welcome. Glad you're already seeing benefits. To me, switching to RTD's felt like going from a truck to a sports car when driving a roast.

June 8th, 2023, 9:57 pm

Brewzologist wrote:To me, switching to RTD's felt like going from a truck to a sports car when driving a roast.

June 9th, 2023, 10:18 am

I cleaned up the installation of the Phidgets components; I needed to bend the BT probe as I installed it in the door. The nice thing about the machine was that it already had a few tapped locations that were available and I did not need to drill any new holes. The new probes were installed in new locations and I kept the old probes in. Tested everything one more time and everything seems to be in order.

June 14th, 2023, 10:09 am

I tried the Phidgets setup, and the wild swings are gone, and it makes it much easier to control the roasting process. I set up the sampling rate at 0.5s (because I can

) and delta span at 6, and the signals are so clean compared to my stock setup; I will keep this set up. However, I must adjust some of my roasting parameters around the first crack and drop to align with the new setup.
I also tried an automated roast which came out so-so again. I will need to buy some cheap coffee to continue the experiment; however, I like manually controlling the roast.

June 26th, 2023, 4:36 pm

I tested the RTD probe (blue) against the stock BT probe (red); both are 3 mm probes located at slightly different locations with the RTD installed in the door, as seen in a prior picture. The RTD measures 10 degrees lower than the stock BT probe. Could it be the probe location or the probe itself? Which one is the correct BT reading?

#38: Post by **Capuchin Monk** » June 26th, 2023, 6:05 pm

blondica73 wrote:Which one is the correct BT reading?

If you have another thermometer, measure the temp at your fingertip, then reach inside the roaster when it's at room temp and Artisan is on, and touch the tip of each BT probe for about 10 seconds. Then see which one shows the closest to your fingertip temp. If both are identical, then it may be the location of probes.

#39: Post by **Brewzologist** » June 26th, 2023, 8:43 pm

If one probe is lower than the other in the drum it could read lower for sure.

Also, depth of insertion into the drum can affect things. I try to keep the tip of all probes at least a couple centimeters away from any metal surfaces to avoid skewing readings.

Also, did you try running the roaster empty to see if the two BT probes match any better/worse when only surrounded by air?

Finally, if the probes can be easily removed you could test them in ice baths and boiling water to see if it's a probe specific issue. Artisan supports an offset for those wanting a particular reading.

June 26th, 2023, 8:56 pm

The probes track fairly close up until the turning point, which is very interesting, then I start to see an offset.