Brewzologist wrote: 3) Your probe temps seem off. Did you test them with freezing/boiling water to see if they need to have calibration offsets in Artisan? Also, which probes are you using (TC, RTD and diameter)? This will help in recommending Artisan settings for DeltaSpan and smoothing.

Yeah, they seem way off. Weird thing is your turning point is pretty high, if we assume your BT is 40°F low (usually 1stC is at 390-395°F).

TP could be high with too small charge. How big is your charge? The Huky does optimal batches of 350 g. If you are roasting 150g or something, your temp probes are meaningless.

Rickpatbrown wrote: TP could be high with too small charge. How big is your charge? The Huky does optimal batches of 350 g. If you are roasting 150g or something, your temp probes are meaningless.

Good point Rick. In one of the profiles it appears it may have been a 225g charge. @Robert; I recommend 350-400g charges on the Huky for best performance. Less than 350g you risk running into the problems Rick stated, and above 400g you may not have enough heat to control the roast.

Thanks for the points, here are some follow up items based on the questions:

• I had purchased some greens in 1lb bags, so I split it into two roasts of 225g each. I assume this would only impact TP, not the overall roast profile?
• My TP when I was running seasoning roasts at full 450g charges was 200F.
• The probes are 3" TDI 3.2mm K-type TCs, all four are the same. BT extends about 2-2.5in into the drum, putting the probe tip in the heart of the bean mass near the faceplate.
• I have the solid drum option if that matters at all.
• My MET is usually around 440-460F at Charge and goes up to around 530F during the height of the Gas 100% setting.
• I did notice the probes seem to be reading odd temps as well and was curious about that. I put all 4 probes plus my kitchen Thermapen (calibrated fast read thermometer) in a closed environment with high heat and can confirm all 5 read accurately and in sync up to around 270F (I stopped at that point). My next roast I may try placing the Thermapen in the trier for a true independent reading to see it is in the same ballpark.
• The roaster is connected with a three-prong cord to earth ground. To eliminate noise as an issue on the curves, I ran all motors and graphed the temp readings with no heat turned on.... flat line for all probes, so I don't believe there is a noise issue.
• I set air to min to pass the lighter test and never adjust it during the roast.
• My burner is a 6"x5" metal-mesh IR burner (no blue flame). I wonder if the IR energy vs blue flame could be a cause at all for the temp diffs and how the roaster/beans/air interact? There is no flame under the metal-mesh, the propane/air mix exits a 1in tube in the burner and fills the entire burner with gas that exits through the mesh where it is lit.
• I have Gas 100% dialed in at 10.4in H20 (6 oz/in2) and 5% at 1.6in H20 for this burner. I could max out more towards 11-12in / H20 if needed. Gas events are marked on the black graphs under the Artisan pics.

From the basic roasting session I attended from Scott Rao, he shared FC is normally around the 375 range and not too worry too much if it is +/- around this range so I am not sure if I need to worry about the 356F FC vs just focusing on the ROR curve and roast time. Thoughts?

I will try a lower charge temp next time for a 225g charge, maybe in the 360 range, to see how that goes. I could also buy coffee next time in multiples of 350g so as not to waste any.

I think the problem with your TP and other temp readings is the 225g charge. You can squeak by at 300g on the Huky, but really should make your minimum charge 350g. Anything less and you run into issues where the BT probe may not cover the bean mass all the time. (The fact your TP was 200F with a 450g seasoning charge points to that being the problem too since a 350-400g charge at 400F typically has a TP around 200F on the Huky).

In short, strongly suggest you start with a 350-400F charge while learning your Huky to make your life much easier. This is the sweet spot for probe coverage and roast control (vs. available heat) on the Huky. Once you get your system down, then feel free to experiment with lower charges to see where the temps start getting wonky and you need to roast by your senses instead.

As far as max gas settings go, I'd still shoot for the ability to hit a max of ~40F/min BT RoR after TP before declining. IMO, if your burner can hit this output on the Huky you can pretty much accomplish any desired roast profile you want.

Given all the automation and effort you put into your rig, I'm also just going to come out and say I recommend you consider upgrading your probes to RTD's at some point. Here's a thread on the subject that includes detailed links to info by Marko from Artisan on why: Huky - Temperature Probe Upgrade - T/C & RTD

You can PM me, Marcel, and Chuck who've all done this to our Huky's. It will make driving a roast on the Huky seem more like a Porsche instead of a Yugo.

Thanks for the response and recommendations. I will do the next round of roasts with 350g charges to get closer to an apples-to-apples comparison with other Huky owners.

I saw the Artisan settings you are using from the link you provided for your RTDs:

Sampling= 1sec
Smoothing= 0
DeltaSpan= 5sec

I will also set that next time to see how my 3.2mm T/Cs respond for comparison sake. Since I just bought the T/Cs, I will work with them for a while but definitely would look at the RTDs if I have to replace one.

You will probably find a LOT of noise if you try my RTD settings with regular TC's. Marko's OnIdle blog linked in that thread gets into the details of why that's the case. Agree that you can incrementally upgrade to RTD's later easily with your rig.

You can play with the 3 variables. Delta Span should be at least 2x the sampling rate. For coffee roasting, 3 sec sampling is fine, but if you don't have noise go faster. Leaving smoothing at 0 is best if you can do it. It's all up to how you see the data and react to it. It's what works for you. There is no single right answer.

And you might want to see how the roast with the 350gram charge goes first before changing anything. If you have a roast with 1 sec sampling you should be able to run Simulator and see it how it would look with 2 or 3 second sampling.

Here is a follow up after another dozen roasts...

Regarding the TC readings and FC temps... I inserted a kitchen thermapen (calibrated) in the trier port when the unit was heating up and the thermapen and BT TC tracked 1:1 up to 420F, so I am pretty confident the TCs are providing accurate readings. The interesting part is, based on the bean, I am getting FC temps of either 350F or 355-356F. I am using a very unique heat source - a rectangular metal mesh infrared heat source that is as large as the drum (top down view). Maybe this has something to do with it? I am not worried too much on this as I am getting the correct roast times as well as decent ROR curves.

Here is a 350g roast of a Guatemala La Esperanza. Black line is BT ROR:



And a 350g roast of a Flores Wawo Muda Jember:



Let me know if you have any thoughts on the curves.

Robert

Both of those look solid
How do they taste?

In the end, probes might be different than everyone else's. So long as they are consistent, you should be good.

You've done a lot of modification, so it makes sense that you. Numbers are diffenet than other Hukys. They are just wayyyy different.