Real-time roasting by weight loss - Sweet Maria's prototype popper [VIDEO]

Discuss roast levels and profiles for espresso, equipment for roasting coffee.
User avatar
baldheadracing
Team HB
Posts: 6170
Joined: 9 years ago

#1: Post by baldheadracing »

The popper is light enough to put it on a scale while roasting. I'm sort-of used to roasting by colour with iRoast2, but I might try this for decaf. 4:01 long
-"Good quality brings happiness as you use it" - Nobuho Miya, Kamasada

User avatar
yakster
Supporter ♡
Posts: 7263
Joined: 15 years ago

#2: Post by yakster »

That's pretty cool. I was actually talking to one of the guys from Acaia during the original Pearl Kickstarter at a Cafe in Santa Clara about uses for scales and brought up the idea of tracking weight loss during roasting. It would be cool to have built-in load cells in roasters to get something real-time, plus a real-time color meter, etc.
-Chris

LMWDP # 272

Milligan
Supporter ❤
Posts: 1446
Joined: 2 years ago

#3: Post by Milligan »

Interesting. Thanks for the video. I would have thought the scale reading would bounce around more due to the vibration. A scale that averages the readings over 1-2s before displaying could be beneficial.

Also! Adding a load cell under my cormorant and plotting that on Artisan as another parameter sounds like a really cool project.

mikelipino
Posts: 256
Joined: 3 years ago

#4: Post by mikelipino »

Interesting approach! I'd also be concerned about noise, and you'd need a pretty slow scale response to smooth it out (smoothing here is not always good, as it's programmatic and an estimate). I'd also be concerned about buoyancy effects, as the weight and volume of the roaster is so much more than the expected weight loss through evaporation. The roaster may act as a fairly sealed vessel and buoy up (think a hot air balloon) and might be a significant chunk of the expected 10g loss

Milligan
Supporter ❤
Posts: 1446
Joined: 2 years ago

#5: Post by Milligan replying to mikelipino »

Thrust from the moving air exiting could be a potential vector for error especially when we are talking about fractions of a gram. Density of the air affects flow and thrust, hotter air as the roast proceeds is lower density. Changing heat level and air flow control could have an effect. How much? Not sure, but something to consider.

randytsuch
Posts: 483
Joined: 15 years ago

#6: Post by randytsuch »

First, let me admit I didn't watch the video.

But I think the recent posts raise the question about how the real time results compare to before/after roast weight measurements?

And now I'm wondering if I could make something for my SCTO that would work here?

I have seen stuff online about guys building their own scales, doesn't seem that hard.

Randy

User avatar
yakster
Supporter ♡
Posts: 7263
Joined: 15 years ago

#7: Post by yakster »

Insert old Sivitz chart here.



Insert old thread reference here.

Weight Loss During Roasting
-Chris

LMWDP # 272

mikelipino
Posts: 256
Joined: 3 years ago

#8: Post by mikelipino »

Right, as Chris says, if it's doable then it would be possible to roast by weight loss as another potential variable to check (weighing afterwards would mean that you couldn't decide and adjust mid-roast).

On larger machines, this might not be too feasible too. E.g. the Cormorant is 14kg, so you'd need a scale that could handle that weight and still be sensitive to 10g, or 0.07%. That's tough to find!

User avatar
baldheadracing (original poster)
Team HB
Posts: 6170
Joined: 9 years ago

#9: Post by baldheadracing (original poster) »

I couldn't see a load cell working on my several-hundred-pound 1kg :), or even something like a Cormorant or Hottop.

I do measure weight loss post-roast as a measure of consistency between roasts, along with colour. However, at my level of lack of roasting skills, I find it best to do, say, three roasts of the same green, and then combine all three roasts for drinking.

On something like a small air roaster - popper, Freshroast, etc. - I think it might help. I really struggle being consistent with decaf. The decaf I have now looks the same to me from just before first crack to blowing bits off beans at the end of second - and I can't hear the cracks with the iRoast2.
-"Good quality brings happiness as you use it" - Nobuho Miya, Kamasada

Milligan
Supporter ❤
Posts: 1446
Joined: 2 years ago

#10: Post by Milligan »

You aren't alone with decaf. I roast by profile alone on decaf. Color doesn't help and I get random pops all over the place. A real time metric like weight loss would be nice.

As for scale not being accurate enough at higher weights. One could use a load cell under each foot. That would cut the weight that each cell sees in a quarter. Could make it easier to source parts anyway.