Same bean, same settings, same day, different profile

[laurence]

Does anyone have any insight as to why these two roasts behaved so differently?
Everything controllable is the same (bean, weight, profile, same day (both in the evening)).
The green line is the profile of the first roast. The black line is the second roast (which I did as soon as the machine had cooled from the first).
Things stayed pretty similar until about 13 minutes.

[Dimal]

What type of roaster are you using Laurence?

Mal.

[soprecise]

Probe/Thermocouple issue? how on earth did you get to up to 300 deg c? And correct me if I'm wrong, but is the 13 min mark you're referring to actually the 8 min mark>? Its midnight, so it may be tiredness, but this log is tripping me out? Love to hear Andys/others explanation....

[DesigningByCoffee]

Hi Laurence
I'm intrigued - but like the others agree we might need a few more details. If it was a corretto, I'd guess you adjusted the gun up in stead of down
Cheers Matt

[laurence]

Hi all. Thanks for your replies.
Soprecise; yes, 8 minutes! Not so precise on my part.
This was on a Belmore 1600.
This is possibly a no-no, but I got to 300 by manually going straight to 100%. I stayed there until first crack (which was just before 15 mins for the first roast and about 16 minutes for the second), at which point I dropped the power off, and then gradually increased it until the beginning of second crack, where I hit cool.
Cheers

[Barry O'Speedwagon]

Have you taped the temperature probe to the heating element or something?

300 generally means beans on fire.

[laurence]

Hi Barry. No, probe is in the recommended position (and beans were not on fire).

[MrJack]

Does anyone have any insight as to why these two roasts behaved so differently?
Everything controllable is the same (bean, weight, profile, same day (both in the evening)).
The green line is the profile of the first roast. The black line is the second roast (which I did as soon as the machine had cooled from the first).
Things stayed pretty similar until about 13 minutes.

[ATTACH=CONFIG]12752[/ATTACH]

Chaos theory?

Just because everything controllable (rather, manipulatable) appeared to be the same, doesn't mean it actually was (I also note a small difference in starting temperature).

Then there the things which are not obvious to/manipulated by the user (control logic, ambient temp, residual temp of roaster, variability between batches of greens, bumping the thermocouple...).

Then there is the possibility that the measured value is not necessarily reflective of "reality".

That said the spike in the middle looks to me like the heat output changed - the next question wouls be why (which is probably more difficult to answer). I'm not that familiar with the behmor, but it does look as if the applied heat rate is somewhat greater during the roast with the spike.

[Andy]

I agree with the above post, so many little variable changes that might add-up to what you see.

I also agree with comments about the 300C... beans will combust at around 260C so something is well off with your multimeter.

Might be worth calibrating that first?
Drop the probe into a glass of ice and water
Drop the probe into a cup of boiling water
Save the profile and JPG and upload them here for diagnosis.

If you are using a multimeter there is a internal adjustment (I'll try and dig one up and get a picture), if you are using the HeatSnob the temperature should be pretty close and fine adjustments can set at the end of the Roast Monitor software preferences.txt file.

You are also running an older version of Roast Monitor, it won't be the issue but running the current version does mean we are all on the same page.

[moreCoffee]

Hi Barry. No, probe is in the recommended position (and beans were not on fire).

I had similar extreme temp readings when I installed the 100mm stainless steel probe in the recommended position. I put it down to a multi-meter compatibility issue and switched to a bead style probe where the recorded environment temps maxed out around 250C... (one of these)

I've experimented with a number of positions, main ones were:
-Back left (recommended position)
-Exhaust
-Inside the drum (bean mass) via Artman's wiper mod.

My 2 cents on Behmor probe placement:

Back left (recommended position): Mainly provides feedback on environment temperature and the whereabouts of the heating elements. With a bead style probe, you can see the elements cycling on/off on the graph. I found the recorded temperatures unreliable for replicating previous roasts, but useful for planning manual override timing etc. Temperature readings here are wildly influenced by ambient temperatures, breezes, position of the moon - you name it. (see MrJack's post above)

Exhaust: Reading start after the fan/afterburner kicks in, found it to be kind of useful for estimating the rate of rise as you are approaching first crack. However, the rate of rise during and after first crack is not a good proxy for bean temperature (as I have since discovered.) Again, readings here are heavily influenced by external factors and can't be relied upon for the purpose of replicating a roast profile.

Inside the drum (bean mass) via Artman's wiper mod.: Still early days, but the dozen or so roasts I've done so far indicate this is a consistent way to estimate bean mass temperature and rate of rise for the entire roast. I'm hitting first crack at the same temperature every roast - that to me suggests the the probe placement is consistent and somewhat 'insulated' from those factors that affect the other placement options.

Finally I have a way to closely replicate successful roast profiles. Happy days....

Cheers,
Pete

[artman]

I have done several roasts now with the "wiper" bean mass bead probe and the results are very consistent.

The temp curve traces the prior roast almost perfectly. It's far easier to replicate the profile than in the corretto I was using. I use the heat slider to track my manual heat input.

Cheers