Just extended a family of curves drawn in the past for various air temperatures in the 232-250 C ballpark, that led to FC in the 9-12 minutes interval.
Drawn the same equation curve, but for 270C, and the result is above. The machine is a popper, of course.

I added a thermocouple to the Popper using SM's guide. It was really easy, you just have to be careful not to drill too low in the chamber and the thermocouple self taps into the housing. It has made roasts much more repeatable in that I can predict DE and 1C against desired times and can now drop at a desired temperature. Definitely worth the effort! And I've read up on how these types of thermocouples might be too noisy to meaningfully do ROR on Artisan, but at least the option is there to play around if I ever get a Phidget.

Espresso roasts have turned out surprisingly good. It's nice to be able to roast my choice of beans to a desired roast level. I love my local roaster, but they have to sell to customers that think espresso = dark.

Noise is not in the sensor used, but in the whole environment.
No particular sensor or measurement interface will calm down such high turbulent environment.
An improvement can be obtained using more sensors placed in different locations, and averaging their reading.
For some sensor types averaging can be done just by connecting them together, without special measurement circuitry.
For example four Pt sensors of same type, if connected in a bridge, do an excellent averaging job, the resulting two wires output behaving as a single sensor of same value.
Or, for thermocouples, even simpler, if the wire is twisted multiple times, starting from the tip, say every one inch make 2-3 twists, to ensure a firm connection, then the resulting output voltage (temperature) will be the average of the voltages in every connection point. Spiral such twisted wire on a thermo-resistant (fiberglass) rod, and you have a much accurate and noise free sensor for such scenario.
I was an industrial automatics engineer 30 years ago, before turning to computers. Those times thermocouples aren't sold in shops or online, but built on purpose, using thermocouple wire sold as 100 meters reel. Even today you can find such wire for those who have special measurement needs.

Especially for the popper roasters case, I gave up measuring anything, after some measurement/simulations/modelling the process.
Instead measuring temperature, all you need is a precise controllable and repetitive power source for the heater.
In such small environment you can equate the power percent to hot air temperature, with a higher precision than measuring the air/beans mix temperature.
Starting with a power level that equate to 150C in the roast chamber at drop, then increase power from there with same allure as the intend profile will led to a very precise and repetitive roast all the time. In my case this is 45 to 70% power range for 1600W full power heater, and 100 grams of greens

Nope. No good. Apart from being hard to dial in a shot, the grounds appear cocoa colored. Still grassy and flat. The beans looked evenly roasted and normal but no way. I guess I'll try a darker roast next.

renatoa wrote: Especially for the popper roasters case, I gave up measuring anything, after some measurement/simulations/modelling the process.
Instead measuring temperature, all you need is a precise controllable and repetitive power source for the heater.
In such small environment you can equate the power percent to hot air temperature, with a higher precision than measuring the air/beans mix temperature.

I wonder what the popper does over the FreshRoast products?. I was surprised to see SM create it. It can only do 100g which gives about 80g out which is pretty small. In these small roasters the ambient temp makes a huge difference, so if you are roasting outside, wind, temp, sun all change the roasting parameters. Also the amount of moisture in the beans will make a large difference; so if you like to use different beans and drying processes, you'll need to experiment quite a bit. Using a FreshRoast 800 with a thick 3rd party tube and a temp sensor under the bean chamber helps, but it is still quite a bit of work trying out different 'algorithms' and then cupping the results, if you are looking to really bring out beans' characteristics. You can see issues users go through on the FreshRoast 800 threads.

Bluenoser wrote:I wonder what the popper does over the FreshRoast products?. ...

It costs $89. An SR540 is $209; the SR800 is $289.

The sub-$100 price point - $99 with four pounds of greens - is very appealing.

Craig is spot on. Cracking under $100 with roast control out-of-box is (at least to my knowledge) new and very appealing.

I've also been very interested in a Fresh Roast, but my internal logic has been what would the added cost get me? Compared to the SR540, I see added fan control, integrated ET, and nominal increase in roast capacity (120g vs 100g). Since the cyclonic air doesn't really push beans out of the chamber, the fan control is mostly another method of regulating heat, and we have more resolution with the Popper analog heat control. Integrated ET is addressed somewhat easily with a thermoprobe and can be done in pairs to get ET/BT (same with the SRs). Then there's quality of life items that I see as split, with the SRs likely having better build quality with glass, but the Popper having a better chaff collection system (doesn't fall back in after the roast) and being less fragile.

I still really like the Fresh Roasts, and if I had one I would probably have no interest in replacing it with a Popper. But for someone new to the game that wants to air roast smaller batches for household consumption or samples, both have their strengths. Both encourage 5-senses roasting rather than by number, or something more similar to cooking than a commercial process. And Dave's right, both can be difficult to get perfectly repeatable across roasts. I'd compare it to frying an egg, which is surprisingly difficult to make repeatable. I'm sure I could throw tech and metrics at it to get something perfect every time, but a lot of us could get by with usually pretty good and sometimes excellent.

Bluenoser wrote:...In these small roasters the ambient temp makes a huge difference, ...

And don't forget mains voltage...
Huge variations rarely happens during a roast, but they are almost unavoidable from location to location.
Telling to a fellow with an identical machine to dial 64%, when his mains is 10V different than yours led to one minute event shift.

renatoa wrote:And don't forget mains voltage...
.

Good point. I wonder how Ikawa handles ambient temperature and mains voltage variations in their new home roaster.

I believe they measure the temperature of the air entering the chamber.

Ira