The i Roast is here! The biggest change from the prototype (reviewed below) is that the design of the chaff collector is improved so chaff will not obstruct the screens and cause uneven roasting even with more heavily chaffed coffees like Ethiopian Harar. Based on some tests on the new machines we just received - Tom has some roast profile recommendations which I am posting below.I believe that this page - the roast curve graph and chart is still accurate. - Maria
Program Tips for the Hearthware i-Roast ver. 5/04
Understanding the Program Temperatures and the Onboard Thermometer Readings: The i-Roast is truly a next-generation home coffee roaster because you can create your own roast programs. Use the pre-sets to get used to the machine function, and remember that you can customize the overall roast time on each of the pre-sets. To find out how to use the custom program function and to adjust the roast times on the pre-sets, see the i-Roast manual.
ThereÕs a trick to using the custom program: the temperatures you program are not the temperatures you are going to read on the thermometer during the roast. This is going to take some getting used to, but after a few trials you will see the correlation. The actual onboard thermo temperature seems to be about 40-50 degrees lower than what you program. So when you set warm-up max temperature and you want to see 380, you will have to program in 430, roughly. But the other factor is the air flow pattern - if the roaster uses a high velocity air flow to maintain the program temperature you choose, it will tend to lower the overall roast chamber temperature. I like this for the first stage because it aids in even roast development. In fact, the i-Roast is the most even roaster I have seen for the early warmup stages, aside from $4000 professional sample roasters! I am still playing with the programs and seeing how different time & temperature combinations affect the roasters behavior in terms of onboard temperature readings and air movement.
In general, setting of 320-400 degrees are good for your warmup period and are not going to get the coffee into the first crack. Settings of 450-470 degrees are going to slowly get the coffee into the first crack range. (First crack occurs with onboard thermometer readings of around 375-380 on my test roaster). If you can avoid using the highest temperatures, 480 to 485, do so ... but if you need them to get the kind of roast you are looking for then they are there to use! If these conflicting temperature settings/readings seems odd, remember that the numbers are not the important thing ... what you want is for a coffee seed to progress evenly and steadily through the roast stages, not too fast or too slow. The i-Roast does that extremely well with the pre-sets, and with the custom programs you can truly get one-of-a-kind roast results. DonÕt program a roast that has the seed advance to a stage of roast (such as the medium brown color the coffee achieves just before 1st crack) and then just stall there for a long period of time. Using your program, keep the coffee moving, slowly and steadily, through the roast stages. With darker roasts, try to have a "controlled" end of roast, which is indicated by a defined pause between the end of the "first crack" and the beginning of the "second crack." (Another indicator is a slow and controlled sound of second crack, not a super-rapid, out-of-control snapping.)
Preset Roast Curves - Specific on the Preset roast curves that come programmed into the iRoast
Preset #2 : total roast time : 11:30 min - shorter stage 1 time, longer middle stge, otherwise the same as preset 1.
Sample Roast Curves for Specific Coffees:
Total roast time: 9:30 min
I am also using a batch size of 130 to 150 grams - which is actually abotu 6 ounces of coffee. This is not what Hearthware recommends - so you might try both and see what you like more.
Island Coffees: Coffees from Jamaica, Hawaii, and to a lesser degree Puerto Rico, have a lower bean density because these island coffees do not have the altitudes of such origins as Kenya. They benefit from a lower initial temperature during the warmup time. Here is the program I am currently using for a City roast:
Brazil Coffees for Espresso: Like the Island coffees, Brazils come from lower elevations. They benefit from a slower warmup and a longer overall roast. For Northern Italian Espresso I like this program, and you can use it for other light espresso roasts too:
Some additional thoughts; roast theory ...
A basic profile I am using is my lower-heat warmup profile. It benefits all coffees, and only causes problems with really dense seeds that need a higher initial temperature. Basically it is
340 for 2 minutes --this will give you a high speed air pattern to aid in really even initial heat distribution
I think the best way to use the curves in the i-Roast is to regulate initial heat distribution. Air roasters, even the professional Sivetz air roasters, have never allowed a slower warm-up "ramp" and this is where you can cup quality by letting the coffee accept and distribute heat according to its physical limits, not forcing the heat on too quickly. Once heat is distributed from the core to the exterior of the bean evenly, then a higher roast temperature that allows the coffee to pass through the cracks can be applied. If this temperature is not too excessive, the cracks will happen in a slow and controlled way, with a clearly delineated pause between 1st crack and 2nd crack.
At first I thought I would be using the stages of the i-Roast program to regulate the finish of the roast, but realized through trial and error (and cupping) that this doesn't make sense. The fantastic thing about programming a temperature over a period of time is to get the batch on the right "track" for the final finish roast temperature. As I said, no air roaster has offered this before, and despite the fact that the buttons and programming technique on the i-Roast is ...well ... lame, and despite the fact that the temperature you program is not the temperature you read on the onboard thermometer, the i-Roast is still a "next step" in air roasting because of this feature.
One last thought ... don't get hung up on overall roast times. 8 minutes, 11 minutes, 14 minutes ... it DOES NOT matter! Roast times are only relative to the method of transferring heat to the coffee. Air roasts are not supposed to be as long as drum roasts because they have the ability to transfer heat rapidly in a high velocity air stream. A drum has to conduct and convect heat to coffee much slower to avoid scorching. With the i-Roast, let the coffee be your guide in programming curves. Watch what the coffee is doing, and adjust your curves to what you observe. If you blast it with hot air, and 1st and second crack blur together. make adjustments. If you stall it, and the coffee just sits there, not developing but being exposed to heat, then you are baking it ... make a change. But a particular roast time is not a goal, and a 10 minute roast in one machine has not relevance to a 10 minute roast in another machine, not in terms of degree of roast, quality of roast, or cup quality. I hope this helps a bit... Tom
|An update on the new chaff collector design, and the exhaust adapter...|
| || |
The i-Roast features a new design of the chaff collector ring, an insert that goes inside the chaff collector top. This is a great improvement from the prototype design I tested; who would guess such a small design change could make such a big difference. The new design has improved the consistency of the roast levels between different types of green coffee, some that produce more chaff during roasting, some that produce little to none (like decafs).
The new chaff collector ring fits in the top as pictured,
A Preliminary Review of the Pre-Production Hearthware I-Roast Home Coffee Roaster
I saw glimpses of the prototype I Roast over as far back as 2 years ago. I worried that the final machine would not deliver all the promise that I saw in the protoype. But I am not disappointed at all! No, it is not the super high tech, expert machine some of you would want. But be reasonable, you are not going to see a roaster with $500 of PID controls and multiple thermocouples and bean probes and a Comm Port for your PC. What we have here is something that works well at default settings and takes home roasting where it has never gone before, into basic profile programming
|First, the downsides.... The main problem with the machine is it is loud! It is quite loud. It makes me want to put on earplugs being near it. Its probably as loud as the Gourmet, like a vacuum cleaner or a hair dryer - loud. They explain this: to roast a higher volume of coffee they needed to put in a higher-speed fan and a beefy motor.|
The great thing is... it works - it roast 150 grams, a full level cup of green coffee, 2x the batch size of the Precision and Gourmet. It roasts it evenly through the early stages of the roast - frankly, I was skeptical if it could but the coffee yellows to a very even color, which makes it a big improvement over the 2 other air roasters on the market. Probably the one other thing I don't like is that the digital controls are hard to read - they are big and well-designed, but you have to get your head down level with the machine to see the settings.
City stage roast results from my test - corresponding thermometer catalog output in a graph format is at the bottom of the page!
The I-Roast Base - wide at bottom for stability. Digital readout is hard to view unless your eyeball is somewhat level with it. It would be nice if it was angled up...
|My machine arrived with a cracked chaff collector because it wasn't shipped double-boxed with foam- we will ship it properly when we get them so this is not an issue "repaired it" with some sheet aluminum and screws, because there's no way I am going to wait to get a replacement before testing this puppy out. So the chaff collector is a little thin, but involves less parts than the Precision one, and will be cheap to replace if you ever have a problem with it.|
Now, the upsides ... A long time ago I tried to tell Hearthware that there would be a market for a "Precision Pro" a model with more user-control, with temperature and time control, even though it would cost more. The I-Roast is that machine. You can use 2 Pre-set roast curves if you like, add or take away the total roast time, hit a button, and the machine counts down roast time, and when it hits cooling cycle it flashes COOL and counts down the 4 minute cool time. This roaster has an overall MAX roast time of 15 minutes, in any mode. That's plenty for an fluid bed roaster. You can of course end the roast at any time by hitting to COOL button. The roaster progresses through 3 stages, indicated on the digital screen. In the third stage, you can add time to the roast if you see that the coffee is not to the degree you want it. What's really cool is that at any time in the roast, in any mode, you hit the TEMP button and you get a reading of the roaster temperature!!! It's great, like having a built-in digital thermometer. But it is the ROASTER INPUT AIR TEMP. you are monitoring, not the bean temperature (which as all you who have mod'ed out roasters know, is VERY hard to read, especially in an air roaster.) But the best thing here, besides larger batch, EVEN roasting, onboard thermometer, countdown timing etc. is PROGRAMMING! Now, it could be a lot more sophisticated, and store a bunch of roast curves etc., and it could have lots of programmable input points etc., but it doesn't and no home roaster ever will. What it does have is a single saved (if the machine is not unplugged) roast program corresponding to the 3 stages of the preset roast curves: you can set the time and temperature for each of 3 stages that form the total roast time. So you can say I want 7 minutes warm-up at 390, then 3 minutes at 420, then finish with 2 minutes at 450 target temperature. (Max total of 15 minutes is possible). Its quick to set the program once you do it a couple times, and it saves the program until you override it, or unplug the machine.
|Ok, the caveat is this: the temperatures you program are, I believe (I will find out) the heat coil temperature, and the actual onboard thermo temperature is about 40 degrees lower. So when you set warm-up max temperature and you want to see 390, you will have to program in 430, roughly. It wont take long to build a table of the program input temperature, the onboard thermo, reading. and a themo-probed bean temperature from an external digital thermometer. I can put that up on the web site, and will probably have something printed to send out with the roasters. What this means is that we can all play with the time and temp.s of these 3 roast phases, and given that the voltage and the ambient temp is about the same, actually share our roast "curves" with eachother for specific coffees/blends/etc.|| |
Note the difference!
I am pretty excited about that. I have played with a few settings already: slower vs faster Stage 1 warm-up, holding the coffee at 350 briefly, fast finish vs. slow finish. etc. It actually seems quite new to me ... things I do on a drum roaster, and that a few of you with PIDs installed in air roasters have done, but basically now everyone can try different roast styles on an air roaster. Now, I did some tests to see how dark the roaster can get, (HW's concern) and I feel it does fine, but I haven't tried testing it on slightly lower voltage to see if it is still adequate. It can go past where I ever take my espresso roasts. Nonetheless, even if you are not a dark roast person, you want your roaster to be capable of it. All heating elements will cool slightly over time, so you want some headroom to keep you covered in the long run. Anyway, the roaster really (sounding like a soundbite here but ...) takes home roasting a step ahead, and I hope they take their time, and make sure they ship a consistent and reliable roaster. Like 'em or not, HW has put an immense effort into this machine, and shown good faith in not rushing it to consumers. No reason to stumble at the threshold.... Tom
Here are some general views of the machine while I have been testing it.
The point of all the pictures is so you can see the probed bean temp. vs. the onboard thermometer temp. Note how they get closer toward the end of the roast.
| || |
| || |
| || |
| || |
Cooling temperature: coffee drops to 124 after 4 minute cool cycle - that's quite acceptable
It is very easy to snake a K-Type or J-Type probe into the chamber - no drilling or modification required!
TOP CHAFF SCREEN after roast
Bottom of chaff collector
Roasting pot with patented "fountain" - slightly different from the one in the Precision.
The coil and sensor can be seen in the base.
Another view of the same...
As mentioned above, the I-Roast temperatures you program are of the input air temperature, not the bean temperatures. This graph shows a relationship for a good Full City roast curve (see program settings in graph) between the temperature I programmed into the I-Roast (yellow), the Onboard Thermometers temperature reading (blue), and the probed bean temperature I logged from an additional digital thermometer with a K-type probe inserted into the coffee (red).
The next step is to come up with a chart that lists the relationship between the program temp, and the bean temp, so you can know that when you program 400 degrees, your green coffee temp will plateau at around 360 (as is the case in stage 1 below.)
Here is the data that the chart above is derived from:
|Sweet Maria's |
|Sweet Maria's |
|Further Reading||The Complete Sweet Maria's Coffee Library Page |
- Coffee Travel Pictorials, New Product Reviews, Roasting Pictorials, Etc!
|Interesting Coffee and Coffee Roasting Web Sites |
- Links to Home Roasting Web Sites, Coffee Industry Sites, Great Coffee Books, Etc!
|Coffee Book Recommendations|