Why I chose IMF for Prodigal

I’ve been asked dozens of times why I chose IMF for Prodigal, and not Loring or Probat. To understand that, I’d like to explain how those machines differ and why I believe IMF makes the best roasting machines in the world. 

Drum roasters vs air roasters

Although IMF and Loring machines have drums, most of us call them “air roasters” because they transfer heat almost exclusively by convection with little-to-no conduction. Some air roasters, such as Sivetz and Neuhaus Neotec move the beans via a “fluidized bed” of hot air; other air roasters such as IMF and Loring use a drum to direct the bean movement, independent of airflow. 

Over the years, marketers have done a good job of muddying the facts about drum vs air roasting. Air roasting advocates have often claimed all drum roasts are “burnt” and only air roasting is “clean” and some drum roaster fans have claimed air roasts have “no body” or “uneven development.”

While such claims may have some merit when a machine is used improperly, I do not think they accurately represent the results a semi-skilled operator would achieve. 

A skilled roaster will find drum roasts offer a little more body, a bit less flavor clarity or delicacy, and less inner-bean development at a given outer-bean roast color. All of those differences are due to drum roasters transferring more heat via conduction. 

Air roasters, on average, make inner-bean development easier and are less likely to impart roasty flavors. The same trade-off that exists between flavor clarity and body exists in roasting; increasing one decreases the other. 

Air roasters allow faster roasting than drum roasters do, since air roasters can operate at extremely high inlet-air temperatures without concern of overheating the drum surface. The common notion that roasts should take “10-12 minutes” or the like is an artifact of most roasters using drums; let’s say you put a massive burner in a 12kg Probat and roasted 12kg in six minutes; the drum would get so hot that it would sear the beans, causing tipping, scorching, and likely inadequate inner-bean development at lighter roast levels. But it’s common to roast in 6-7 minutes in air roasters; hot air does not damage beans (for reference, in my ROEST, I use an inlet temperature as high as 700 fahrenheit during my 200-g batches with no cosmetic or flavor damage. 

One could argue that roast consistency is a little easier with an air roaster, though I think nearly equal consistency is achievable in most machines, provided they don’t have features or control systems that interfere with an effective between-batch protocol. 

Which type of roaster is better? 

I don’t believe there is a one-size-fits-all roaster design. If you prefer darker roasts and maximum body, you should probably use a drum roaster. At darker roast levels, inner-bean development won’t be much of a concern, and you probably won’t mind the lack of flavor delicacy. 
If you prefer light roasts and high flavor clarity, you’ll probably want to choose an air roaster. 

But, dear reader, please do not misinterpret what I’ve written here: nowhere have I said that you cannot develop light roasts well in a drum roaster or impart roasty flavors in an air roaster.  These are merely comparisons of their relative performance.

IMF vs Loring

Given that I prefer light roasts and delicate flavors, I was definitely going to buy an air roaster for Prodigal. The choice came down to IMF and Loring. 

Both machines use air-recirculation: instead of having a burner for the roasting process and an afterburner to clean the exhaust air, they have one burner held at afterburner-level temperatures to clean the exhaust air and provide hot air for roasting. Both designs are more fuel efficient than using a drum roaster with an afterburner. But the similarities end there. 

Airflow: There are effectively two different airflows in IMFs and Lorings, what I’ll call “air throughput” and “recirculated air.” Think of them like this: imagine you are filling a leaky bucket with water from a faucet. The water entering the bucket and draining from the bucket are the “throughput.” If you simultaneously stir the water, that is the recirculation. The “types” of airflow in the IMF and Loring intermingle but their levels are independent. 

Both machines recirculate air through their burners, and assuming one maintains the burner temperature high enough, the air is “cleaned” of smoke and pollution, to prevent the recirculation from tainting coffee flavor. 

The amount of throughput in Lorings is correlated with the gas setting, in order for its fuel injection system to maintain a constant air:fuel ratio. So, when using a gas setting of 80%, the throughput is roughly twice as high as when the gas is at 40% (I’m not sure if it’s exactly a 1:1 ratio, but it’s close). I do not like this system for two reasons: it prevents the Loring from lending itself to inlet-temperature profiling (see below), and it means the air throughput is very low when using low gas settings late in roasts. This, to me, is the reason why the IMF is better at both controlling roasts through first crack and executing dark roasts. 

The IMF exhaust fan and burner temperature are independently controlled. That means we can have whatever throughput level we want at anytime. It also allows one to use inlet-temperature profiling, which is the best way to control an air roaster. 

Incoming air in the IMF is a combination of hot air from the burner and air from the roasting room, blended dynamically by a fluttering “vortex” valve that produces the desired inlet temperature at all times. The level of air throughput is determined by the exhaust fan setting, which is set by the operator.

To me, the only benefit of Loring’s lower overall air throughput is it makes the fuel efficiency higher than that of the IMF. The IMF’s higher air throughput offers superior control of roast curves at all times, especially when most needed late in a roast. 

Airflow: advantage IMF

Fuel Efficiency: advantage Loring

Drum: The Loring’s drum is solid steel, while the IMF’s is perforated. While I don’t think that’s a big deal, I would always choose a perforated drum in order to decrease potential conductive heat transfer.  

Drum: advantage IMF 

Productivity: Both machines can roast full batches faster than a drum roaster. Some Loring machines seem more powerful relative to stated capacity than others (The 15kg seems especially underpowered.) But IMFs are more powerful than Lorings at any size, and one can roast full batches in less than 8:00 in IMF with no compromise in quality. 

Productivity: advantage IMF 

Control System: Loring offers three ways to control roasts: manually changing gas settings, using automated burner recipes, and “profile roasting.” Burner recipes are like manual control, but with valve settings preset by the operator, and executed by the machine (Cropster Gas & Replay, and Artisan software offer this same option for any machine that has a digital gas valve capable of receiving commands from the software.) In profile roasting, which I wrote about here, the Loring uses a PID to change the gas setting automatically, in an attempt to trace the bean-temperature curve of a reference roast. The system does not perfectly trace curves, because profile roasting gets confused during the first minute of a roast while the BT reading drops. After the turn, the system does a relatively good job of tracing curves, assuming the operator has programmed a curve achievable by the automation. I find the profiling system useful for very dark roasts, but a skilled operator can usually roast more consistently in manual mode than in profiling mode. 

The IMF control screen allows the operator to create a twelve-step recipe using bean-temperature set points. The operator can choose the inlet temperature, drum speed, and fan setting at each bean-temperature set point. I cannot overstate the superiority of controlling roasts by inlet temperature rather than valve settings. For example, let’s say you have a Probat and have a recipe that looks like this: 

gas 80@ BT300°F

gas 70@ BT320°F

gas 60@ BT340°F

The beans do not experience the gas-valve settings, they experience the inlet temperatures produced by those settings. However, since ambient air is drawn through the burner, unless the ambient temperature is tightly controlled, a given valve setting will produce different inlet air temperatures as ambient conditions change. For example, it’s typical for a certain gas setting to produce hotter inlet temperatures at the end of a roast session than at the beginning, if the roastery gets progressively warmer throughout a session. Likewise, inlet temperatures tend to be higher in summer than winter for a given valve setting.  

Directly controlling the inlet temperature prevents inconsistencies due to changes in ambient conditions. I doubt many roasters until now have recognized the benefit of such a system. The improvement in predictability and consistency yielded by an inlet-temperature control system is why I asked ROEST to add that feature to its machines, and why machines like ROEST and IMF make roast replication much easier than machines that require control by gas-valve settings. 

One cannot operate a Loring by inlet temperatures, and even if they could, the constantly changing air throughput of the Loring would make the heat transfer more challenging to manage. 

Advantage: IMF, by a landslide

Price

While price was never a consideration when deciding which machine to purchase for Prodigal, IMFs are, across the board, cheaper than Lorings. I doubt that will be the case for long, once roasters come to understand the benefits of IMF. 

Advantage: IMF

Conclusion

While I consider Loring and IMF to be two of the best roasting machines on the market, I chose IMF for all of the reasons above. I confess that I didn’t start Prodigal earlier because I was frustrated with the choices of sample roasters and production roasters until recently. Now that I have found IMF, and ROEST has added inlet-temperature profiling, I couldn’t be happier with o results. 

PRODIGAL IS BACK. GET PRODIGAL TODAY