I know that there are a few topics on air flow in roasting, however I was interested that two different sources seemed to point in the same direction and wanted to ask people's thoughts on this. Having read (and enjoyed) Rao's Roasters Companion I took note that he provided a simple suggestion for determining "correct" air flow. If memory serves correctly, he takes the trier out and holds a lighter flame close to the outlet. If the flame leans slightly towards the trier outlet then he suggests that air flow is at the right level Giesen roasters seem to use the same principle. One can set fan speed to automatic on a Giesen and it sounds like the roaster maintains a certain level of negative pressure in the drum.

I am a little surprised given that I expect that altering the ratio of conductive/convective heat transfer to the beans (one can obviously do this ceteris paribus by altering fan speed or damper settings) would alter the characteristics of the roast.

Your thoughts would be appreciated,
T

There is a logarithmic empiric equation for the effect of airflow on heat transfer (reproduced in Illy) which, in the region of oven and roaster design is close to linear, so the faster the airflow, the more heat is transferred. This means that you can use lower ETs and get the same roast time with higher airflow. The same equation also addresses the thermal volume of the air and the beans; the upshot of that section is if you lower the charge the beans in a drum roaster, you can transfer more heat out of the air, and get either a faster roast at the same ET or a roast at lower ETs.

On the Quest, I can roast 250 grams to 100 grams (i,e, 100% to 40% charge); I assume this range is available on any drum. The lower the amount, the faster the roast I can get and the lower the ETs I can use. For commercial roasters, low charge roasts are a waste time and money; but for hobbyists, varying the charge on a drum roaster is the easiest way of going from high to low ET roasts while keeping the roast times the same.

Negative pressure is the minimum, not the "right" airflow for a "classic drum" roaster (to use Scott's term).

another_jim wrote:There is a logarithmic empiric equation for the effect of airflow on heat transfer (reproduced in Illy) which, in the region of oven and roaster design is close to linear, so the faster the airflow, the more heat is transferred. This means that you can use lower ETs and get the same roast time with higher airflow.

This is true, but highly misleading as a practical matter (at least in smaller roasters -- say 15kg and under) because increasing airflow tends to decrease ET by pushing therms out the exhaust in an amount greater than the increased efficiency of higher-airflow convection can keep up with. I.e., in the lot sizes we're likely to roast, more airflow means lower RoR.

A typical airflow profile for a roast master using an IR roaster such as a Diedrich, is 25% for Drying, 50% for Ramp, and 75% for Development. This is typical both for Diedrich (the roast master) and for Boot (with most of his published profiles).

For my usual profile with Central SHBs, using my open flame "classic drum" USRC Sample Roaster, I run about 25% air through Drying and most of Ramp, then increase to around 80% as the BT approaches 1stCs. Worth noting perhaps that my USRC roaster doesn't have the power to give me the Ramp RoR (~85F, 3:45) I want at 50% air; and my HT and Amazon are even less powerful.

In an open flame drum roaster, airflow is not only useful for transferring heat to the bean it's an aid to controlling temperature. Higher airflow means finer control with the gas dial.

During Development, high airflow is not only helpful to control a very moderate RoR, many roast masters believe that it's important to clear smoke and chaff. Some roasting theorists don't believe that a lot of smoke in the chamber has a deleterious effect on the beans, but many roast masters disagree.

I could go on, but the point is there's no "right" airflow for a classic, drum roaster -- it's circumstance and roast master dependent.

Rich

These roasting discussions seem to be getting more science free by the moment. Obviously, if you increase airflow, you have to increase the heat to compensate and maintain ET.

Thanks guys...

Jim, if I'm reading you correctly your post focuses on airflow relative to charge mass and how that adds flexibility to the hobbyist's roasting arsenal.

Rich, I assume that you are, at the very least, highly sceptical about the thought of an automated air flow solution as per Giesen.

T

kwantfm wrote:Jim, if I'm reading you correctly your post focuses on airflow relative to charge mass and how that adds flexibility to the hobbyist's roasting arsenal.

Exactly. Rich is right, of course, in saying that changing airflow without having a fast responding ET measurement is going to create big problems; however, I do hope roasters prefer better instrumentation to a candle ritual and a "do not touch this" sign.

kwantfm wrote:Rich, I assume that you are, at the very least, highly skeptical about the thought of an automated air flow solution as per Giesen.

I don't know enough about automated roasting to form an opinion, much less express one. So, I'll restrict myself to the observation that programmed, automated roasting defeats the point of hobby roasting.

In the Roaster's Companion (Chap. 17, pp. 76 and 77), Scott talks about a tendency for automated profile machines to over compensate for deviations from the programmed, ideal plot. I don't know if that's universal at the current state of the art; but doubt that it's inherent limitation and that improvements in hard and software software sophistication will solve the problem because... y'know... technology marches on.

In any case, just as too little air flow is not a good thing for a roast, too much air flow/air velocity isn't desirable either. Presumably, if you're using a drum roaster, you don't want the signature characteristics of a fluid bed which come from combining "too much" heat with "too much" air.

I don't consider myself an expert roaster or roasting theorist, but am pretty sure that the best way to approach the subject is with sensitivity and flexibility rather than a recipe. Ultimately, you'll develop your own style(s) and basic profile(s) based on your own tastes and equipment.

Rich

boar_d_laze wrote:I don't consider myself an expert roaster or roasting theorist, but am pretty sure that the best way to approach the subject is with sensitivity and flexibility rather than a recipe. Ultimately, you'll develop your own style(s) and basic profile(s) based on your own tastes and equipment.

Rich

Thanks... I'm fully manual on a Quest M3 at the moment and am contemplating the next big move. This is really a question of equipment choice and my early tendency (perhaps not that well informed) is to stay away from anything where there is "one answer"... I'm a little concerned that the Giesen's approach to fan speed may be limiting.

I'd be surprised if the Giesen Semi-Automatic couldn't be operated as a full manual as well as "semi-automatically," and they do make a full manual as well; but you need to talk to someone in Oz who knows their line.

Curious... Why a Giesen as opposed to something more manual and less expensive?

Also, I'd be a bit skeptical about taking Scott Rao's opinions about indirect heating and double-walled drums too seriously when it comes to 1kg roasters. I'm not saying he's wrong or staking any particular claim, but it seems to me there are all sorts of changing price and performance contingencies as you ascend the size ladder from 500g to 15kg.

Rich

The issue of air circulation roasting faster versus requiring higher temperature all depend on where the air is going.

If it is re-circulating like a convection oven, then the bean swill roast faster.

If it's being vented, it may require more heat input, but can assist in parts of the process like drying, etc.