Why do roasters turn heat off at charge?

#1: Post by **lyle** » February 13th, 2016, 5:53 pm

I appears to me that a lot of roasters seem to follow a practice where they 1) preheat the roaster, 2) charge, 3) turn off heat, 4) wait until the bean probe temp turns around or some set amount of time, and 5) turn heat back on. I've seen this prescribed for gas roasters (the guys at Mill City) as well as some electric roasters (the M3).

My question is: why? Wouldn't a similar temp curve and ROR be achieved by a lower charge temp?

I plan to experiment with this, but I'm curious if there is some theoretical reason that turning the heat/burner off is preferred. Maybe I'm wrong and this practice isn't as common as I think it is. Thanks for your thoughts.

Lyle

February 13th, 2016, 6:48 pm

They are likely just doing the initial drying phase.

Len

#3: Post by **Boldjava** » February 13th, 2016, 7:25 pm

lyle wrote:...

My question is: why? Wouldn't a similar temp curve and ROR be achieved by a lower charge temp?

No, not in my experience.

I go after a steep curve (as encouraged by Rao) in the initial 3 minutes of the roast. I can best do that by having the burner off for the first minute. I used to charge lower and leave the burner on. In working through my roasts, I have found better cups by having that steeper initial curve via burner off. Lay two profiles (as you plan on doing) one atop the other. You will see it.

I couldn't begin to explain the food chemistry that creates those; I can experience a difference in the cup. Maybe it is a better way of activating the water energy. Who knows. Your mileage may vary. I don't believe there is "only one way to roast." When Joe Marrocco (Cafe Imports) works on our machine, you will see him throttle the gas way down rather that turn it off. I asked him why? "I don't want to struggle get the burners re-lit." In 18 months, I have never had our burners fail to reignite so I just turn them off via the electronic ignition.

#4: Post by **treq10** » February 13th, 2016, 9:50 pm

I think this is because you want a burst of energy to hit the beans in the initial charge without scorching/overdrying.

The basic challenge of coffee roasting is to cook the bean evenly from the outside in without drying out all the desirable flavor/aromatic compounds. Too long of a roast will dry it out so that nothing good is left, and too short of a roast will likely mean that the core of the bean didn't get the chance to develop.

A high charge temp gives the beans a head start. The beans do not caramelize until well over 300F, and so you don't really need to take that long to take the beans from room temp to 200F. The charge's purpose is to do this as fast as possible without scorching.

As it takes drum roasters quite some time to change in temperature because of thermal mass, starting from a low temp would create a lag in bean temp that would take too long to go from dry to ramp.

If you charged at a high temp but kept the heat up, the ET would be too hot and the beans can easily scorch. The initial hit of the cool beans will cause a significant drop in ET because they have soaked up a whole lot of energy. With the heat off, the ET will drop closer to BT, and once BT turns (or slightly before) the heat can be added to control the RoR so that it's gradually increasing according to the desired profile. If the heat is kept on at the charge temp from the beginning to end, the beans will cook really fast on the outside (probably scorching/burning) while leaving the core underdeveloped.

It's all about coaxing the heat to give the beans the perfect heat bath (or sauna?) so that it cooks evenly and sufficiently outside in.

#5: Post by **Boldjava** » February 13th, 2016, 10:40 pm

treq10 wrote:...

It's all about coaxing the heat to give the beans the perfect heat bath (or sauna?) so that it cooks evenly and sufficiently outside in.

I don't disagree with much you have written. However, it doesn't speak to OP's question re: why not charge lower temp, leave gas on and employ the same profile (I suggested it would be flatter, not the same). You can get the same finished time/temp with his approach. I have suggested he try it, lay the cups side by side, and slurp. There will be a difference even if he drops the two roasts at the same time and temp.

#6: Post by **day** » February 14th, 2016, 12:02 am

I don't think the idea that they would have the same delta bt graph really makes sense. Let's go extreme and then Just make up numbers ...(I'm a total amateur though but have been reading to understand fluid bed, so this concept was important to me) let's say that the beans are at room temperature, you drop in to a room temp drum and turn up the heat. Obviously the gas has to heat the drum, the air, and the bean simultaneously so the beans will have a fairly low ror, most importantly here the drum and, depending on design, air temp, would heat up faster than the beans. Thus, the ror of the et would be higher than that of the bt.

If we preheat to 300 degrees, turn off the gas and just leave it off, then at some point the bean temperature and environmental temp will stabilize and be equal, at first the beans will heat up very quickly But as they get closer to the et the ror will slow down. Thus, let's call that initial ror 60f/min. As the bt reaches 150degrees the drum is still much hotter than the beans but is losing heat and the two are much closer in temp, so now the ror will be lower, let's call it 40f/min

say we now turn on the gas, at bt 150f, theoretical delta bt 40f/min. If heat is then applied it must first start with the drum, which is now losing heat to the beans and getting heat from the gas. In this way you can control the descent from 40f/min until drop.

Let's say instead we charge at 150f. Just by natural logic we can tell that at the moment of drop the beans have to have a lower delta than at 300f charge. If it was 60f/min at 300 then it must be lower at 150. Thus, the ror might now be 45f/min. Continuing to apply heat must first heat the drum, and then the beans, thus the bt ror will not equal 60f/min, it just won't drop as it would with the heat off. stopping my long winded speech here, it seems that in order to reach the benchmark goals the delta graph will have to look different. As to how they slurp, I don't have enough experience to speak to that.

#7: Post by **TomC** » February 14th, 2016, 12:05 am

treq10 wrote:I think this is because you want a burst of energy to hit the beans in the initial charge without scorching/overdrying.

The basic challenge of coffee roasting is to cook the bean evenly from the outside in without drying out all the desirable flavor/aromatic compounds. Too long of a roast will dry it out so that nothing good is left, and too short of a roast will likely mean that the core of the bean didn't get the chance to develop.

A high charge temp gives the beans a head start. The beans do not caramelize until well over 300F, and so you don't really need to take that long to take the beans from room temp to 200F. The charge's purpose is to do this as fast as possible without scorching.

As it takes drum roasters quite some time to change in temperature because of thermal mass, starting from a low temp would create a lag in bean temp that would take too long to go from dry to ramp.

If you charged at a high temp but kept the heat up, the ET would be too hot and the beans can easily scorch. The initial hit of the cool beans will cause a significant drop in ET because they have soaked up a whole lot of energy. With the heat off, the ET will drop closer to BT, and once BT turns (or slightly before) the heat can be added to control the RoR so that it's gradually increasing according to the desired profile. If the heat is kept on at the charge temp from the beginning to end, the beans will cook really fast on the outside (probably scorching/burning) while leaving the core underdeveloped.

It's all about coaxing the heat to give the beans the perfect heat bath (or sauna?) so that it cooks evenly and sufficiently outside in.

You should write here more often. It's easy to see when someone knows what they're talking about.

#8: Post by **kwantfm** » February 14th, 2016, 12:16 am

I also like treq10's post.

There's one other reason why I turn heat off at charge. A little of my roasting background is required. I've just moved from a Quest M3 to a Proaster 1.5. It's been an enormous upgrade with the key lesson being the need to accommodate the much greater thermal mass of the Proaster. I roast a lot of different charge weights but always find that the thermal mass of the drum has me charging at about 180*C. So when I roast down to 400 g charge weights the amount of thermal mass tends to overwhelm the beans. At that low charge I have to leave the heat off for the first four minutes to get anything resembling a reasonable roast duration. It took quite a while to work that out!

On the other hand, a full capacity 1.5 kg roast requires heat on all the time with ETs tending towards the high end of acceptable for the entire roast.

#9: Post by **TomC** » February 14th, 2016, 12:24 am

day wrote:I don't think the idea that they would have the same delta bt graph really makes sense. Let's go extreme and then Just make up numbers ...(I'm a total amateur though but have been reading to understand fluid bed, so this concept was important to me) let's say that the beans are at room temperature, you drop in to a room temp drum and turn up the heat. Obviously the gas has to heat the drum, the air, and the bean simultaneously so the beans will have a fairly low ror, most importantly here the drum and, depending on design, air temp, would heat up faster than the beans. Thus, the ror of the et would be higher than that of the bt.

If we preheat to 300 degrees, turn off the gas and just leave it off, then at some point the bean temperature and environmental temp will stabilize and be equal, at first the beans will heat up very quickly But as they get closer to the et the ror will slow down. Thus, let's call that initial ror 60f/min. As the bt reaches 150degrees the drum is still much hotter than the beans but is losing heat and the two are much closer in temp, so now the ror will be lower, let's call it 40f/min

say we now turn on the gas, at bt 150f, theoretical delta bt 40f/min. If heat is then applied it must first start with the drum, which is now losing heat to the beans and getting heat from the gas. In this way you can control the descent from 40f/min until drop.

Let's say instead we charge at 150f. Just by natural logic we can tell that at the moment of drop the beans have to have a lower delta than at 300f charge. If it was 60f/min at 300 then it must be lower at 150. Thus, the ror might now be 45f/min. Continuing to apply heat must first heat the drum, and then the beans, thus the bt ror will not equal 60f/min, it just won't drop as it would with the heat off...

Pretty accurate summary of heat moving into beans, if I might say.

I think we can study and discuss the 'right' and 'wrong' way to roast coffee. I don't think we can unanimously agree or even elucidate the proper way to do so, for any given bean ( it's never going to be truly formulaic, each coffee is so different). I've long since advocated that the most important part of roasting coffee is the first third of the profile. I can speak only anecdotally, but I strongly believe it has the greatest influence. It's the exact opposite of Venetian artists, who had their understudies draft up the basic forms and colors in a painting, only to come along and bring to life the details of the hands or face. I think if a truly master coffee roaster set the profile properly early on, they could turn over the controls to a fledgling and have them end the roast whenever they chose, and still have a better result. But that's just my opinion and I could be wrong.

#10: Post by **9Sbeans** » February 14th, 2016, 12:42 am

There are many possible reasons.

Two of professional roasters I know of using the same machine as mine don't turn off the gas completely after the charge. My roaster can be switched to PID controlled and it's easier to leave an automatic very low flame on instead of completely turning off the gas. I also suspect the drum material makes a difference. The inner wall of the dual-wall drum is stainless steel, which has slower heat conducting rate then the outer cast iron. I haven't experienced the Facing, the roasting defect commonly attributed to too aggressive charge temperature / too much gas / or too slow drum speed at the charge, and therefore think that it's OK to leave the flame on. Tipping is another story though.

If I were operating a big production roaster, I would have optimized my work flow and reduced my adjustment steps into minimal necessary. IF (big if) I can achieve the same roasting profile time segments and produce the same cup quality, I would always favor charging at higher temperature and turn off the gas so that I won't waste my time waiting for the drum cooling down. I would then turn the gas full on in a big single step instead of breaking it into many small increments so that I would have time packaging the previous batch of bean on the cooling tray.