This completes my first impression of the roaster.

Bottom line: This roaster is comparable in price, capacity, and build quality to the Hottop; but the emphasis is very different. The Hottop has a friendlier interface, more automation, and more consumer orientation. The Quest is like a commercial decaf grinder or catering espresso machine, a solid no-frills design meant more for light commercial than home use. The ideal use for it is as a sample roaster for a small cafe/roaster or the more prolific amateurs. For my needs, it is a good fit, better than the Hottop would be.

I particularly like the simple controls, and the full range of feedback -- there's an accurate thermometer, a trier, a window, and the air exhaust is good for sensing the aromas without getting a nose full of smoke. If I were to order it again, I'd ask for a second well for a dial thermometer or TC, located in the front plate, near the top edge, so one could monitor the ET at the same spot I use, but more elegantly than my jury-rigged setup.

One note: The M3 is a DIY machine when it comes to repairs and maintenance. The manufacturers have been terrifically helpful, but you'll need to repair and service it yourself with parts they supply.

Hi.

I've been playing with the fan now that I have a good set gauges. I can visually detect any microscopic changes and how they affect the temperatures. Till now, I always kept the fan at a setting of 5 (middle) all through out the roast.

When the fan is maxed, the BT change of rate is highest. I was like wuau, look at that baby go up!. In an empty drum, the BT line went from almost flat to almost vertical when the fan setting went form 5 to 9. This high Delta BT possibly means a higher heat transfer rate by convection. At the same time, the MET will drop if in constant manual mode (cooler air coming in for the same power setting). I have my MET probe in an air sensitive location which may have something to do. But if the MET is PIDed then there is no drop.

It seems that the two heaters, the hot outer layer of the drum, and the hot surface of the roaster walls heat up the incoming air. The hot air transports heat to the interior of the drum fast. The higher the fan setting the higher the heat transported. I guess that this effect is bigger the lesser the amount of beans since more beans would restrict the air circulation.

I don't know yet what would be the best fan strategy. For example, in a final stage if a put the fan at max setting from medium setting (in manual mode no PID), it would bring down the MET but it would also initially transfer more heat to the beans by convection. Another scenario, if in the middle of the roast I start with full maxed fan setting and then bring it down to half at the final leg, it would cool the BT (less convection) but raise the MET (if not PIDed). Maybe a combination of power and fan is best.

Some food for thought.

Cheers

The ramp from about 300F (150C) to the first crack can be a little slow (unless you want nutty & malty flavors), so cranking up the heat and air is probably the best way to speed it up without tipping the beans.

On the other hand, I like a high drop in, so I turn both the air and heat low during the drying phase. But I still haven't figured out all the possibilities or how they taste.

If the roaster were perfectly efficient, the heat transfer or rate and he rise-rate of he beans would be set by the amperage of the heater, while the airflow would set the ET, higher with low airflow, lower with high airflow. But I think its a bit more complicated than that. Rafael, with all the instrumenting you've done, you should b e able to say for certain fairly soon.

I also think it is complicated.

My only problem is with low density beans. I can so sidamo beans 1C at 5:30. They come out pretty good because of high MET and high charge temp (BT 440F).

But how to transfer the most heat while keeping a very low MET for low density beans? Low density beans also need a low drop temperature. It seems that a combination of low to med. load and a high fan setting would give a faster roast time.

I would like to know what type of test I could do other than DeltaBT. Maybe there is somewhere a test procedure to bench mark a roaster?

Cheers

Much seems to do with the effects of conduction, convection and radiant heat on taste. And also that different types of beans are effected differently by the three. I get very little conduction with my roaster. So I control a balance between convection and radiant. Adjusting the batch size to the heat transfer to keep a reasonable MET for a bean type. I have 2 turbo oven tops one has the stock 2000+rpm the other has a 5600rpm motor with rheostat. I keep adjusting the balance to taste. The highest convection roasts are generally not my favorite. The only time I will crank up the custom top is when I'm pushing my batch size.
I've been monitoring your trials to see the taste effects of adjustments of the heat types. Jim coming from strong convection roasting with the popper. It seems that the magic of a great Roaster is not merely about BT/ET/MET/Times but the manipulation of heat types throughout the roast.

Thanks for the input.

farmroast wrote:Adjusting the batch size to the heat transfer to keep a reasonable MET for a bean type.

That seems to be the ticket here also. For low density beans, the MET has to stay low to avoid staining the beans with burn flavors/bad acidity. But keeping the MET low also brings down the gap between BT and MET, which makes it slower. Some of the low density beans I have are supposed to have fruity flavors but I am afraid that long roasting times kill it. I don't mean that the roasts are bad (defective) but I think they are not at their best if roasted long (+16 mins). So keeping a matching load size to a matching low MET, and using the fan to enhance the heat transfer, are the only ways so far to bring shorter times. I am still learning and this opinion may change through time.

Rgrds

Seeing is believing.

These are two test roast. Identical except for the fan setting. No adjustments made

Beans: Guatemala Acatanango - Finca La Soledad

First batch:

charge 150 grams
FAN setting = 9 (max) all through out
Drop charge = 210F
MET = 500F
1C = 8:50 at 380F
Roast ended at 11:50 403F

Second batch:

charge 150 grams
FAN setting = 5 (med) all through out
Drop charge = 210F
MET = 500F
1C = 11:13 383F
Roast ended at 14:45 409F

(notes: the timer shut down the roaster at about 400F in the second batch. See the dip in temp. However, it didn't affect much and I restarted the roaster immediately)

I'll be testing the coffees tomorrow using drip instead of espresso

First batch:




Second batch:



PS: Note that the graphs look similar but the second one started much earlier (longer roast time). I still have to work with the software to mark the charge drop and show 1C and 2C. By the end of next week it will be all working. I may have to use the system clock instead of a timer to keep it more accurate. The times for 1C mentioned above were measured with a stopwatch anyway and not the software. I also moved the MET probe closer to the heater. The closer it is, the more sensitive (before didn't fluctuate that much).

Hi.

Yesterday I marked one bag with a pice of tape but now I don't remember which one it is So much for all the work jeje I remember that the first batch had some 'toasty' bite to the bean (fast slope phase) but it didn't show any defects in the cup, on the contrary.

None of them had 'grassy' flavor (short drying phase). One has more acidity and sweetness than the other, which I like more. I went back to chew the roasted beans to try to identify which one is which. I think that the sweeter/brighter one is the first one. So even though it had some 'toasty bean bite,' it ended up with a better flavor in the cup. The second batch was flatter/uninteresting but not defective. I wasn't able to taste the apricot flavor the mention in the SweetMarias page.

This is my learning check up for both of them:

First Drying phase:
Grassy flavor: No
Dusty flavor: No

Second Roast Phase
Toasty flavor: No
Bready flavor: No

Third Finish Phase
Overly acidic: No
Flat: No

Cheers

sounds like a fun roaster. I kinda wish i had gone with it instead of my Hottop. The biggest advantages seems to be the more manual control, the ability to actually have airflow through the roast (hottop fan seems useless in this regard) and the type of heat transfer. In the hottop the heating elements are in the roasting chamber, so the beans recieve a good bit of infrared heat and the drum itself is simply a wire mesh. This roaster seems to mimic commercial roasters better with a solid drum and indirect heating.

I dug out my Sivetz, a coffee tech book from the 70s, which discusses the different drum designs.

The perforated drum design without forced air is by Jabez Burns from the 1930s. In this design, the thermal gradient from the bottom to the top of the roaster creates adequate convection current airflow through the perforated drum, providing there is an air intake at the bottom, and an exhaust at the top of the roaster.

The design with forced air drawn into the back of the drum, and the beans rotated toward the front, is by Probat, also from the 1930s.

The Burns design became the standard for craft roasters in the US; the Probat for Europe. Sivetz comments that both designs will roast a load of beans in eight to ten minutes without exceeding 500F environmental temperatures. In his judgement, this made them the first good roasters, i.e. that didn't spoil the coffee either by flattening it with over long roasts or scorching it with too hot drums. (Sivetz insists on 8 minute roasts; but since the old roasters took about 20 to 25 minutes, he was right about them being too slow)

Bottom line, for best roast quality, the two designs are equivalent. The Probat is easier to use, because the extra control over the air makes the roaster a bit more maneuverable.

Neither design remained state of the art for long. The Burns Thermalo from the 1940s, a solid drum with recirculated air, was able to do the required eight minute, 500F roasts with a lot more beans in the drum, and for less power expenditure, then either the Burns or Probat design. This was the first industrial roaster. I saw one used at Green Mountain (with modern controls), and it was no larger than a one bag shop roaster, but did 500 pounds at a time.